# OrcaSlicer.net - Full Content Archive This file contains the full prose content of every article on orcaslicer.net, intended for AI/LLM ingestion. For metadata-only index, see llms.txt. Site: https://orcaslicer.net | Articles: 120 --- # Section: Beginners & Setup ## OrcaSlicer System Requirements: GPU, RAM, OpenGL Explained URL: https://orcaslicer.net/orcaslicer-system-requirements/ I've now installed OrcaSlicer on a Raspberry Pi 5, a battered Core i5 Windows laptop from 2015, and an M2 Mac mini, partly because I was curious and partly because every "system requirements" page I found online disagreed with the next one. What I kept hitting was the same gap: the project's own README doesn't actually publish minimum specs, yet half the internet quotes a confident "4 GB RAM minimum, 8 GB recommended, Core i3, OpenGL 3.3" table that nobody can trace back to a primary source. So this is my honest writeup. I'll show you what the OrcaSlicer GitHub repo actually says (and doesn't say), where the real per-OS floors come from, what GPU and RAM and CPU truly do during a slice, and which numbers floating around in 2026 are just guesses dressed up as facts. If you want the installer itself, grab it from the official OrcaSlicer GitHub releases page rather than any mirror. Table of contents What the OrcaSlicer team actually publishes OS minimums you can trust The OpenGL 2.0 reality GPU is for the viewport, not slicing RAM and CPU, what actually matters The 4 GB / Core i3 myth Linux gotchas: GLIBC, FUSE, Flatpak What to actually buy FAQ What the OrcaSlicer team actually publishes for system requirements I went straight to the source before writing a single recommendation. The OrcaSlicer GitHub repo is the canonical reference, and its README is what you'd reasonably treat as the official word. I read the whole thing twice. There is no "System Requirements" section. There is no minimum RAM line. There is no CPU floor. There is no listed OpenGL version, no Vulkan target, no GPU compatibility table. The install section is purely instructional: download the Windows installer, drag the macOS app to Applications, grab the AppImage or the Flatpak on Linux. The only runtime dependencies the README mentions at all are on Windows: Microsoft Edge WebView2 Runtime and the Visual C++ 2019 redistributable. That's it. Everything else you've read about minimum specs has been inferred by someone and copy-pasted across mirror sites and AI summaries until it looked official. OrcaSlicer GitHub releases page showing the v2.3.2 download assets for Windows, macOS universal DMG, Linux AppImage and Flatpak So this guide is written from the position the README forces on anyone honest: there's no published minimum spec, only the runtime hints buried in the code, the error strings the app actually throws, and the real-world bug reports that show what passes and what doesn't. OS minimums you can actually trust Here's what I can verify from primary sources, broken out per OS. Windows. The README lists two runtime deps: MicrosoftEdgeWebView2RuntimeInstallerX64 and vcredist2019_x64. Both ship with reasonably modern Windows 10 and 11, but if either is missing the app won't launch. The v2.3.2 build is 64-bit only (asset name: OrcaSlicer_Windows_Installer_V2.3.2.exe). My 2015 i5 laptop runs it fine on Windows 10 22H2. Worth flagging Issue #13239, where a user with an i7-13700K, 32 GB DDR5-6000, and an RTX 4090 reported a ninety-second launch on Windows 11 25H2 with OrcaSlicer 2.3.2. That's not a hardware bottleneck, that's the app itself, and it's a reality check against the "throw more hardware at it" instinct. macOS. The CMakeLists.txt includes set(CMAKE_OSX_DEPLOYMENT_TARGET "11.3" CACHE STRING "Minimum OS X deployment version" FORCE). That's the build deployment target (the SDK level the binaries are compiled against), not a runtime line saying "macOS 11.3 minimum." Conceptually the project floor is Big Sur 11.0, but the published universal DMG is built against 11.3, so on 11.0 through 11.2 the official binary may refuse to launch. On my M2 mini running Sequoia, it just works. Also: v2.3.2 ships a single OrcaSlicer_Mac_universal_V2.3.2.dmg, a universal binary covering both Apple Silicon and Intel. The README still says "pick the arm64 or x86_64 DMG," but as of v2.3.2 there's only one file. Linux. The v2.3.2 AppImage is named OrcaSlicer_Linux_AppImage_Ubuntu2404_V2.3.2.AppImage, and the Ubuntu2404 in the filename matters. It means the build is linked against GLIBC 2.38 from Ubuntu 24.04. Older distros won't launch it. The Flatpak (flatpak install flathub com.orcaslicer.OrcaSlicer) bundles its own runtime and dodges the whole issue, which is what I'd recommend for anyone not on a current Ubuntu base. The OpenGL 2.0 reality (and the error nobody wants to see) Search for "OrcaSlicer minimum OpenGL" and you'll find people quoting 3.3. I quoted it myself in an earlier draft. It's wrong. The actual minimum, the one the app enforces with a popup, is OpenGL 2.0. I know because Issue #2928 shows the verbatim error string the application displays: The application cannot run normally because your OpenGL version is lower than 2.0 That issue was filed against OrcaSlicer 1.8.0 on Windows 10 inside a remote-desktop VM, and the same string crops up in Discussion #3881 (Qubes OS) and Issue #34. So 2.0 is the wall. If your GPU can do OpenGL 2.0, you can launch OrcaSlicer. The OpenGL 2.0 error dialog OrcaSlicer shows when the system GPU or driver does not meet the minimum If you're stuck below 2.0 (typical in VMs and RDP sessions), there are two workarounds from the tracker. Launch with the --sw-renderer flag, which forces software rendering and skips the GPU entirely. Or copy the opengl32.dll from a PrusaSlicer install into your OrcaSlicer folder, which several users in Issue #2928 report fixes the VM case. Neither is in the README. If you're still stuck, our OrcaSlicer won't open troubleshooting guide walks through the OpenGL failure modes step by step. GPU is for the viewport, not for slicing This matters because I see it misstated everywhere. OrcaSlicer doesn't use your GPU to slice. The GPU does one job: render the 3D viewport via OpenGL. That's the model preview, the toolpath visualizer, the layer-by-layer playback. All slicing computation (perimeter generation, infill, support trees, gcode output) runs on your CPU. How do I know? The README never claims GPU acceleration for slicing. The CMake build doesn't link any CUDA, OpenCL, or Vulkan compute libraries. The slicing engine is inherited from PrusaSlicer and Slic3r, both CPU-only. Put a stopwatch on a slice and check task manager: CPU pegged, GPU idle. I confirmed this on both the Windows laptop and the Mac mini. OrcaSlicer 3D viewport during a slice preview showing the GPU-rendered model with toolpaths Where it gets interesting is multi-plate work. Issue #11991 is an open feature request asking for parallel processing of separate build plates when you hit "Slice All." Right now, even with a 16-core CPU, OrcaSlicer processes plates sequentially, and within a single plate the parallelism is partial. Don't let anyone tell you OrcaSlicer "scales perfectly to all your cores." It doesn't. RAM and CPU, what actually matters in 2026 Since the README publishes no minimum, the only honest way to frame RAM and CPU is by what real workloads stress. For CPU, the brutal example is Issue #9064, a real benchmark from a user testing organic supports. On an i5-1235U laptop with 16 GB and an RTX 2050 (irrelevant, since slicing's CPU-bound), they measured the same model: OrcaSlicer 2.2 sliced it in 3.8 seconds. OrcaSlicer 2.3 took 1 minute 48 seconds on the same hardware. PrusaSlicer 2.91 did it in 1.8 seconds. The 30x slowdown only happened during the "Generating Supports" phase when organic supports were enabled. The lesson is that single-thread CPU performance matters far more than core count for the kind of work that actually chokes the slicer, and that "more cores" doesn't always help. If your slices feel sluggish, our slow slicing fix guide covers the usual culprits (organic supports being one of the top offenders). Windows Task Manager performance tab during an OrcaSlicer slice of a model with organic supports For RAM, the situation's the opposite of what mirror sites suggest. Issue #9094 documents a user with 31.3 GiB of RAM whose OrcaSlicer 2.3.0 [truncated] --- ## How to Install OrcaSlicer on Linux: AppImage, Flatpak, AUR URL: https://orcaslicer.net/orcaslicer-install-linux/ The first time I installed OrcaSlicer on Linux, it was a fresh Debian 12 box, and I got a confident "Permission denied" the second I double-clicked the AppImage. Two minutes later, after I'd chmod'd it executable, I got the famous "dlopen(): error loading libfuse.so.2" splash, which is Linux's special way of saying "welcome, please read the docs." I've since installed OrcaSlicer on Ubuntu 22.04, Ubuntu 24.04, Pop!_OS, Fedora Workstation, EndeavourOS, and a Steam Deck in desktop mode, and every single one wanted something slightly different. That's Linux installs in a nutshell: there isn't one path, there are three, and the right one depends on your distro, your GPU, and how much you care about sandboxing. This guide walks you through all three official routes (AppImage, Flatpak, AUR) with the exact gotchas I've hit on real machines in 2026. Table of contents The Linux install picture in 2026 Pick your route: a quick chooser Route 1: AppImage Route 2: Flatpak from Flathub Route 3: AUR (Arch, Manjaro, EndeavourOS) Where OrcaSlicer stores your profiles The GLIBC 2.38 gotcha Wayland and NVIDIA: the one breaking combo System requirements that actually matter Common gotchas and fixes FAQ The Linux install picture in 2026, three routes, one binary Here's the situation. OrcaSlicer publishes three official Linux artifacts per release on GitHub, and there's a community-maintained set of AUR packages for the Arch crowd. There's no Snap, no signed .deb in the official assets, no .rpm, and no system tray installer that does everything for you. You pick the route that fits your distro and your tolerance for sandboxing. The three real options are the AppImage (the portable single-file binary), the Flatpak from Flathub (sandboxed, bundled runtime, cross-distro), and the AUR packages (if you're on Arch or a derivative). I'll walk through each one in detail, but if you just want a recommendation right now, here's the short version: if you're on a current Ubuntu, Pop!_OS, or Mint, the AppImage is fastest. If you're on Debian 12, an older Mint, Zorin, or anything immutable like Fedora Silverblue, jump straight to Flatpak. If you're on Arch, you already know to check the AUR. One thing worth flagging up front, because it bit me hard and it's biting a lot of people in 2026: the official AppImage filename literally includes Ubuntu2404 because that's the build host. That means it's linked against GLIBC 2.38 or newer, and if you're on Debian 12 (GLIBC 2.36), Linux Mint 21, or Zorin OS 17.x (Ubuntu 22.04 base), the AppImage will refuse to launch. There's a verified GitHub issue (#10990) tracking this. We'll come back to it, but Flatpak is the answer if you're stuck. Pick your route, a quick chooser Before we get into the commands, let me save you ten minutes of guessing. Here's how I'd pick if I were standing in front of your machine. You're on current Ubuntu (24.04+), Pop!_OS 24.04, or Mint 22: AppImage works fine, install libfuse2t64, and you're done in under a minute. You're on Ubuntu 22.04 LTS: AppImage works, but install libfuse2 instead (different package name on this release). Still fast. You're on Debian 12, Linux Mint 21, Zorin 17.x, or anything with GLIBC older than 2.38: skip the AppImage entirely. Use the Flatpak. It bundles its own runtime so your host GLIBC version stops mattering. You're on Fedora, Fedora Silverblue, Kinoite, openSUSE, NixOS, or any immutable distro: Flatpak. It's not even close. You're on Arch, Manjaro, or EndeavourOS: AUR. Specifically orca-slicer-bin unless you have a reason to compile from source. You're running NVIDIA proprietary drivers + Wayland session: any install route works, but you'll likely need to log out and pick the X11/Xorg session at your display manager. More on that in the Wayland section. That's the cheat sheet. Now let's get into each route. Route 1, AppImage: the official portable binary The AppImage is OrcaSlicer's flagship Linux artifact. It's a single file (~125 MB), ships with everything it needs except FUSE2 and a recent enough GLIBC, and you don't have to install it system-wide. I keep mine in ~/Applications/ and that's it. Grab it from the canonical GitHub releases page: github.com/SoftFever/OrcaSlicer/releases. As of v2.3.2 the asset you want is OrcaSlicer_Linux_AppImage_Ubuntu2404_V2.3.2.AppImage. Don't grab anything off random mirrors. The GitHub releases page is the only canonical source. Once it's downloaded, here's the three-step incantation. Open a terminal in your downloads folder and run: chmod +x OrcaSlicer_Linux_AppImage_Ubuntu2404_V2.3.2.AppImage ./OrcaSlicer_Linux_AppImage_Ubuntu2404_V2.3.2.AppImage If it launches, you're done. If it doesn't, you've hit one of two things: the FUSE error, or the GLIBC error. They look different. Let's handle them. The FUSE error: pick the right package for your Ubuntu version If your terminal spits back something like dlopen(): error loading libfuse.so.2 AppImages require FUSE to run, that's because Ubuntu (and a lot of derivatives) stopped shipping libfuse2 by default a couple of years back. AppImages still need FUSE2 specifically (not FUSE3, which is a different thing entirely, despite the name). The fix is one package, but the name depends on your Ubuntu release: Ubuntu 22.04 LTS (and derivatives based on it, like Pop!_OS 22.04 and Mint 21): sudo apt install libfuse2 Ubuntu 24.04+ (and Pop!_OS 24.04, Mint 22): sudo apt install libfuse2t64 That t64 suffix isn't a typo, it's the 64-bit time_t transition that Debian and Ubuntu pushed through in 2024, and a bunch of library packages got renamed as part of it. FUSE2 and FUSE3 happily coexist, so installing libfuse2t64 won't break anything else on your system that needs FUSE3. The OrcaSlicer team marked their original FUSE issue (#5482) as "not planned" because this is fundamentally an AppImage runtime requirement, not something OrcaSlicer can fix at their end. The official AppImage troubleshooting docs are the canonical reference if you want to read more. Desktop integration, optional but nice A raw AppImage doesn't show up in your application menu by default. If you want the OrcaSlicer icon in your launcher (and you probably do), install AppImageLauncher from your distro's repos. It handles registration, icon extraction, and updates automatically. You don't strictly need it, but I run it on every Linux machine I own because clicking an icon beats cd ~/Applications && ./Orca* every time. If you don't want AppImageLauncher, you can create a .desktop file by hand in ~/.local/share/applications/, but honestly, just install the launcher. It's five MB. Route 2, Flatpak from Flathub: the easiest cross-distro path If you're on an older Ubuntu derivative, Debian, Fedora, anything immutable, or you just like the idea of sandboxed apps, Flatpak is the path of least resistance. The OrcaSlicer team published the official Flathub listing as of v2.3.2, and the bundled runtime means your host distro's GLIBC version stops mattering entirely. That's huge if you've been chasing GLIBC errors on the AppImage. Before you can install anything from Flathub, you need to enable the Flathub remote on your distro. The Flathub setup pages cover this for every major distro, but the rough shape is: Ubuntu / Pop!_OS / Mint: install flatpak from apt, then add the Flathub remote (the Flathub Ubuntu setup page walks you through it). Debian: the Flathub Debian setup page covers the apt commands. Fedora: Flathub is already configured out of the box on Workstation, Silverblue, and Kinoite. Skip this step. Arch: the Flathub Arch setup page has the pacman commands. Once Flathub is enabled, the install is a single command. This is the part where the app ID change matters, so pay attention. The new canonical Flatpak ID, as of v2.3.2, is com.orcaslicer.OrcaSlicer. Older guides (and the original Flathub discussion thread) mention io.github.softfever.OrcaSlicer, which was the legacy ID. The v2.3.2 release notes explicitly call out that migration code is included, [truncated] --- ## OrcaSlicer Won't Find My Printer on the Network: Mac and Windows Fixes URL: https://orcaslicer.net/orcaslicer-wont-find-printer-network/ After three hours of clicking Scan on a Friday night and watching a blank Device tab stare back at me, I finally drank some coffee, opened a terminal, and admitted the problem wasn't OrcaSlicer's button. I'd just moved my Bambu P1S onto an IoT VLAN that morning, and the slicer on my Mac was happily sitting on the main LAN, asking the network politely if any printers wanted to say hi. None did. The printer was fine. The slicer was fine. The plumbing between them was the thing in pieces. If you're staring at the same blank Device tab right now, or a spinner that never resolves, or a Discover button that returns nothing, I've been there at least a dozen times across Bambu, Klipper, OctoPrint, and PrusaLink hosts. The good news: there are only really four root causes, and once you know which one bit you, the fix is usually five minutes. This guide walks through the diagnostic, then breaks fixes out by printer type and by operating system, with a by-IP fallback that works for every host so you can keep printing while you debug the proper fix. Table of Contents Why OrcaSlicer Can't Find Your Printer (the 4 Root Causes) The 30-Second Diagnostic Flowchart Bambu Lab Fixes: LAN Mode, Access Code, Bambu Connect Klipper, Mainsail, and Fluidd Fixes via Moonraker OctoPrint Fixes: API Key, Port 5000, Discovery Plugin PrusaLink Fixes: Port 80 and HTTP Digest The Connect-by-IP Fallback That Works for Everything Windows Fixes: Defender Firewall, Bonjour, Network Profile macOS Fixes: Firewall, Stealth Mode, Bonjour Linux Fixes: Avahi, UFW, NetworkManager Router and VLAN Fixes: AP Isolation, IoT VLAN, mDNS Reflector FAQ Why OrcaSlicer Can't Find Your Printer (the 4 Root Causes) I'll save you the suspense. Every Device-tab failure I've helped someone debug falls into one of these four buckets, and they're worth memorizing because the rest of this article is basically "which bucket are you in, here's the fix." One: authentication mismatch. The printer is reachable, the protocol is right, but the credential is wrong, stale, or got wiped. For Bambu this means the 8-character Access Code, for OctoPrint an API key, for PrusaLink an HTTP Digest username and password, for Klipper either a Moonraker API key or an entry in its trusted-clients list. I've seen this look exactly like a network problem when it's just a typo. Two: wrong protocol or port. OrcaSlicer talks to different hosts on different ports. Bambu LAN mode opens MQTT over TLS on 8883 and FTPS on 990, OctoPrint defaults to 5000, Moonraker to 7125, PrusaLink to 80. Point the Device tab at the wrong port and you'll get a perpetual spinner, even though the printer is on the LAN and pingable. Three: mDNS or Bonjour is broken. .local hostname resolution depends on multicast DNS, which uses UDP port 5353 on the multicast group 224.0.0.251. If that traffic is blocked by your firewall, missing on your operating system, or filtered by your router, then printer.local simply will not resolve, and OrcaSlicer's discovery scan returns nothing. The fix is almost always either a firewall rule, an Avahi or Bonjour install, or skipping the hostname and using a raw IP. Four: subnet or VLAN isolation. mDNS multicast doesn't cross subnets by default. A lot of consumer routers also have AP Isolation enabled on a guest SSID, or split 2.4 GHz and 5 GHz onto separate subnets. If your printer is on an IoT VLAN and your laptop is on the main LAN, you cannot ping each other, never mind discover each other, until you add either a static route or an mDNS reflector. That's it. Four. Work through the diagnostic below and you'll know which bucket within about thirty seconds. OrcaSlicer Device tab showing "No printer found" empty state The 30-Second Diagnostic Flowchart Open a terminal (PowerShell on Windows, Terminal on macOS, your shell of choice on Linux) on the same computer running OrcaSlicer. Don't skip this step. I've watched people spend an hour reinstalling OrcaSlicer when the problem was that their printer had silently lost Wi-Fi. Check 1: can you ping the printer's IP? Grab the IP from your router's DHCP client list or the printer's own touchscreen. Then run ping 192.168.1.45 (substituting your printer's address). If you get replies, the printer is on the network and reachable from this exact machine. If you get "Destination host unreachable" or "Request timed out", you have a layer-3 routing problem, and no slicer setting is going to fix it. Most likely culprits: VLAN isolation, AP isolation on the SSID, or the printer is actually offline. Check 2: can you open the printer's web UI in a browser? If you can ping the IP, try http://192.168.1.45 in your browser (for OctoPrint try http://192.168.1.45:5000, for Moonraker http://192.168.1.45:7125). If the web UI loads, your TCP path is clear and the problem is almost certainly inside OrcaSlicer (wrong host type, wrong credential, stale cached IP). If the web UI doesn't load but ping works, you have a firewall blocking that specific port, or the host service isn't running. Check 3: does the .local hostname resolve? Run ping printer.local or ping octopi.local (whatever your printer announces). If it resolves to an IP, mDNS is working on your machine. If it returns "cannot resolve host" or "Ping request could not find host", you have an mDNS problem. On Windows that usually means a firewall rule for the wrong network profile, on macOS it's usually stealth mode or a third-party firewall, on Linux it's Avahi not running. That's the diagnostic. Three commands, maybe ninety seconds. You now know which of the four root causes you're in, and you can jump to the right section below. A note on what OrcaSlicer's Discover button actually does: it scans the local subnet using a mix of mDNS queries (for OctoPrint, Klipper, PrusaLink) and SSDP multicast on UDP 2021 with multicast group 239.255.255.250 (for Bambu Lab printers in LAN mode). If your computer can't send or receive multicast on the printer's subnet, Discover returns nothing, even if you can ping the printer just fine. That's a really common source of confusion. Bambu Lab Fixes: LAN Mode, Access Code, Bambu Connect Bambu owners are the biggest cohort I see with this problem, and the picture got more complicated in January 2025 when Bambu shipped firmware 01.08.03.00 with a new authorization system. I'll walk through the whole binding sequence and then the post-firmware caveats. First, enable LAN-Only mode on the printer itself. On a P1S or A1, go to Settings → WLAN → LAN Only Mode on the touchscreen, toggle it on, and the screen will display an 8-character alphanumeric Access Code. Per the Bambu Lab Wiki, this code is formatted as eight characters in any combination of letters and numbers, and it's specifically for LAN-Only authentication (different from the original cloud-bound Access Code). Write it down. You'll need it in OrcaSlicer in a moment. LAN-Only Mode enabled and Access Code visible" class="wp-image" loading="lazy" /> Bambu Lab printer screen showing LAN-Only mode toggle and 8-character Access Code In OrcaSlicer, go to the Device tab. If the printer's already been bound, it'll show up as a tile. If it hasn't, click No Printer → Bind with Access Code, then enter the printer's IP (find it on the printer's About screen) and the 8-character LAN code. Hit confirm. If everything is healthy, the Device tab populates with temperatures, AMS status, and a preview pane within a few seconds. If binding fails, the most common cause I've found is a firewall blocking the Bambu ports. The verified ports you need open from your computer's subnet to the printer's subnet are 990/tcp for FTPS file upload, 8883/tcp for MQTT over TLS (control and status), 6000/tcp for the LAN-mode camera stream on P1 and A1 series, and 2024 through 2025/tcp for additional FTP-like traffic. Discovery happens over UDP 2021 with the SSDP multicast group 239.255.255.250. These port assignments come from an independent IoT VLAN write-up at nuxx.net that tested them with Bambu Stud [truncated] --- ## SimplyPrint vs Obico vs OctoEverywhere: OrcaSlicer 2026 URL: https://orcaslicer.net/orcaslicer-simplyprint-obico-octoeverywhere/ I've spent the last six months running OrcaSlicer with all three of these cloud services bolted on, sometimes two at the same time on the same Raspberry Pi, and I can tell you the marketing copy doesn't match the lived experience. I'm a maker with a Voron 2.4, a Bambu P1S, and an old Ender 3 running OctoPrint, and I've burned enough late-night hours debugging "why can't I see my webcam from the office" to have strong opinions about what actually works. If you're here, you've probably already hit the wall I hit. OrcaSlicer 2.3.x is a brilliant slicer, but its cloud story is short. The Device tab talks to Klipper through Moonraker, to OctoPrint through the host plugin, to Bambu Cloud and LAN mode, and to PrusaLink. That's it. There's no SimplyPrint button. There's no Obico tab. There's no OctoEverywhere checkbox. Anything you've read claiming otherwise is wrong, and I'll show you exactly why the integration story is identical across all three services in the next section. This isn't a hype piece. I genuinely use all three, I pay for one of them, I self-host another, and I bounce between them for different printers. Let's get into it. OrcaSlicer doesn't have remote-print cloud baked in How OrcaSlicer talks to SimplyPrint, Obico, and OctoEverywhere SimplyPrint deep dive Obico deep dive OctoEverywhere deep dive Side-by-side comparison table Security and privacy AI failure detection compared Which one should you pick for OrcaSlicer? FAQ OrcaSlicer doesn't have remote-print cloud baked in, and here's what that actually means I want to start with the architectural reality because it's the single most important thing in this article. OrcaSlicer is a slicer. It takes your STL, slices it, and hands the G-code to a printer or a host. The "cloud" part of remote printing, the part where you watch a webcam from your phone at the airport and pause the print because the bed scraped, isn't done by OrcaSlicer. It's done by the host running on the Raspberry Pi or single-board computer next to your printer, and by whatever cloud service has its tunnel open to that host. So when somebody tells you "OrcaSlicer added SimplyPrint support" in a forum post, they're either confused or they're describing the workflow incorrectly. What they probably mean is that they installed SimplyPrint on top of OctoPrint, then used OrcaSlicer's existing OctoPrint connector to upload to that host. The cloud part happened in the background, invisible to OrcaSlicer. Same story for Obico. Same story for OctoEverywhere. This matters because it shapes how you should evaluate the three options. They don't compete on "which one works better with OrcaSlicer," because they all work with OrcaSlicer identically. They compete on dashboard quality, AI failure detection accuracy, mobile experience, pricing, fleet features, and openness of source. That's the comparison this article is built around. If you haven't yet got your printer talking to OrcaSlicer through OctoPrint or Klipper, you'll want to fix that first. I've got a separate write-up that walks through the Device tab basics in our general OrcaSlicer troubleshooting hub. Come back here once you can hit "Send" from OrcaSlicer's Device tab and watch a print start. One more thing before we dive into each service. The three groups of people reading this article are different, and the right pick is different for each. If you run one or two printers and you just want a phone app plus AI failure detection without poking holes in your router, you're in group one. If you run a print farm with five to fifty printers and you need a fleet dashboard, you're in group two. If you're a Bambu owner shopping for a non-Bambu cloud option after Bambu's authorization-control changes shook things up, you're in group three. I'll flag which service suits which group as we go. How OrcaSlicer talks to SimplyPrint, Obico, and OctoEverywhere (it's always through OctoPrint or Klipper) The Device-tab handoff Here's what physically happens when you hit Send in OrcaSlicer. The slicer wraps your G-code into an upload request and pushes it over your local network to whatever host you configured in the Device tab. If that host is OctoPrint, OrcaSlicer talks to OctoPrint's REST API. If it's Klipper, OrcaSlicer talks to Moonraker, which is the HTTP API layer that sits on top of Klipper. If it's Bambu in LAN mode, OrcaSlicer talks to Bambu's local MQTT and FTP services. If it's PrusaLink, OrcaSlicer talks to Prusa's HTTP API. That's the universe of native connections. None of those endpoints care that you've also installed SimplyPrint, Obico, or OctoEverywhere alongside. From OrcaSlicer's perspective, you're sending a file to OctoPrint or Mainsail/Fluidd. From the host's perspective, the file lands in its uploads folder and gets queued like any other print. The cloud service only enters the picture once the print is running, because that's when the service's daemon, which has been quietly maintaining an outbound tunnel, starts streaming status, snapshots, and webcam frames to the cloud dashboard. Device tab with Moonraker and OctoPrint connection types visible" class="wp-image" loading="lazy" /> OrcaSlicer Device tab showing Klipper and OctoPrint host options Where each cloud service inserts itself All three services follow the same install pattern. You SSH into your Raspberry Pi, you install the service's plugin into OctoPrint or its system-package equivalent into the Klipper stack, you paste a one-time pairing code from the service's web dashboard, and the tunnel comes up. Once it's up, the service's cloud has a persistent outbound connection to your printer's host. You don't open ports. You don't set up dynamic DNS. You don't expose OctoPrint to the public internet. Everything goes through the service's relay servers. This outbound-tunnel pattern is the security model all three share, and it's why they all work behind carrier-grade NAT or hotel WiFi. If any third party tells you their service "uses UPnP to open a port," they're lying. The official docs for all three describe outbound tunnels, and you should trust those docs over random YouTube comments. If you're starting fresh and haven't picked OctoPrint or Klipper yet, the choice is a separate rabbit hole. The short version is OctoPrint is friendlier for most Ender-class printers, Klipper plus Mainsail/Fluidd is what you want for high-speed CoreXY printers and Voron builds. Either path lets you bolt on any of these three services later, so don't agonize. You can read my walkthrough on getting a Klipper printer connected to OrcaSlicer if you're leaning that way. The one exception, SimplyPrint's standalone host image SimplyPrint is slightly different because it offers its own Raspberry Pi image and its own host installer that can replace OctoPrint for many printers. If you go that route, your Pi runs SimplyPrint's host stack instead of OctoPrint, and SimplyPrint's marketing positions this as the simpler turnkey option for newcomers. SimplyPrint's homepage puts it this way: "By far most 3D printers work with SimplyPrint. The general rule of thumb is; if your printer works with OctoPrint / has a way to plug in a USB cable, it'll work with SimplyPrint." How OrcaSlicer's Device tab talks to a SimplyPrint standalone host is something you should verify in SimplyPrint's docs before assuming, because the exact endpoint surface wasn't something I could pin down to a quoted line. If you're already happy with OctoPrint, just install the SimplyPrint plugin on it and move on. SimplyPrint deep dive, the Danish print-farm dashboard I want to be upfront. SimplyPrint is the option I'd pick first if I ran more than three printers, and that's not a marketing line. The dashboard is genuinely the most polished of the three, and the fleet-management thinking is built into the product from the ground up rather than bolted on. The company is based in Dyssegard, Denmark (VAT DK41306505), which puts the service squarely under GDPR jurisdiction, and I'll co [truncated] --- ## OrcaSlicer Initial Setup Wizard: Picking Your Printer & Region (2026) URL: https://orcaslicer.net/orcaslicer-setup-wizard/ I clicked through OrcaSlicer's setup wizard twice in three months and got it wrong both times. The Region picker is the silent trap: pick the wrong one and Bambu users land on the Chinese cloud server, but the wizard explains none of this. Here's what each option actually does, and how to recover when you realize you picked wrong. This guide walks through every screen of the v2.3.2 wizard in order, with the actual screen names pulled from ConfigWizard.cpp in the OrcaSlicer source. I'll tell you what to pick, when it matters, and when it absolutely doesn't. If you're brand new and you also need the install steps that come before this, the Windows install guide and the macOS install guide cover that, and the download page is a step earlier still. Once you finish the wizard, the natural next stop is the beginner's first print walkthrough. Table of contents What the setup wizard actually is From installer to first launch Screen 1: Get Started Screen 2: Region (the one everyone gets wrong) Screen 3: Printer Selection Screen 4: Filament Selection Optional: Bambu Network Plug-in and login What happens after you click Finish Re-running the wizard later Common wizard pitfalls and fixes FAQ What the setup wizard actually is The setup wizard is a single modal window that pops the first time you launch OrcaSlicer. It's mandatory. There's no skip button, and there hasn't been one for years. Issue #10187 on GitHub asked for a skip option and it was closed as "not planned," and the older issue #4458 confirms the same. So you're going to walk through it whether you want to or not. The good news: it takes 60 to 120 seconds, and you can re-run it any time from Help → Setup Wizard if you mess up. The wizard has five screens in v2.3.2: Get Started, Region, Printer Selection, Filament Selection, and an optional Bambu Network Plug-in step that only shows up if you ticked a Bambu Lab printer. I've cross-referenced these against the source file src/slic3r/GUI/ConfigWizard.cpp and the v2.3.1 macOS issue thread #10970, which lists the live screen names exactly. The Help menu entry is the literal string _L("Setup Wizard") from MainFrame.cpp, in case anyone tells you it's called "Configuration Wizard" (that was the old PrusaSlicer-era name). One thing I want to defuse up front: nothing you pick in this wizard is permanent. You're not signing a contract. New selections later are additive, and there's a checkbox to wipe and reset profiles if it ever gets weird. From installer to first launch Before the wizard shows up, you've got to actually install OrcaSlicer. On Windows the installer asks where to drop the binaries, and I leave it at the default path under Program Files unless I'm specifically setting up a portable copy on a thumb drive. Default install location on Windows. Leave this alone unless you've got a reason not to. The installer is a standard NSIS-style wizard. It runs in maybe 15 seconds on an SSD. The "Choose Install Location" dialog you see above is the only screen most people actually need to read; everything else is Next-Next-Finish. If you're on macOS you're dragging the app to /Applications and there's no installer step at all, and Linux is whatever your distro's flavor of AppImage or .deb gives you. When the installer finishes, you'll get the Completing OrcaSlicer Setup dialog with a "Run OrcaSlicer" checkbox usually ticked by default. Click Finish. Hit Finish and OrcaSlicer launches straight into the wizard. That's the handoff. The instant the app opens, the configuration wizard takes the foreground and blocks the main window. You can't get to the slicing canvas until you finish. Screen 1: Get Started The first screen is called "Get Started" and the title bar reads "Welcome to the OrcaSlicer Configuration Wizard." It's mostly a language picker. There's a dropdown defaulting to English, and on Linux you'll also see a checkbox labeled "Perform desktop integration (Sets this binary to be searchable by the system)" which I always tick on a fresh Ubuntu box. On Windows and macOS that checkbox isn't there. What to do: pick your language if you don't want English, leave everything else alone, click Next. That's it. This screen takes five seconds. One landmine to know about: if you're on Linux running Wayland with integrated AMD graphics, the wizard window can come up totally blank. I've seen this reported in issues #4935 and PR #12159. The fix is launching with WEBKIT_DISABLE_COMPOSITING_MODE=1 as an environment variable. If you're on Windows or Mac this doesn't apply. If you ever previously installed OrcaSlicer and you're reinstalling onto the same user account, the Get Started screen also shows a "Remove user profiles (a snapshot will be taken beforehand)" checkbox. On a true first install it's hidden. Leave it unchecked unless your old profiles are corrupted and you specifically want a clean slate. Screen 2: Region (the one everyone gets wrong) Here's the screen I came here for. The Region picker shows five cards: China, Asia-Pacific, Europe, North America, and Others. The wizard doesn't tell you what they actually control. Most published guides hand-wave through it. I'll be blunt. The honest version: this screen primarily controls Bambu Lab cloud routing. If you don't own a Bambu Lab printer, this screen is functionally meaningless. Pick whatever's geographically closest and move on. If you do own a Bambu, it matters a lot, because the wrong pick breaks cloud binding silently. What each region option actually does RegionBambu cloud endpointMakerWorld URLWhat changes if you pick it Chinabambulab.cnmakerworld.com.cnChina-only API gateway. Required for printers bought in mainland China. North Americabambulab.commakerworld.comDefault global endpoints, US/Canada CDN. Europebambulab.commakerworld.comSame as NA, mild EU CDN routing difference. Asia-Pacificbambulab.commakerworld.comSame as NA, APAC CDN routing. Othersbambulab.commakerworld.comCatch-all bucket, functionally identical to NA. If you own a Bambu Lab printer Pick the region that matches the country where you bought the printer, not where you live right now. If you imported an X1C from China and you're in the US, you may still need to pick China to bind to the original cloud account it was registered to. China users especially: pick China or Bambu cloud login fails with no useful error. I've seen people spend an hour troubleshooting "wrong cloud account" errors that were just a wrong wizard pick. If you're on a global X1 Carbon, A1, A1 mini, P1S, or P1P bought from a Western retailer, North America or Europe is correct. For more on what happens after this when you're choosing between cloud and LAN-only mode, the Bambu Cloud vs LAN guide covers the actual networking implications. If you own anything else Creality, Prusa, Anycubic, Elegoo, Voron, Sovol, Qidi, Snapmaker, Flashforge, Raise3D, BIQU, Ratrig, anything that's not a Bambu Lab box: pick the region closest to your continent and don't think about it again. It changes essentially nothing for you. The Obico walkthrough confirms this bluntly, stating the region "is only relevant to those machines" (Bambu). Most published tutorials make this screen sound life-or-death. It isn't for 80% of users. Can I change it later? Yes. Two ways: re-run the wizard from Help → Setup Wizard, or open Edit → Preferences and change the Login Region directly. The preferences route doesn't reset anything else, and it doesn't touch your printer or filament profiles. So if you realize a week from now you picked wrong, fix it in 10 seconds. Screen 3: Printer Selection This is the screen that actually matters. You're looking at a grid of vendor brand cards, and OrcaSlicer v2.3.2 ships with 67 vendor brands baked in. I counted by listing the resources/profiles/ folder on the main branch. That includes Bambu Lab (filed under BBL in the source folder), Creality, Prusa, Anycubic, Elegoo, Voron, Sovol, Qidi, Flashforge, Snapmaker, FLSun, Raise3D, Artillery, BIQU, Ratrig, [truncated] --- ## How to Install OrcaSlicer on macOS: Apple Silicon & Intel (2026) URL: https://orcaslicer.net/orcaslicer-install-macos/ Last updated: May 2026 I plugged my M2 MacBook in, downloaded OrcaSlicer, opened the DMG, dragged it to Applications, double-clicked, and macOS told me the app was damaged and should be moved to Trash. The app isn't damaged. macOS just doesn't trust unsigned binaries, and that warning has been the same generic "damaged" message since Catalina. Here's the actual install path, end to end, for both Apple Silicon and Intel Macs in 2026. I've installed OrcaSlicer on three Macs at this point: an M1 mini I keep on my print bench, an M2 MacBook Air I write on, and an old 2018 Intel MacBook Pro I keep around because it still runs Final Cut faster than I expect it to. The flow is the same on all three, but the dialog macOS throws at you differs slightly between macOS releases, and the gotchas on Intel are different from the ones on Apple Silicon. This guide covers everything I hit and everything readers email me about. If you just want the executive summary: download from the official GitHub releases page, drag to Applications, right-click and choose Open the first time, and if it still says "damaged," paste one Terminal command. That's it. The rest of this article is the why and the edge cases. For the bigger picture on download sources and version differences, see the OrcaSlicer download hub. On Windows? The Windows install guide covers the SmartScreen flow over there. Table of contents Before you install: what you actually need Apple Silicon or Intel: which Mac do you have? Where to download OrcaSlicer for Mac Install OrcaSlicer on Apple Silicon (M1, M2, M3, M4) The "damaged and can't be opened" message: what it really means The xattr command and its variants Install OrcaSlicer on an Intel Mac First-run setup: printers, filament, region Common install problems and fixes Where OrcaSlicer stores its files on macOS How to uninstall OrcaSlicer cleanly Is it safe to install OrcaSlicer on Mac? FAQs Before you install: what you actually need I'll keep this honest. OrcaSlicer isn't a heavy app by modern standards. If your Mac can run a current copy of macOS at all, it can probably run OrcaSlicer. But there are three hard requirements worth checking before you waste time on a download that won't open. System requirements (the realistic version) The official DeepWiki page lists macOS 11.3 Big Sur as the floor, 4 GB RAM as the minimum, and OpenGL 2.0 as the GPU baseline. I think those numbers are technically right and practically misleading. Yes, the app will launch on 4 GB. No, you don't want to slice a 12-plate Voron print on it. Here's what I actually recommend: macOS: 11.3 Big Sur or newer. Monterey, Ventura, Sonoma, Sequoia and macOS 26 all work. If you're on Catalina 10.15 or Mojave 10.14, you're stuck. Update macOS first or stay on an old OrcaSlicer release. RAM: 4 GB minimum (per the project's own docs), 8 GB recommended in practice, 16 GB+ if you slice large multi-material plates often. Disk: About 1 GB for the install itself, plus another 5 GB free for cached profiles, sliced G-code and STL imports. The app footprint is small. The clutter accumulates. GPU: OpenGL 2.0 capable for the bare minimum, OpenGL 3.3+ for the full 3D preview. Every Apple Silicon Mac clears this comfortably. Any Intel Mac from 2015 onward does too. Display: 1280x720 minimum, 1920x1080 recommended. You'll cry trying to use the slicer at 720p, so consider this a real recommendation, not a tick-box. A note on the source you download from I'll repeat this throughout the article because it matters. The only place you should be downloading OrcaSlicer for Mac is the official GitHub releases page at github.com/SoftFever/OrcaSlicer/releases. Some search results push you toward random mirror sites with rebranded installers. Skip them. The official repo moved organisations at one point, so github.com/OrcaSlicer/OrcaSlicer/releases also resolves, but they're the same source. If you want a deeper breakdown of why this matters, I covered it in is OrcaSlicer safe to install. The official OrcaSlicer repo on GitHub. This is where the Releases page lives. Click the Releases link on the right sidebar to get to the DMGs. Apple Silicon or Intel: which Mac do you have? I get this question more than any other when people DM me about install issues. Half the time, the person doesn't know which CPU is in their Mac. That's fine. Apple stopped making the answer obvious somewhere around 2020. Here's the 15-second test: Click the Apple menu (the apple icon, top-left corner of your screen). Choose About This Mac. Look for either Chip or Processor in the panel that opens. If the line says Chip: Apple M1 (or M2, M3, M4, M1 Pro, M2 Max, M3 Ultra, etc.), you have an Apple Silicon Mac. Grab the Apple Silicon build. If it says Processor: 2.6 GHz 6-Core Intel Core i7 (or any Intel Core or Xeon variant), you have an Intel Mac. Grab the Intel build. There is no third option in the current Mac lineup. What you see in About This MacChip familyWhich DMG to download Chip: Apple M1 / M1 Pro / M1 Max / M1 UltraApple Silicon (arm64)Apple Silicon build Chip: Apple M2 / M2 Pro / M2 Max / M2 UltraApple Silicon (arm64)Apple Silicon build Chip: Apple M3 / M3 Pro / M3 MaxApple Silicon (arm64)Apple Silicon build Chip: Apple M4 / M4 Pro / M4 MaxApple Silicon (arm64)Apple Silicon build Processor: Intel Core i5 / i7 / i9 / XeonIntel (x86_64)Intel build Apple shipped the first M1 Macs in November 2020. Anything you bought new from late 2020 onward is almost certainly Apple Silicon. The last Intel iMacs trickled into the channel until mid-2021, but most consumer portables and desktops sold after 2020 are on Apple's own silicon. If you're not sure when you bought it, just check About This Mac. It's faster than digging through receipts. Where to download OrcaSlicer for Mac Open Safari (or Chrome, or Arc, the browser doesn't matter) and go to github.com/SoftFever/OrcaSlicer/releases. The latest stable release sits at the top. As of this writing it's V2.3.2, but you should always grab whatever's marked as the latest stable. Skip pre-releases and beta builds unless you have a specific reason to test them. Scroll down to the Assets section under the release notes. This is where the actual download files live. And here's something worth flagging: the current OrcaSlicer Mac DMG is a single universal binary. The 2.3.2 release ships one file that runs natively on both Apple Silicon and Intel Macs. Older releases (1.x, early 2.x) split this into two separate DMGs labelled "AppleSilicon" and "Intel," and you'll still see references to that split in older threads and tutorials. If you're looking at a screenshot or video from 2023 or 2024 that shows two Mac DMGs in the Assets list, that was true at the time. It isn't now. The README's Important Security Alert section. Read it once. It explains what the ad-hoc signature is and why the install flow involves a Gatekeeper bypass. Click the DMG link. Safari will start the download. The file's around 120 MB, so on a decent connection it's done in well under a minute. Once it's in your Downloads folder, you're ready to install. The Assets list on an older OrcaSlicer release page. This screenshot shows the previous split, with separate AppleSilicon and Intel DMGs. The current 2.3.2 release ships one universal DMG instead, but the download position is the same: scroll to Assets, click the DMG link. What about Homebrew? If you live in Terminal anyway, you can install OrcaSlicer via Homebrew with one command: brew install --cask orcaslicer Homebrew handles the download and the move-to-Applications for you. The cask is labelled "Mac_universal," so it pulls the universal DMG. I use this on my dev machines because it's faster than clicking through Safari, but it's not magically better than the manual path. Both end with the same OrcaSlicer.app in the same Applications folder. Install OrcaSlicer on Apple Silicon (M1, M2, M3, M4) This is the path 80% of readers are on. The flow is short. Five steps, no Terminal required [truncated] --- ## How to Install OrcaSlicer on Windows 10 & 11 (Step-by-Step 2026) URL: https://orcaslicer.net/orcaslicer-install-windows/ I told a friend to install OrcaSlicer last week and he hit five different speed bumps before getting it running: SmartScreen blocked it, Defender quarantined the .exe, the installer wanted admin rights he didn't have on his work laptop, the first-run wizard wouldn't accept his printer, and the bambu_network plugin offered itself before he understood what it did. That's a lot for a free slicer. Here's the version where none of that catches you off-guard. I've installed OrcaSlicer on something like thirty Windows boxes over the last two years, across both 10 and 11, on everything from a five-year-old work Dell to a fresh AMD desktop. The install is genuinely simple, but Windows does its best to make you doubt that. If you just downloaded the installer and Windows yelled at you in big blue text, you're in the right place. This guide walks through the installer route, the portable route, the SmartScreen panic, the antivirus false positives, the first-run wizard, and how to uninstall cleanly when you're done experimenting. I'll flag the spots where I'd actually slow down and double-check, and the spots that look scary but really aren't. One quick note before we start. orcaslicer.net doesn't host the OrcaSlicer binaries. Every download link in this guide points at the official GitHub releases page, because that's the only place I'd ever pull the installer from. If you haven't grabbed the file yet, our OrcaSlicer download hub has direct links to the latest stable, beta, and nightly builds along with file sizes and changelogs. On a Mac? Skip this entirely and follow the macOS install guide instead. The flow is similar but the Gatekeeper warning works differently. Table of contents What you're actually installing System requirements for Windows 10 and 11 Step 1: Download the right file from GitHub Verifying the download is genuine Installing OrcaSlicer with the .exe installer Installing the portable version Why Windows shows "unrecognized app" and what it means Defender and antivirus false positives First-run setup wizard Common install problems and fixes How to uninstall OrcaSlicer cleanly Other install routes (winget, Chocolatey) FAQ What you're actually installing OrcaSlicer ships two Windows artifacts in every release. They're the exact same application. The only difference is plumbing: how it lands on your disk and how Windows knows about it. I'll get the trade-off out of the way first, because picking the wrong one wastes ten minutes and a small amount of patience.   Installer (.exe) Portable (.zip) Filename (v2.3.2) OrcaSlicer_Windows_Installer_V2.3.2.exe OrcaSlicer_Windows_V2.3.2_portable.zip Download size 124.9 MB 157.7 MB Needs admin rights? Yes (UAC prompt) No Default location C:\Program Files\OrcaSlicer\ Wherever you extract it Start Menu shortcut? Yes (automatic) No Add or Remove Programs entry? Yes No .3mf file association? Optional during install Not registered User profiles stored at %APPDATA%\OrcaSlicer\ %APPDATA%\OrcaSlicer\ (yes, really) Triggers SmartScreen? Yes Yes Notice that last row. Both versions trigger the same SmartScreen panel because they're both unsigned executables from a developer Microsoft hasn't seen before. I'll get to why that happens in a bit. The other thing to call out: the portable build is not really portable across machines by default. It still reads and writes your profiles to %APPDATA%\OrcaSlicer\ on whatever PC you launch it from, which means if you copy the folder to a USB stick and plug it into a different PC, your settings don't come along. The fix is a --datadir shortcut flag, and I'll cover it in the portable section. Which one should you pick For 9 out of 10 people, pick the installer. You want OrcaSlicer to feel like a normal Windows program. You want it in your Start Menu. You want to double-click a .3mf and have it open. The installer does all of that. I'd pick portable only if one of these is true: you're on a locked-down work laptop where you don't have admin rights, you want OrcaSlicer running off a USB stick, or you want to test multiple OrcaSlicer versions side by side without them stepping on each other. That last one is genuinely useful if you're chasing a regression between two releases. System requirements for Windows 10 and 11 OrcaSlicer is not a heavy app by modern standards, but it does ask for a real GPU. The wiki's "official" minimums are conservative, and the practical floor is a bit higher. I'll list both. Spec Minimum Recommended OS Windows 10 64-bit (build 1809 or later) Windows 11 CPU Any 64-bit multi-core with SSE4.2 Modern 6+ core, AVX2 support RAM 4 GB 8 GB (16 GB+ for big models) GPU / OpenGL OpenGL 2.0 OpenGL 3.3 or newer (4.5 for smoothest preview) Disk space 1 GB for the app 5 GB to allow for project files and profile bundles Display 1280x720 1920x1080 or higher About OpenGL. The repo wiki says 2.0 minimum. I've watched OrcaSlicer hang on splash on machines that "technically" had OpenGL 2.0, so in practice I'd treat 3.3 as the real floor. Any Intel HD Graphics chip from 2012 or later clears that. Any discrete GPU from the last decade clears 4.5 without thinking. If you're on a really old laptop with an Intel GMA or first-gen HD, you're going to have a bad time and you should consider PrusaSlicer instead. Two runtime dependencies are worth knowing about because they show up as cryptic error messages if missing: Microsoft Visual C++ Redistributable 2015 to 2022. This is what supplies VCRUNTIME140.dll, VCRUNTIME140_1.dll, and MSVCP140.dll. Most modern Windows boxes already have it from Steam, Office, or another app. If OrcaSlicer launches with a "VCRUNTIME140.dll not found" pop-up, you don't have it. Grab vc_redist.x64.exe from aka.ms/vs/17/release/vc_redist.x64.exe and run it. Microsoft Edge WebView2 Runtime. Bundled on Windows 11, usually arrives via Windows Update on Windows 10. OrcaSlicer uses it to render the MakerWorld and printer dashboard panels. If it's missing, those panels are blank instead of crashing the app, so you might not notice immediately. You also want an internet connection for the first launch. OrcaSlicer pulls vendor profile bundles from GitHub during the first-run wizard. After that, it runs fine offline. Step 1: Download the right file from GitHub The official download page is github.com/OrcaSlicer/OrcaSlicer/releases. Type that URL by hand the first time you visit it, then bookmark it. I cannot stress this enough: there are at least three lookalike sites (orca-slicer.com, orcaslicer.com, orcaslicer.uk.com) that are not affiliated with the project, and the first time you Google "orcaslicer download" the ad slot at the top is a coin flip on which random domain shows up. Don't click ads. Go straight to GitHub. At time of writing the latest stable is OrcaSlicer v2.3.2, released March 23, 2026. Scroll to the "Assets" section under the latest release. You'll see something like this: The Assets list on the OrcaSlicer release page. Grab the file with "Windows_Installer" in the name. Click OrcaSlicer_Windows_Installer_V2.3.2.exe for the installer (124.9 MB) or OrcaSlicer_Windows_V2.3.2_portable.zip for the portable build (157.7 MB). Edge or Chrome will probably warn you "this file isn't commonly downloaded." That's the browser's reputation system, not your antivirus. Click the three-dot menu on the download bar and choose Keep, then Keep anyway. If you skip this step the download gets quarantined into a folder you'll struggle to find later. Note that OrcaSlicer mirrors its releases between two GitHub orgs: SoftFever/OrcaSlicer (the original) and OrcaSlicer/OrcaSlicer (the more recent mirror). Both point to the same binaries. Either is fine. Verifying the download is genuine Here's where I have to be honest with you. OrcaSlicer does not officially publish SHA-256 checksums on its release page. The feature was requested back in Issue #7005 and closed as "Not Planned." That's a legitimate gap in the project's release hygiene, and pretending otherwise would be a [truncated] --- ## OrcaSlicer Download: Latest Version, Safe Sources & What's New (2026) URL: https://orcaslicer.net/orcaslicer-download/ I've installed OrcaSlicer four times this year on three different machines and the same thing happens every time: the antivirus flags it, Microsoft tells me it might be malware, my friend asks if it's safe. The short answer is yes, but the long answer is that the download experience is genuinely confusing for anyone who hasn't done it before. This is the unfiltered version of what version to grab, where to grab it, and what's new in the current 2.3.2 line. I'll walk you through every step, every warning dialog you're going to hit, and exactly which of the six download files on the GitHub releases page you actually want. No filler, no marketing fluff, and definitely no pretending the SHA-256 hash situation is something it isn't. Table of contents Download OrcaSlicer (latest stable: v2.3.2) Which file do I actually need? Is OrcaSlicer safe to download? What's new in OrcaSlicer 2.3.2 Stable or nightly: which build should you grab? Installing it once you've downloaded Common problems and the fixes that actually work Updating from an older version without wrecking your profiles Should you actually be using OrcaSlicer? System requirements OrcaSlicer download FAQ Download OrcaSlicer (latest stable: v2.3.2) The current stable release is OrcaSlicer v2.3.2, published on the GitHub releases page on March 23, 2026. If you're on this page because you want the file, here's the file. Pick your operating system, click through to the official release page, and grab the matching asset: Get the latest stable build here: github.com/OrcaSlicer/OrcaSlicer/releases/latest That's the only URL I'd trust. Not orcaslicer.net (this site, which is a third-party content guide and doesn't host any binaries), not orca-slicer.com, not orcaslicer.info, not any of the dozen lookalike domains Google will happily serve you. The OrcaSlicer maintainers publish builds directly to GitHub and link to that page from their own official site at orcaslicer.com. Anywhere else is somebody else's mirror at best and somebody else's repackaged installer at worst. Here's exactly what's on the release page right now: The v2.3.2 release notes on GitHub. Skim the "What's Changed" prose before downloading, especially if you're updating from 2.3.0 or earlier. Platform Filename Size What it is Windows (installer) OrcaSlicer_Windows_Installer_V2.3.2.exe ~119 MB Standard installer, registers file associations and Start menu entries. Windows (portable) OrcaSlicer_Windows_V2.3.2_portable.zip ~150 MB Extract and run. Settings sit next to the .exe. Good for USB drives. macOS (universal) OrcaSlicer_Mac_universal_V2.3.2.dmg ~214 MB One DMG covers both Apple Silicon and Intel. No more "pick the right architecture." Linux (AppImage) OrcaSlicer_Linux_AppImage_Ubuntu2404_V2.3.2.AppImage ~119 MB Built on Ubuntu 24.04 base. Needs libfuse2t64 on newer Ubuntu. Linux (Flatpak x86_64) OrcaSlicer-Linux-flatpak_V2.3.2_x86_64.flatpak ~114 MB Flathub-distributed sandboxed build for desktop x86_64 systems. Linux (Flatpak aarch64) OrcaSlicer-Linux-flatpak_V2.3.2_aarch64.flatpak ~110 MB ARM64 Flatpak. For Raspberry Pi 4/5 desktops or ARM workstations. Six files. Most readers only need one. If you're not sure which, the next section breaks it down. Which file do I actually need? The release page lists six assets and that's a lot of choice for what should be a one-click action. I've made every one of these decisions wrong at some point, so here's the picker I wish I'd had the first time. Windows If you've never installed OrcaSlicer before, you want OrcaSlicer_Windows_Installer_V2.3.2.exe. Double-click, click through the wizard, done. It installs to C:\Program Files\OrcaSlicer\, drops a shortcut on your Start menu, and registers .3mf and .stl file associations so double-clicking those files opens OrcaSlicer. Your user profiles, presets, and printer configs live in %APPDATA%\OrcaSlicer\, which expands to C:\Users\<you>\AppData\Roaming\OrcaSlicer\. The installer bundles Microsoft Visual C++ Redistributable 2019 and the WebView2 runtime if you don't already have them, so you don't need a separate download. If you've got a fresh install of Windows and the installer asks to install those first, let it. The portable ZIP is for a narrower audience. I use it when I want to run a stable build and a nightly side by side without one stomping on the other's settings, or when I'm testing on a machine I don't have admin rights on. Extract the ZIP anywhere, double-click orca-slicer.exe, and OrcaSlicer reads and writes its settings to a data/ subfolder right next to the executable. That makes it portable in the literal sense: drop the folder on a USB stick and your full configured slicer goes with you. If you're a regular user just trying to slice some prints, take the installer. I'm not going to talk you into the portable build unless you specifically need it. macOS Grab OrcaSlicer_Mac_universal_V2.3.2.dmg. That's it. The "universal" in the filename means it works on both Apple Silicon Macs (M1, M2, M3, M4) and older Intel Macs in a single binary. If you were here a year ago and remember picking between an arm64 DMG and an x86_64 DMG, that split ended with the 2.3.1 release. The 2.3.2 line ships one DMG for everybody. Mount the DMG, drag OrcaSlicer to your Applications folder, eject the DMG. The first launch will throw a Gatekeeper warning, which I'll cover in the safety section, but the fix is a single command and you'll be slicing within two minutes. If you live in your terminal and use Homebrew, you can also do brew install --cask orcaslicer and Homebrew handles the download and install for you. That cask conflicts with orcaslicer@nightly, so you can only have one of the two installed at a time through Homebrew. Linux Linux is where it gets opinionated. There are three reasonable paths and you should pick based on your distro and how much sandboxing you want. Flatpak via Flathub is what I recommend to almost everyone now that the maintainers have made it the official Linux distribution channel. Run flatpak install flathub com.orcaslicer.OrcaSlicer, then launch with flatpak run com.orcaslicer.OrcaSlicer or from your application menu. Updates are handled by your system's Flatpak runner, libraries are bundled and isolated, and you don't fight your distro's package versions. The one gotcha: USB serial access to a printer is sandboxed, so if you're using a direct USB connection (rather than network or SD card), you'll need to grant device permissions via Flatseal or the --device=all flag. AppImage is the right choice if you want a single self-contained file you can copy to any machine. It's built on Ubuntu 24.04, so on newer distros it just runs. On Ubuntu 22.04 or Mint 21, you'll need to install libfuse2 (older releases) or libfuse2t64 (Ubuntu 24.04 renamed the package). I'll cover the troubleshooting steps in the problems section below. AUR is for Arch and Manjaro. You've got four options: orca-slicer-bin (pre-built, easiest), orca-slicer (builds from source), orca-slicer-git (latest main branch), and orca-slicer-nightly-bin (pre-built nightly). I'd pick orca-slicer-bin unless you have a specific reason to want otherwise. The quick picker Your situation File to grab First-time Windows userWindows installer .exe Want stable + nightly side by side on WindowsPortable ZIP Mac (Apple Silicon or Intel)Mac universal DMG Mac with Homebrewbrew install --cask orcaslicer Generic Linux desktopFlatpak via Flathub Linux, want one self-contained fileAppImage Arch / Manjaroyay -S orca-slicer-bin Raspberry Pi 4/5 desktop or ARM workstationaarch64 Flatpak Is OrcaSlicer safe to download? Yes, with caveats that you should actually understand. Let me walk you through what's real and what's noise. OrcaSlicer is open-source software licensed under AGPL-3.0. The source code lives at github.com/OrcaSlicer/OrcaSlicer and anyone, including security researchers, can read every line of it. The project is a fork of Bambu Studio, which is itself a fork of PrusaSlicer, whi [truncated] --- ## How to Connect OrcaSlicer to OctoPrint: IP, API Key, and the Common Errors URL: https://orcaslicer.net/orcaslicer-octoprint-setup/ I still remember the first time the Test button finally went green. An old Ender 3 V2 on the bench, a Pi 3B+ underneath running OctoPi 1.10, OrcaSlicer freshly installed on the laptop next to it. I'd pasted the IP wrong twice, forgotten the protocol once, and grabbed the API key with a trailing newline before catching it. On the fourth try, that little green checkmark popped up with "OctoPrint 1.10.3 detected," and the rest of the afternoon felt smooth. This guide is the version of that afternoon I wish I'd read first. I'll walk you through wiring OrcaSlicer to an OctoPrint instance on a Raspberry Pi, where the API key actually lives in current OctoPrint (not where the older guides say), what the Octo/Klipper Host Type really means, and the error strings you're most likely to see if something goes sideways. If your problem is part of a wider OrcaSlicer issue, the OrcaSlicer troubleshooting master covers the bigger map. This article zeroes in on OctoPrint. Quick honesty pass: OctoPrint is the right answer for an Ender 3 family machine, an SV06, an MK3S+, a CR-10, a Voron with caveats. It's the wrong answer for a Bambu Lab printer, a Creality K1, a V3 KE, an MK4, or a Core One. I'll cover that in the per-printer section. If you're not on the latest OrcaSlicer, the download page is worth a stop first; the Print Host integration improved a lot since 2.0. What's in this guide Why OctoPrint plus OrcaSlicer is still a good combo in 2026 Before you start: OctoPi, IP or mDNS, and the Application Keys plugin The 60-second version Generate the API key in OctoPrint Add the printer in OrcaSlicer (Host Type: Octo/Klipper) Slice and send your first job Verify the upload arrived (the "User: API client" check) Per-printer support: who can use OctoPrint, who can't When to use PrusaLink, Mainsail, or native Bambu instead Common errors during setup After it works: camera, Obico, OctoEverywhere Frequently asked questions Why OctoPrint plus OrcaSlicer is still a good combo in 2026 OctoPrint turns 13 this year. It still runs on a humble Pi 3B+ with 1GB of RAM, still accepts G-code over a small REST API, and still gives you a webcam stream, plugins, and a job queue without a cloud subscription. For the Marlin-based bedslinger crowd, nothing else gets close on cost-per-feature. OrcaSlicer's contribution is the upload pipe. When you click Print or Send, OrcaSlicer hits OctoPrint's REST API on `/api/files/local` with an `X-Api-Key` header and (if you used Print) tells OctoPrint to start the job. No proprietary protocol, no telemetry, no slicer-side websocket. The Device tab is a Chromium webview pointed at OctoPrint's own UI, which is why it shows the same temperatures and webcam feed as a regular browser tab. The reason this combo is still relevant in 2026, honestly, is the alternatives haven't replaced it for the older-machine crowd. Klipper plus Mainsail is faster and prettier, but it's a config-file-and-flash-the-MCU lift not every Ender 3 owner wants to take on. PrusaLink is great if you have a Prusa, useless if you don't. So OctoPrint keeps doing the job. Before you start: OctoPi, IP or mDNS, and the Application Keys plugin Five things need to be true: OctoPi (or a manually installed OctoPrint) is up and serving its web UI. You can open it in a browser from the machine you'll run OrcaSlicer on. The Pi has a stable address. Either `octopi.local` resolves over Bonjour/mDNS, or you've pinned the Pi to a static lease in your router's DHCP table. Access Control is enabled in OctoPrint (default since 1.4), and you have an admin user. OctoPrint is at least 1.5, ideally 1.10.x. I tested the steps below against 1.10.3. Your printer is one OctoPrint supports. If you're on a Bambu, K1, or MK4, skip to the per-printer section first. On the address question, `octopi.local` works on Mac and Linux out of the box, and on Windows 10/11 if Bonjour Print Services is installed. On guest WiFi or a managed network, mDNS is often blocked, in which case you need the actual IP. The cleanest way to find it: open your router's admin page, look at the DHCP table, find the device whose vendor MAC starts with "Raspberry Pi Trading." Set a DHCP reservation for it. I cannot stress this enough. If the Pi grabs a new IP after a router reboot, OrcaSlicer will silently fail and you'll waste 20 minutes thinking the API key got revoked. The 60-second version If you've done this before on PrusaSlicer or Cura and you just need the OrcaSlicer-specific bits, here's the compressed version: OctoPrint browser tab open. Wrench icon (Settings) > Application Keys > Generate new application key, name it `OrcaSlicer`, click Generate, copy the string. OrcaSlicer top-right, click the small WiFi icon next to the printer name in the printer dropdown. The Physical Printer dialog opens. Host Type: Octo/Klipper. Hostname: `http://octopi.local:5000` or `http://192.168.1.42:5000`. API Key / Password: paste the key. Click Test. Green check, OctoPrint version string. Click OK. Slice. Click the dropdown next to Print, pick Send (uploads only) or Print (uploads and starts). If that worked, you're done. If it didn't, keep reading; the failure modes are the meat of this article. Generate the API key in OctoPrint This is where the older guides are wrong, so be careful. There are two API key concepts in OctoPrint and the difference matters. The legacy one is the global API key under Settings > API. It grants full account access to whatever user it's tied to. It still works. The docs recommend against it now, because if it leaks you've handed over the printer farm. The modern one is the Application Keys plugin, bundled with OctoPrint since 1.5. It lets you mint per-app, per-user keys with names, so you can revoke just OrcaSlicer's key without breaking anything else. This is what you want. To generate one: open OctoPrint, log in, click the wrench (Settings), then click the Application Keys entry in OctoPrint's Settings sidebar. In "Generate new application key," type `OrcaSlicer` and click Generate. A long alphanumeric string appears with a copy button. Use the copy button. Don't triple-click and copy by hand; that's how trailing whitespace sneaks in, and whitespace silently fails with a 403. One upgrade for any rig that lives on the home network long term: under Settings > Access Control, create a dedicated `slicer` user with only three permissions (Status, File Upload, Print), then generate the Application Key against that user. If the key leaks, the worst someone can do is upload a benchy and start it. Add the printer in OrcaSlicer (Host Type: Octo/Klipper) Now hop over to OrcaSlicer. The connection settings live deeper than you'd expect; they're not in the main Preferences panel. Printer Settings is the same area where Print Host lives. The WiFi-icon shortcut next to the printer name in the top-right sidebar opens this same dialog directly. The fastest entry point is the WiFi icon to the right of the printer name in the printer dropdown at the top of OrcaSlicer's right sidebar. It's small, about 16 pixels wide on a 1080p monitor. Click it and the Physical Printer dialog opens. Fill it like this: Host Type: Octo/Klipper. Yes, that name is deliberate. The Octo/Klipper option covers both OctoPrint and Moonraker (Klipper's API server), because Moonraker ships an `octoprint_compat` shim that speaks enough of OctoPrint's REST API for slicer integrations to work. Hostname, IP or URL: full URL with protocol. `http://octopi.local:5000` is the standard. With a static IP, `http://192.168.1.42:5000`. Don't omit the `http://`; OrcaSlicer is strict about it. API Key / Password: paste the application key. Verify no leading or trailing space. Use HTTPS: leave unchecked unless you've set up a reverse proxy with a real cert. Self-signed certs will fail OrcaSlicer's bundled curl validation. HTTP Authorization (Username / Password): leave blank. These are for HTTP Basic auth from a reverse proxy, not OctoPrint's own login. Click Test. A green check wit [truncated] --- ## How to Connect OrcaSlicer to Klipper (Mainsail and Fluidd): The 2026 Setup URL: https://orcaslicer.net/orcaslicer-klipper-setup/ I spent a Saturday afternoon last spring chasing a ghost on my Voron 2.4. Mainsail loaded fine in the browser, the printer was happy, the toolhead homed without complaint. OrcaSlicer just kept refusing to connect. The Test button stared back with a polite "Connection failed" message, and I'd already double-checked the IP three times. Turns out I'd pasted Mainsail's URL into the Hostname field. That's the small mistake almost every Klipper newcomer makes once, because OrcaSlicer's connection dialog has two URL-shaped fields that look interchangeable but absolutely are not. One talks to Moonraker's API. The other just embeds the web UI. Mix them up and nothing works. This guide walks through every field and gotcha I've personally hit across a Voron 2.4, a rooted Creality K1, a Sovol SV08, and a stock MainsailOS Pi 4. If you came from the main OrcaSlicer troubleshooting guide or searched "orcaslicer klipper not connecting," you're in the right place. We'll cover Mainsail, Fluidd, the rooted K1 weirdness, the K1C 2025 firmware that broke port 80, and why MK4 owners keep ending up here even though PrusaLink isn't Klipper. What's in this guide What you need before you start The 90-second version Hostname vs Device UI: the distinction most tutorials miss Step-by-step walkthrough Authentication: trusted_clients vs API key Per-printer port cheat sheet PrusaLink is not Klipper Common errors and quick fixes Known OrcaSlicer bugs to watch for After connection works, what to set up next Frequently asked questions What you need before you start Three things. First, your printer needs Mainsail or Fluidd already running, since OrcaSlicer doesn't install anything on the printer side, it just talks to Moonraker over HTTP. Second, the printer's IP address. Third, OrcaSlicer 2.3.x on your laptop, because the network agent was refactored in 2.3.2 and earlier versions had quirky behaviour with multi-printer Moonraker setups. If you're not sure which version you're on, grab the latest from the download page. OrcaSlicer's network plugin also needs to be enabled. It's on by default in fresh installs, but I've seen it disabled on machines that came from PrusaSlicer profiles. Open Preferences and confirm "Enable network plugin" is ticked. The Preferences dialog is where the network plugin gets toggled. If this is off, no host upload of any kind works, no matter how perfect your Moonraker config is. Find the IP however you prefer. Easiest path on a Klipper printer: open Mainsail and click the Machine tab, the IP shows up under Network. On a stock Creality K1 or KE it's on the touchscreen under Settings, Network, Wi-Fi. On a Voron or DIY Pi build, SSH in and run hostname -I. Set a static DHCP reservation in the router so the IP doesn't drift between reboots, because dynamic leases are the number-one reason "it worked yesterday" prints fail the next morning. One last check: confirm your printer actually runs Klipper plus Moonraker plus a Mainsail or Fluidd web UI. PrusaLink on the MK4 looks superficially similar (it's an HTTP API, it lets you upload G-code), but it's a different protocol with its own host type. We'll cover that below. Bambu printers in LAN mode are also their own thing, see the Bambu LAN mode guide. The 90-second version If you're on a stock MainsailOS or FluiddPi install on a home LAN, here's the fast path. It works for probably 80% of readers, and if it doesn't, the rest of this guide explains why. In OrcaSlicer, click the printer name dropdown at the top-left, then click the small Wi-Fi/globe icon next to it. The Physical Printer dialog opens. Set Host Type to Octo/Klipper. Paste the printer's IP into the Hostname field. Just the IP, no http://, no port. Paste http:// plus the same IP into the Device UI field, e.g. http://192.168.1.45. Leave the API Key field blank. Click Test. Green banner means you're done. Click OK to save. That's the recipe. If Test went green, jump to the after connection works section to set up your G-code flavor and macros. If it didn't, keep reading, because the rest of the article explains every way this can fail and how to recover. Hostname vs Device UI: the distinction most tutorials miss This is what tripped me up on the Voron, and it trips up almost everyone with a non-standard setup. OrcaSlicer's Physical Printer dialog has two URL fields that do completely different jobs. The Hostname, IP or URL field is the address OrcaSlicer uses to talk to the Moonraker HTTP API on port 7125 by default. When you click "Send to printer," Orca POSTs your G-code to http://hostname:7125/server/files/upload. When you click Test, it hits /printer/info on Moonraker. If this field points anywhere that isn't Moonraker, the connection fails. The Device UI field is the URL Orca's Device tab embeds in a webview. That's whatever browser URL loads Mainsail or Fluidd. On stock MainsailOS, that's http://IP on port 80. On a rooted Creality K1, Fluidd lives at http://IP:4408 and Mainsail at http://IP:4409. OrcaSlicer's dropdown only has one entry called "Octo/Klipper," not separate ones for Mainsail and Fluidd, because Moonraker exposes an OctoPrint compatibility shim that lets the same code path handle both ecosystems. GitHub discussion #1169 goes deep on this, but the short version: Hostname goes to Moonraker, Device UI goes to your browser, and on most stock setups they're the same IP with different (or no) port. The four fields that matter. Hostname for Moonraker (port 7125 by default), Device UI for the browser-facing Mainsail or Fluidd page, API key only when you actually need one, and Test to confirm it all works. If you have multiple Klipper instances on the same hostname (Docker, single SBC running multiple Klipper instances on different ports), use IP:port in Hostname but only IP in Device UI, because of issue #2507. Orca routes uploads by hostname, and shared hostnames with only port differing can route uploads to the wrong machine. Step-by-step walkthrough Every click in order. I'm assuming OrcaSlicer 2.3.x is open with a Klipper-flavored printer profile selected from the top-left dropdown. If your printer isn't in the built-in list, pick "Generic Klipper" and customise later. Open the connection dialog Top-left, next to the printer name, click the small Wi-Fi/globe icon. The Physical Printer dialog opens. (If the icon's missing, your network plugin is disabled, head back to Preferences.) Set the Host Type The dropdown lists about a dozen options: PrusaLink, OctoPrint, SimplyPrint, AstroBox, Repetier, Flashforge, Duet, FlashAir, MKS, CrealityPrint, Obico, Octo/Klipper. Pick Octo/Klipper. This one option covers every Mainsail and Fluidd setup since they all speak Moonraker. The Host Type dropdown set to Octo/Klipper. The yellow redaction blocks on the IP and API key are a good habit when sharing screenshots, do the same with yours. Fill in Hostname Paste the printer's IP. Just 192.168.1.45, no http:// prefix. If Moonraker is on a non-default port (Docker, rooted K1), use IP:port. mDNS hostnames like mainsail.local work in theory, but Flatpak builds on Linux can't resolve them due to issue #9311, so I use raw IPs. API Key field Leave it blank on a home LAN with stock MainsailOS or FluiddPi. The default trusted_clients block covers RFC1918 ranges, so your laptop is already trusted. If force_logins is enabled, trusted_clients is locked down, or your laptop is on a different subnet, you need a key, see the authentication section. Device UI The full browser URL of Mainsail or Fluidd. On stock MainsailOS it's http://192.168.1.45 with no port. On a rooted K1 it's http://192.168.1.45:4408 for Fluidd or http://192.168.1.45:4409 for Mainsail. HTTPS / Ignore SSL errors Tick this only if Mainsail is behind a reverse proxy with a self-signed certificate. Most home setups don't need it. Test and save Click Test. A green "connection ok" banner means OrcaSlicer reached /printer/info on Moonraker. Click OK to save. Slice and send After slicing, the bottom-right ac [truncated] --- ## OrcaSlicer for Beginners: Your First Print in 15 Minutes URL: https://orcaslicer.net/orcaslicer-beginners-first-print/ I bought a 3D printer in February. The thing sat in its box for three weeks because every slicer tutorial I clicked on assumed I already knew what a "process profile" was, what a wipe tower did, and why "pressure advance" was supposed to matter on day one. Then someone in a Discord told me to just open OrcaSlicer, pick my printer from the wizard, drop a Benchy on the plate, and hit Slice. Fifteen minutes later I had a finished print. So this is the version of the guide I wish I'd found first. I'm writing this against OrcaSlicer v2.3.2, the current stable build (released March 2025, 13.7k GitHub stars as of May 2026). If you're on a newer release when you read this, the layout might shift a touch but the workflow below has been steady since v2.0. Quick context: why OrcaSlicer instead of Bambu Studio or Cura OrcaSlicer started life as a fork of Bambu Studio, which itself was a fork of PrusaSlicer, which forked from Slic3r. It was built by a developer called SoftFever back in 2023 and is now community-maintained at the OrcaSlicer/OrcaSlicer GitHub repo. The headline reasons people pick it over the alternatives: It's open-source under AGPLv3 and free, with no telemetry tied to any printer brand The built-in calibration suite (temp tower, flow rate, pressure advance, retraction, input shaper) is the most complete of any FDM slicer in 2026 It ships profiles for hundreds of printers including Bambu Lab, Creality, Prusa, Voron, Anycubic, Elegoo, Sovol, Qidi, Flashforge, and Snapmaker It works directly with Klipper, OctoPrint, Bambu Cloud, and Mainsail without a plugin That last point is what hooked me. I've got an Ender 3 V3 KE running Klipper and a Bambu A1 Mini, and I really didn't want two slicers open at the same time. If you've read about OrcaSlicer being "for power users", that's mostly a hangover from the early days. The 2.3.x releases simplified the wizard and the default profiles are now genuinely beginner-safe. Installing it Grab the installer from the official download page (don't trust random forks on third-party sites; the 3MF path-traversal fix in v2.3.2 was a real CVE). Windows users get an installer .exe, macOS users get a .dmg, and Linux users get an AppImage or Flatpak. The whole thing is around 200MB. If you've never installed a slicer before: install it like any normal app. The installer doesn't ask for anything strange, doesn't bundle toolbars, doesn't phone home unless you log in to Bambu's cloud (and that bit is optional). The first-launch setup wizard The first time OrcaSlicer opens, you get a four-screen wizard. It takes about ninety seconds. Region/login. Pick where you live. If you've got a Bambu Lab printer and want cloud uploads, sign in here; otherwise just hit Skip. I always skip and use LAN-only. Printer selection. This is the screen most people get wrong on day one. OrcaSlicer's printer-selection dialog filters by brand and nozzle size. Bambu Lab profiles ship by default, but Creality, Prusa, Voron, Anycubic and Elegoo are all in the same list. Find your actual printer model. Don't pick "Custom Printer" or "FDM Generic" out of impatience, those profiles are nearly empty and nothing will print well. The list filters by brand. Tick the nozzle sizes you own (0.4mm is what your printer shipped with unless you've been modding). Hit Confirm. If your printer isn't listed, OrcaSlicer probably has a clone profile that works. Common examples: most Sovol SV0x printers run fine on the Ender 3 V2 profile, the Kingroon KP3S works on the older Ender 3 profile, and most Klipper-converted printers can use the Voron 0.2 profile as a starting point. Filament selection. Tick the filaments you actually use. PLA almost always; PETG, ABS, TPU only if you have spools at home. You can add more later from the filament dropdown. Network setup. Optional and easy to come back to. If you've got a Klipper or OctoPrint printer, this is where you'd paste its IP, but you can skip it for now and configure from the Device tab later. The interface in 90 seconds The screen splits into five tabs across the top: Prepare, Preview, Device, Project, Calibration. You'll spend 90% of your time on the first two. The Prepare tab is where you spend most of your time. Printer settings up top, filament panel in the middle, process settings under that, and the build plate filling the right two-thirds of the screen. The left side stacks three settings panels: Printer: shows your active printer (a Bambu Lab P1S 0.4 nozzle on a Textured PEI Plate in the screenshot above). Don't fiddle here on day one. Filament: which filament loaded in which slot. The four numbered buttons are AMS slots if you've got a Bambu Lab AMS attached; on a single-extruder printer it just shows slot 1. Process: the print quality preset. The default is "0.20mm Standard @BBL X1C" or whatever your printer shipped with. The five sub-tabs (Quality / Strength / Support / Multimaterial / Others) hide every setting you might ever want. The right side is the build plate, rendered in 3D. Drag with right-click to rotate the view, scroll to zoom, middle-click to pan. The icons across the top of the plate are the manipulation tools: move, rotate, scale, mirror, paint, cut, support paint, seam paint, and so on. The Slice plate / Print plate buttons in the top-right are what you'll hit when you're done. Loading your first model OrcaSlicer ships with a built-in 3DBenchy. Right-click anywhere on the empty plate, pick "Add Handy Model", and there it is. (If you don't know what a Benchy is: it's the standard test boat everyone in the 3D-printing world uses to compare their print quality. Print one, post it on r/3Dprinting, get torn apart by strangers about your stringing. Rite of passage.) You can also drag-drop any .STL, .3MF, or .OBJ file directly onto the plate. Most files you'll download from Printables or MakerWorld are already in 3MF format and land with print settings pre-baked. Arranging and scaling Once you've got something on the plate, the alignment tools matter. Especially if you drag in two or three parts at once. The transform popup shows X/Y/Z scale and exact size in millimetres. The little "Print By Object" reminder in the corner is OrcaSlicer suggesting auto-arrange so parts can't collide mid-print. Click any object and the transform popup appears. The Scale row lets you resize uniformly (tick the Uniform scale box) or per-axis. Size shows the exact final dimensions in millimetres so you don't have to math out percentages. If your parts overlap or sit too close together, OrcaSlicer pops a red error toast warning about collisions. Hit the auto-arrange button (the icon that looks like four squares being aligned) and the slicer figures out a layout that fits. On a Bambu Textured PEI Plate it'll respect the corner exclusion zones too. Two beginner gotchas here: Don't trust auto-arrange to optimise for surface quality. It optimises for fitting, not orientation. If a part has a clear "good face" you want pointing up, rotate it manually first. The Print By Object toggle changes the print order. It's powerful for multi-part plates but it can also make the print head slam into earlier finished parts. Leave it off until you've read the warnings. A safe default settings cheatsheet You can spend a year tuning these. For your first print, just match these and tweak later: SettingBeginner default Layer height0.20mm First layer height0.24mm Nozzle temp (PLA)210°C (first layer 215°C) Bed temp60°C (first layer 65°C) Outer wall speed50–80 mm/s on most printers Infill speed100–200 mm/s First-layer speed20–30 mm/s on a budget printer, 40–50 mm/s on a high-speed printer Cooling fan100% after layer 3 Infill15% gyroid SupportsOff (turn on later if your model has overhangs steeper than 45°) This isn't optimised for any specific filament brand. It's the "you will get a usable print" default. After your first successful Benchy, the next thing to do is run OrcaSlicer's flow rate calibration and pressure advance test, both of which [truncated] --- # Section: Calibration ## How Often Should You Recalibrate OrcaSlicer? 2026 Maintenance Schedule URL: https://orcaslicer.net/orcaslicer-recalibration-schedule/ I spent most of a Saturday last spring running every calibration test in OrcaSlicer on a fresh roll of generic black PLA. Temp tower, two-pass flow rate, pressure advance line, pressure advance pattern, retraction, max volumetric speed, the works. Six hours of printer time, about 80 grams of filament, and a perfectly tuned profile for a $19 spool I burned through in three weeks. The print quality was good. It was also good before I started, because I'd already calibrated the previous spool of the same brand and color two months earlier. That was the moment I admitted I had a problem. Not a printer problem. A FOMO problem. Every "you must calibrate every new spool" YouTube video had wormed its way into my workflow until I treated calibration like brushing my teeth, something you do on schedule because you're supposed to, regardless of whether anything had actually changed. The math is brutal once you sit down with it. If you swap filament weekly and "calibrate everything" takes six hours, that's a part-time job for a hobby that's supposed to relax you. So I rebuilt my approach from scratch, and the rule I landed on is the one I want to make the case for in this article: recalibration is trigger-driven, not calendar-driven. The right question isn't "how often should I recalibrate?" It's "what actually changed since my last good print?" Because what changed determines which test you need, and most of the time, the answer is "nothing changed, just print." Table of contents The two myths to kill first The mental model: what causes calibration to drift The master maintenance schedule When you DO need to recalibrate (11 triggers) When you DON'T need to recalibrate Time investment per calibration Per-printer differences (Bambu, Klipper, Marlin) Profile management is the real maintenance discipline A decision tree before every print FAQ The two myths to kill first Before any schedule makes sense, two pieces of bad advice need to go in the bin. Myth 1: You must recalibrate everything for every new spool. This one comes from a generous place. Creators want to be thorough, and "do all the tests" is the safest possible recommendation. It's also wildly overkill for the average home printer. A new spool of the exact same brand, exact same color, exact same material, opened the same week from the same vendor box, will have flow and PA values within about 1 to 2 percent of the spool you finished yesterday. You won't see the difference on a benchy. You will see the difference on your wallet and your weekend. Myth 2: Bambu handles all of this automatically, so you don't have to do anything. Also wrong, in a more interesting way. Bambu's auto features are real and they're great. The X1C, X1E, and H2D run Flow Dynamics Calibration (Bambu's PA equivalent) with LiDAR, and every Bambu printer runs vibration calibration (auto input shaper) at print start. That's two genuinely automated calibrations per print. But flow rate is still manual on every Bambu printer in 2026, including the X1C, and temperature tower is still manual on every Bambu printer ever shipped. The community thread at P1S Flow Calibration on the Bambu Lab forum is full of P1S owners discovering this the hard way after months of assuming the printer was tuning itself. If you take only one thing from this section: automatic on Bambu means K-value and shaper. It does not mean flow rate, temp, retraction, or MVS. Those are still on you, and the rules in the rest of this article apply to your Bambu just like they apply to a Voron. The mental model: what actually causes calibration to drift Once you've killed the calendar-driven approach, the next thing to install is a framework for thinking about what can drift and why. There are exactly four categories of change that move calibration values, and every trigger in the rest of this article falls into one of them. 1. Filament path changes. Anything that alters what's flowing through the hotend. New brand, new material, new color (in some cases), new moisture state. These changes hit flow rate and pressure advance first, temperature tower second. The reason is straightforward: PA and flow are direct functions of the melt's viscosity and elasticity, and viscosity is a function of the resin formulation. Two PLAs from two manufacturers can have measurably different additive packages (heat stabilizers, slip agents, pigment loadings) that shift behavior at the nozzle. 2. Hardware changes. Anything mechanical: nozzle (material or diameter), hotend assembly, heatbreak, extruder gears, belts, leadscrews. These changes hit the mechanical calibrations: input shaper for belts, e-steps for extruders, MVS and PA for nozzles. A new nozzle is the most invasive single change you can make. It can require almost a full recalibration sweep. 3. Firmware changes. Klipper config edits, Bambu firmware bumps, OrcaSlicer version updates. Klipper config changes that touch rotation distances or accel limits invalidate input shaper. Bambu firmware updates have, in some past releases, shifted the K-value behavior. OrcaSlicer version updates rarely require recalibration on their own, but if Orca flags a profile incompatibility on import, a quick visual check on a 20mm cube is prudent before trusting the saved values. 4. Environmental changes. Seasonal humidity, ambient temperature, enclosure use. These hit drying state primarily, and PA secondarily on hygroscopic materials. A spool that prints beautifully in February (35% indoor humidity) can string and puff in August (70%) without you changing a single setting in OrcaSlicer. If none of those four categories of change has happened, your calibration is still valid. Print the thing. The single most important calibration discipline isn't running more tests, it's recognizing when zero tests are the correct answer. For the full lineup of what each test does and how to run it, I keep coming back to the consolidated OrcaSlicer calibration guide rather than chasing individual tutorials. That's the hub. Everything below is the schedule for using it. OrcaSlicer's built-in calibration menu lists every test you might run. Knowing which ones to skip is the actual skill. The master maintenance schedule This is the table I wish I'd had two years ago. Rows are triggers (the thing that changed), columns are calibrations. Cell values are Yes (run it), No (skip it), or Optional (run only if you see the symptom). Read across a row to see what your specific trigger demands, then drop down to the trigger-by-trigger walkthrough below for the why. Trigger Temp Tower Flow Rate Pressure Advance MVS Retraction Input Shaper Bed Mesh E-steps New filament brandYesYesYesOptionalOptionalNoNoNo Same brand, new color (plain)NoOptionalNoNoNoNoNoNo Same brand, new color (filled, silk, glow)NoYesOptionalNoNoNoNoNo Same spool after long storageNoYes (after drying)NoNoNoNoNoNo Nozzle material change (brass to hardened or back)YesYesYesNoYesNoNoNo Nozzle diameter changeYesYesYesYes (first)YesNoNoNo Hotend replacementYesYesYesNoNoNoNoNo Firmware update (Klipper config-changing)NoNoOptionalNoNoYesNoNo Firmware update (Bambu major)NoNoOptional (spot-check)NoNoNo (auto)NoNo Firmware update (OrcaSlicer only)NoNoNo (visual check)NoNoNoNoNo Belt re-tension or mechanicalNoNoNoNoNoYesNoNo Extruder gear swap (BMG and similar)NoYesYesNoNoNoNoYes (first) Seasonal humidity shiftNoNoOptional spot-checkNoNoNoNoNo Bed disturbance or ~50 printsNoNoNoNoNoNoYesNo Bed surface change (PEI to smooth PEI)NoNoNoNoNoNoYesNo Nothing changedNoNoNoNoNoNoNoNo How to read "Optional": run that test only if you visually see the symptom. Corner blobbing or sharp-corner bulge means run PA. Visible over-extrusion (z-seam blobs, top-layer pillowing) or under-extrusion (gaps, weak walls) means run flow rate. Otherwise, skip and save the filament. The bottom row is the most important one in the whole table, and it's the row most users refuse to act on. If nothing has changed since your last good print, your calibration [truncated] --- ## OrcaSlicer PA Line vs PA Pattern vs PA Tower: Which Pressure Advance Test (2026) URL: https://orcaslicer.net/orcaslicer-pa-line-vs-pa-pattern/ I lost an entire Saturday to pressure advance once. I had a brand new spool of matte PLA, a freshly tuned Voron, and a Benchy that kept printing with corner bulges so fat you could see the seam from across the room. So I opened the OrcaSlicer Calibration menu and stared at three options that all claimed to fix the same problem: PA Line, PA Pattern, PA Tower. Three tests. One bug. I picked PA Line first because it was at the top of the list, ran it, eyeballed the result, saved a value, reprinted the Benchy, and got the exact same bulges. Different number, same bulges. It turned out I'd been reading the test wrong, running it on a printer whose firmware couldn't act on the value, and then forgetting that OrcaSlicer doesn't actually save your calibration result back into the filament profile. Nothing about that experience was obvious from the menu. So I'm writing the article I wish I'd had: the honest comparison of OrcaSlicer's three pressure-advance tests, which one to use for which printer, how to read each one without falling into the "pointiest corner" trap, and what to do when the test refuses to slice. I'll also clear up the cooling-tower question, because that's a search query I see a lot. Short version: OrcaSlicer doesn't ship one. The longer version comes later. For now, if you came here to choose between PA Line, PA Pattern, and PA Tower, keep reading. Table of contents TL;DR decision matrix The cooling-tower question: does OrcaSlicer have one? What PA Pattern actually is PA Line vs PA Pattern vs PA Tower How to read each test (it is not the pointy corner) Saving the value manually Ten common mistakes that ruin a PA calibration Material cooling cheat sheet When PA Pattern doesn't work FAQ TL;DR decision matrix If you skim only one section, make it this one. The three tests are not interchangeable, and picking the wrong one wastes filament. I've boiled down the upstream wiki, the Ellis Print Tuning Guide, and a stack of GitHub issues into a single table. TestBest forAvoid ifFirmwareTime to print PA LineQuick first-pass scan, any printerYour first layer is unreliableKlipper, Marlin (with Linear Advance), Bambu, RepRapUnder 5 min PA PatternFine-grained, most accurate readingMarlin without LA compiled in, nozzle larger than 0.4 mm, layer height above 0.24 mmKlipper (validated). Marlin and Bambu are experimental per the Adaptive PA wiki.Under 10 min PA TowerFirst layer is shaky, you trust calipers more than your eyesYou want sub-0.005 PA resolutionKlipper, Marlin Linear Advance15 to 25 min The quick decision tree: if you're on Klipper with input shaper already calibrated, run PA Pattern. If you're on a stock Bambu X1C or P1S, run PA Line and skip the Flow Calibration toggle. If your first layer is shaky and you don't want to fight it before every test, run PA Tower. If you're on Marlin and you haven't compiled Linear Advance into your firmware, none of these tests will do anything because the printer can't act on the value. I'll come back to that one. If you only have time to do one calibration pass today and you're not sure where to start, do PA Line. It runs in under five minutes, it works on every firmware Orca supports, and it'll get you 80 percent of the way to a clean corner. PA Pattern is the precision pass you do after you've nailed flow ratio and run a temp tower for your filament. The cooling-tower question: does OrcaSlicer have one? I'm putting this here because the question gets asked enough that ignoring it would be dishonest. OrcaSlicer does not ship a cooling tower or fan-speed-tower calibration test. The Calibration menu has nine entries and none of them is a part-cooling test. The full list is Temperature, Volumetric Speed, Pressure Advance, Flow Ratio, Retraction, Tolerance, Cornering, Input Shaping, and VFA. The official wiki confirms this on the Calibration page. There's an open feature request on GitHub asking for one. Discussion #2906, "possibility to do fan speed tower test," is the canonical thread. As of writing, it's still a community request rather than a shipped feature. Issue #919 covered the same ground back in May 2023 and went stale. So if you searched for "cooling tower" and landed here expecting a one-click test, that's why you couldn't find it. If you actually want to dial in part-cooling, the community workflow is to print a fan-speed STL like the caste1200 Ultimate Fan Speed Test from Printables and step M106 S<value> commands in the printer's layer-change G-code at fixed Z bands. It's a manual workflow, not a slicer feature. For most use cases, tuning the filament profile's overhang_fan_speed, overhang_fan_threshold, full_fan_speed_layer, and close_fan_the_first_x_layers keys gets you most of the way without printing a tower at all. Now, the reason this matters in a pressure advance article is that the symptoms overlap. Bulged corners can be a PA problem or a cooling problem. Droopy overhangs are a cooling problem. Bulged corners on every wall, regardless of whether they're overhanging, are a PA problem. If you fire up PA Line and every single line looks rough, the issue is probably upstream of pressure advance. Run a temp tower first, check your flow rate, and then come back to pressure advance. What PA Pattern actually is Menu path: Calibration, then Pressure Advance, then Pattern Method. That's three clicks. What you get is a hand-written 2D pattern test adapted from Andrew Ellis' pattern method generator, which was itself derived from the Marlin pattern method developed by Sineos. OrcaSlicer's Pressure Advance calibration wiki documents the lineage, and Obico's comprehensive guide says the same thing in plain English. Mechanically, here's what the slicer is doing. It generates a thin rectangular prism that acts as a frame. Inside that frame, it writes a custom G-code pattern of straight lines with sharp acceleration and deceleration corners. Each row of the pattern is printed at a different pressure advance value, and the value is labeled in extrusion next to the row. The "infill" is not real slicer infill. It's hand-written G-code embedded into the layer instructions of the prism. That detail matters because it explains why the test breaks if you change layer height or nozzle size: the G-code generator wasn't designed for those configurations. The K-factor calibration concept that PA Pattern inherited from Marlin and adapted. Each row tests a different pressure advance value at the slow-fast-slow transition. When you open the Pattern Method dialog, you'll see a configuration window with start PA, end PA, step PA, and a couple of presets. Bowden defaults use a wider range, 0.02 to 1.0, because Bowden setups need a much higher PA value to compensate for the PTFE compliance. Direct drive defaults sit in the 0.0 to 0.1 range with a step of 0.002. Pick the preset that matches your hardware. If you're on a Voron Mini Stealthburner, that's direct drive. If you're on an Ender 3 with the stock Bowden setup, that's Bowden. One configuration trap worth flagging now: if you're on a Bambu X1C or P1S, do not tick the Flow Calibration box during PA Pattern. Bambu firmware handles flow calibration separately and the two tests interact in ways that make the pattern unreadable. I've seen this trip up a lot of new Bambu owners. Run flow rate as its own test, in its own calibration pass. Total print time on PA Pattern is under ten minutes for most configurations. It's the test I run when I want a precise reading and I have a working first layer. Speaking of which. PA Line vs PA Pattern vs PA Tower Now we get into the real comparison. I'll go test by test, then summarize. PA Line PA Line is the fastest and simplest of the three. OrcaSlicer prints a series of straight extrusion lines on the bed, each at a slightly different pressure advance value, with the value labeled next to the line. The whole test is one layer tall. It runs in under five minutes and uses almost no filament. You read it by looking at where each line [truncated] --- ## OrcaSlicer VFA Test: Diagnose Vertical Fine Artifacts (2026 Guide) URL: https://orcaslicer.net/orcaslicer-vfa-test/ The first time I noticed VFAs on a print, I thought my nozzle was clogging. I was running a tall vase in PETG on a Bambu P1S, raking light hitting it sideways on the workbench, and the whole front face looked like someone had brushed fine corduroy onto the plastic. Tiny parallel stripes, perfectly regular, marching up the wall every two millimetres or so. The dimensional accuracy was fine. The corners were sharp. But that surface ruined what should have been a showpiece print. I spent the next three evenings chasing it. I dropped the outer wall speed. I bumped it back up. I re-ran the input shaper auto-cal. I rechecked the belt with the guitar-tuner app I keep on my phone. Nothing on the slicer side moved the needle, and I started to suspect I was looking at a hardware problem masquerading as a slicer one. That's when I finally sat down and ran the OrcaSlicer VFA test the way it's meant to be run, with a calibration order behind it and a plan for reading the tower. This guide is what I wish I'd had on the bench that week. I'll walk you through what VFA actually means inside OrcaSlicer 2.3.x, how the test prints and what each band on the tower really represents, a decision tree for telling ringing apart from belt-meshing apart from salmon skin, and the honest list of things the slicer can't fix no matter how clever you get with profile keys. If you'd rather start from the beginning, my OrcaSlicer calibration guide covers the whole pipeline. The VFA test is the last stop on that road. What's in this guide What VFA stands for in OrcaSlicer What the VFA test actually prints The diagnosis decision tree Mechanical causes vs slicer causes Where VFA fits in the calibration order Speeds, accelerations, and the dialog fields How to read the printed tower Feeding results into Resonance Avoidance Per-printer notes (Bambu, Klipper, Marlin) Common mistakes that waste a tower When the VFA test can't fix it FAQ What VFA stands for in OrcaSlicer VFA is short for Vertical Fine Artifacts. The phrase covers a category of surface defect, not a single one, and that distinction matters more than anything else in this article. The Wevolver technical overview and the OrcaSlicer wiki both describe VFAs as microscopic ridges on vertical surfaces, typically ten to forty microns wide, appearing at regular intervals on smooth walls. On GT2 belt systems the spacing often lands right around two millimetres, which happens to match one tooth of belt travel through the pulley. So when somebody says "I've got VFAs," they could mean ringing. They could mean belt-meshing banding. They could mean salmon skin. They could mean Z-banding from a bent leadscrew, or a layer shift, or even a chronic seam that they didn't realise was a seam. The OrcaSlicer VFA test is built to separate one of those (speed-dependent resonance) from the rest of them. The other subtypes show up on the tower too, but you fix them in different places. Inside OrcaSlicer 2.3.x the test lives at Calibration → More → VFA in the top menu bar. The same path is confirmed on the OrcaSlicer wiki page for VFA, on the Obico walkthrough, and on the orcaslicer.com mirror of the wiki. In 2.3.x the docs sidebar happens to nest VFA under Input Shaping, but the in-app entry point hasn't moved. You get a dialog with three fields (Start speed, End speed, Step) and a Generate button. That's it. The slicer builds the tower model on the fly based on the range you typed in. What the VFA test actually prints The model OrcaSlicer generates is a tall, hollow tower with thin vertical walls and a stack of hexagonal cutouts running up the front and back faces. The hexagons aren't decorative. They give you a flat reference patch between each cutout where the outer wall is unbroken, which is exactly where VFAs show up if they're going to. Each vertical "band" of the tower (the strip from one row of hexagons to the next) is printed at a different outer-wall speed. The generated VFA tower in OrcaSlicer's Preview pane, coloured by volumetric flow. Each row of hexagons sits between a band of unbroken vertical wall, and that wall is what your eyes (and a raking light) are going to read. The speed assignment is mechanical. If you typed Start 160, End 500, Step 20, the lowest band prints at 160 mm/s, the next at 180, then 200, and so on up to 500. The Obico guide gives a tidy formula: Speed at notch N = Start + (Step × (N − 1)). You can verify that by hovering over a band in the Preview pane and checking the speed colour-map at top-right. Same tower, but coloured by speed instead of flow. The vertical stripes in the colour map line up with each speed band, which makes it easy to confirm that the Start/End/Step values you entered actually produced the spread you wanted. One thing the wiki page is firm about and that I'd repeat twice: there is no shipped default speed range. The temp tower has a 5 °C step baked in. The VFA dialog does not. You have to type your own Start, End, and Step values based on the printer you're calibrating, and if you skip that thought you'll either run a tower that's too coarse to be useful or one that bumps into Max Volumetric Speed halfway up and stops being meaningful. I'll come back to MVS in the speeds section because it's the single most common reason a VFA test fails silently. The diagnosis decision tree This is the part I find people skipping, which is funny because it's the part that saves the most filament. Before you generate a tower, look at the print that's bothering you and decide what kind of artifact you actually have. Different artifacts come from different parts of the printer, and only one of them is what the VFA test fixes. Here's the mental model I use on the bench: Echo or wavy lines that fade after a corner. That's classic ringing (sometimes called ghosting). The wavelength tells you the resonant frequency of the axis. It fades because the oscillation damps out as the head moves in a straight line. Fix domain: input shaper, acceleration, and belt tension. Even, repeating fine vertical lines spaced about 2 mm across the whole wall, not fading. That's belt-tooth meshing, sometimes called belt transmission error. Two millimetres is one tooth of GT2 belt pitch travelling through the pulley. It will not go away in software. Fix domain: belt quality, idler tension, pulley grub screws, and in some cases a quality-control problem with the belt itself. A diagonal sheen on flat walls, especially right after a seam or short move. That's salmon skin, driven by pressure advance miscalibration on Klipper and Bambu firmware, or by Linear Advance on Marlin. Fix domain: re-run the pressure advance calibration. Horizontal bands at a constant Z spacing. If the spacing matches one full leadscrew rotation (often 2 mm or 4 mm), it's Z-banding. Bent leadscrew, binding Z, or POM nut wear. Fix domain: hardware. A sudden horizontal step in the wall. That's a layer shift, not a VFA. Belt tension, stepper current, or a decoupled pulley. A vertical line down one side of every part. That's a seam, and it's a slicer issue. Aligned, Random, or Painted seam settings. A rough zone that only shows up above a certain speed. Now we're talking. That's speed-dependent resonance, and it's exactly what the VFA test surfaces. Rough on one axis, smooth on the perpendicular face. Axis-specific resonance. Input shaper per axis, or belt tension on the offending axis. The rule of thumb I keep in my head: ringing fades, true VFAs don't. If you can run your fingernail down the wall and the texture is exactly the same one centimetre after a corner as it is ten centimetres after a corner, you're not looking at ringing. You're looking at a steady-state artifact, and changing your acceleration won't help. Mechanical causes vs slicer causes Once you've decided what kind of artifact you have, the next question is whether you can fix it in OrcaSlicer at all. I've split this into two honest lists. The first is everything that's actually [truncated] --- ## OrcaSlicer Tolerance & Fit Test: Functional Parts (2026 Guide) URL: https://orcaslicer.net/orcaslicer-tolerance-fit-test/ I designed a snap-fit case for an Arduino Mega last month, slid the lid on, and it bound up so tight I had to break it back off with pliers. The CAD was perfect, the printer was calibrated, but the OrcaSlicer X-Y hole compensation was still at the default 0 mm. One tolerance test print and a 0.1 mm offset later, the lid clicked together like it should. This is the calibration most people skip until they need it, and then they're sitting at the workbench wondering why a 10 mm hole that measures 9.85 mm on the calipers won't pass an M3 bolt. If you print functional parts, especially anything with mating geometry, you cannot get away without running a fit test at least once per filament. Not "should". Cannot. The whole reason the test exists is that no two printers produce the same hole at the same dimension, and no two filament rolls shrink the same way. The Bambu wiki, the Obico guide, and most YouTube walkthroughs gloss over what to do with the result. I'm going to be specific. This guide walks through the OrcaSlicer Tolerance Test from menu click to bolted-down enclosure. It covers the hex test model itself, the math behind XY hole compensation, per-material shrinkage, the three fit types you'll actually use in CAD, and the upstream calibration issues that quietly poison your tolerance result before you ever start. If you haven't done the rest of the calibration stack first, head over to the OrcaSlicer calibration guide and come back when flow and pressure advance are dialed. Table of Contents Why functional parts need a fit test at all The Orca tolerance test model, what it actually is Running the tolerance test in OrcaSlicer, step by step Reading your result without spiraling The three fit types and the gap each one needs XY Hole vs XY Contour Compensation, explained Applying the compensation value in the slicer CAD-baked tolerance vs slicer compensation Per-material shrinkage and what changes Upstream causes that wreck the test Real forum pain points, real fixes Wrap, and what to calibrate next Why functional parts need a fit test at all Decorative prints forgive you. A vase or a low-poly fox doesn't care if its silhouette is 0.15 mm wider than the CAD. Functional parts are different. A snap-fit lid that's 0.2 mm too tight cracks at the corner. A printed gear with a 5.05 mm bore won't slide onto a 5 mm shaft. An M3 clearance hole that prints at 3.1 mm instead of 3.4 mm bottoms out the bolt head before the threads even engage. Tolerance isn't an aesthetic concern. It decides whether the part works. Here's what's actually happening. When the nozzle traces a closed hole, the molten plastic curls slightly inward as it cools. The hole shrinks. On the outer contour of a part, the same physics applies, except now the curl pulls the outer dimension inward too. So a 10 mm peg ends up at 9.95 mm and a 10 mm hole ends up at 9.85 mm, and your beautifully-matched CAD doesn't assemble. It's not the printer being broken. It's basic thermal contraction plus extrusion path geometry, and every FDM printer has it to some degree. The tolerance test gives you a single number that tells you how much your printer overshoots inward on hexagonal recesses. From that number, you decide whether to fix it in CAD or in the slicer, and which fit type each feature of your design needs. Without the test, you're guessing, and guessing on functional parts wastes filament and bench time. I've burned through entire afternoons reprinting a single bracket because I didn't bother to run a 20-minute calibration. The Orca tolerance test model, what it actually is OrcaSlicer ships with a built-in tolerance test model. It's a flat base with six hexagonal recesses, each one sized at a different positive offset from a printed hex tester. The offsets are 0.00 mm, 0.05 mm, 0.10 mm, 0.20 mm, 0.30 mm, and 0.40 mm. The tester is a printed hexagon shipped with the model, and a standard 6 mm Allen key fits the same socket if you'd rather use that. The choice of hex over a round peg-and-hole isn't aesthetic. Round holes hide error. A 10 mm circular hole that prints as 9.85 mm with a slightly oval cross-section can still pass a peg test if the peg is also slightly oval in the same orientation. The two errors cancel. Hexagons don't let you cheat that way. A hex hole has six straight wall segments and six corners, and the corners are where extrusion is hardest to control. If the corner radius blunts inward, the tester won't drop in regardless of how the rest of the hole looks. It also matches real CAD design intent. Nut traps for M3 nylock nuts, hex sockets for cap screws, knurled brass inserts, snap-fit pawls. They all have flat reference faces. Testing a hex shape gives you data that directly applies to the parts you actually print. If your printer can pass a 0.10 mm hex tolerance, your M3 nut traps will work at +0.10 mm clearance too. The Orca tolerance test loaded on the build plate. Six hex recesses on the base, one hex tester to the side. Plate orientation matters less than you'd think because the test is symmetric, but keep the tester flat to the bed. Running the tolerance test in OrcaSlicer, step by step There are two ways to load the model. Both end up in the same place. Route A, calibration menu. Open OrcaSlicer, click Calibration in the top menu, then Tolerance Test. The model loads as a fresh project on the active build plate, with the tester already positioned alongside the base. This is the easier of the two routes. Route B, plate context menu. In Prepare mode, right-click anywhere on the build plate and choose Add Handy Model. Pick the tolerance test from the dropdown. This is useful if you want to add the test to an existing project rather than starting fresh, or if you want to drop in two copies for comparison. Before you slice, do these things in order. Skip any of them and you'll get a misleading result. Use the printer profile and filament you're actually testing. The tolerance result is filament-specific. PLA at 215 C with brand A gives you a different number than PLA at 200 C with brand B. Don't run the test in your "generic PLA" profile and apply the result to Polymaker PolyTerra. It won't transfer cleanly. Zero out both compensation values. Go to Process settings, then Quality, then Precision. Set X-Y Hole Compensation to 0 mm and X-Y Contour Compensation to 0 mm. The Obico guide flags this explicitly, and it's the single most common mistake. If either value is non-zero when you run the test, you're measuring your settings, not your printer. Standard layer height. Use 0.20 mm on a 0.4 mm nozzle. Don't try to "be safe" with 0.08 mm. Finer layers introduce their own variables (pressure advance behaves differently, line width effects shift). Match what you'll actually print functional parts at. Default wall count. Two outer plus one inner is the standard Orca default. Don't crank to four walls hoping for accuracy. The compensation behavior depends on wall count near small features, and you want the test to reflect your normal slicer config. Leave Precise Wall on. The default in 2.3.x is true, and that's what you want. Precise Wall adjusts outer wall spacing to improve dimensional consistency. Turning it off here would defeat half the point of testing. Confirm flow and pressure advance are tuned. If either is off, the tolerance test result is meaningless. Run flow rate calibration and pressure advance first. I cannot stress this enough. I've seen tolerance threads on GitHub where the actual fix was flow ratio off by 4%. Slice, print, and wait for the part to cool fully before testing. PLA needs about 30 minutes after the print finishes to stop dimensional creep, PETG closer to 60 minutes, ABS 90 minutes or more depending on whether you've got an enclosure. Yank a hot PETG print off the bed and try to fit the tester immediately, and you'll get readings that change overnight. Don't trust hot measurements. Reading your result without spiraling Once the print is cool, take the tester [truncated] --- ## OrcaSlicer Cornering & Junction Deviation Calibration (2026) URL: https://orcaslicer.net/orcaslicer-cornering-calibration/ Input shaper is what everyone talks about for high-speed printing, but the calibration nobody mentions is junction deviation. My Voron's corners looked sharp at 100 mm/s and rounded into smooth curves at 250 mm/s, and tuning input shaper didn't fix it. The fix was a single JD value I'd never touched, and OrcaSlicer has a calibration test for it that takes thirty minutes start to finish. I'm writing this because the cornering test is the most ignored entry in OrcaSlicer's Calibration menu. It sits one row below Input Shaper, it ships with sensible defaults, and it solves a specific problem that input shaper alone cannot solve. If you're printing at 200 mm/s plus on a Marlin or Klipper machine and your corners look like soft chamfers instead of sharp 90 degree angles, this is the calibration you've been missing. I'll walk through the three firmware models, the OrcaSlicer test itself, the Klipper gap that nobody documents, and the input shaper interaction that the Klipper docs warn about but most tuning videos skip. Table of contents Who actually needs this calibration The three cornering models in 3D printer firmware OrcaSlicer's Cornering test, end to end Reading the printed tower Marlin integration and saving the value The Klipper SCV gap and how to work around it Input Shaper and SCV, the warning nobody reads Saving the value in your OrcaSlicer printer profile Common mistakes and edge cases FAQ Who actually needs this calibration I'll save you reading time. Not everyone benefits from this test, and the existing OrcaSlicer docs are too polite to say so. Here's the honest breakdown. You should run it if you own a Marlin 2.x machine running at 100 mm/s outer wall or faster, a Klipper-based Voron or Voron-clone, a Ratrig, a Switchwire, a heavily modded Ender or CR series, or any DIY CoreXY where you've personally configured the firmware. These are the platforms where the firmware exposes Junction Deviation or square_corner_velocity as a tunable, and where the defaults are conservative enough that you'll see real improvement. You can skip it if you own a Bambu Lab printer. The X1C, P1S, A1, and H2D handle cornering internally inside Bambu's motion planner. The Junction Deviation field in OrcaSlicer's Bambu printer profile is a time estimation hint that was added in 2.3.2; it doesn't get passed to the firmware as a runtime parameter. Bambu owners running the cornering test will see identical bands on the tower, which is correct behavior for that firmware, not a bug. Don't waste filament. You probably won't see much from it if you own a Prusa MK4, MK4S, or Core One. Prusa's Buddy firmware uses a Marlin-derived planner with input shaping baked in and JD defaults that are already aggressive. You can technically run the test, but most owners report no measurable corner improvement. If you've already tuned input shaper through PrusaSlicer or the printer's built-in routine, leave JD alone. For everyone else, especially the Marlin-Voron and Klipper-on-Voron crowd, this is one of those calibrations where the print time stays the same but the print quality jumps a visible step. I keep it in my standard calibration sequence right after input shaper. The three cornering models in 3D printer firmware Before we open the slicer, a quick physics detour. A 3D printer can't take a corner at full speed because the change in direction at a sharp corner is effectively infinite acceleration. The motion planner has to slow down for the corner. The question is by how much, and that's what these settings control. Three rules exist across the firmwares OrcaSlicer supports. Firmware Model Parameter Default Unit Marlin (legacy) Classic Jerk Jerk X/Y 8.0 (typical) mm/s Marlin 2.0+ Junction Deviation JUNCTION_DEVIATION_MM 0.013 mm Klipper Square Corner Velocity square_corner_velocity 5.0 mm/s RepRap/Duet Instantaneous speed change M566 varies mm/s (per axis) Classic Jerk, the legacy model Classic Jerk is the oldest model. It defines a fixed instantaneous velocity change in mm/s that the firmware will tolerate at any corner. Set it to 8 mm/s and the printer hits any corner at exactly 8 mm/s, regardless of the corner's angle. It's crude. A 90 degree corner gets the same treatment as a 170 degree near-straight, which means either your sharp corners are overshooting or your nearly-straight corners are bottlenecking. Marlin still ships it as CLASSIC_JERK for users who want it, but it stopped being the default after Marlin PR #15481 in October 2019. Marlin's Junction Deviation Junction Deviation is a single value in millimeters that represents how far the toolhead is allowed to deviate from the geometric path while taking the corner. The firmware fits a virtual circle that touches both segments at the junction; the radius of that circle combined with your acceleration limit gives the max cornering speed. Larger JD means a larger virtual circle, higher cornering speed, and more visible rounding on real corners. Smaller JD gives a tighter circle, sharper corners, and slower cornering. Marlin's default is 0.013 mm, set in Configuration_adv.h. That default is deliberately conservative. Most tuned Marlin builds end up between 0.02 and 0.08 mm depending on frame and accelerations. Klipper's square_corner_velocity Klipper used to expose Junction Deviation directly, then replaced it with square_corner_velocity because the units are more intuitive. SCV is literally the speed in mm/s that the toolhead is allowed to take a 90 degree corner at. Doubling SCV doubles cornering speed. From the Klipper docs themselves: "the maximum velocity (in mm/s) that the toolhead may travel a 90 degree corner at. A non-zero value can reduce changes in extruder flow rates by enabling instantaneous velocity changes of the toolhead during cornering." For other angles Klipper derives the speed from a centripetal-acceleration model, so sharper corners get slower and gentler corners get faster. How the units relate The two modern models, JD and SCV, are equivalent in what they control. They just use different units. A common rough conversion used by tuners is: SCV ≈ sqrt(2 * JD * acceleration) At 2000 mm/s² acceleration a JD of 0.02 mm corresponds to SCV roughly 8.9 mm/s. At 5000 mm/s² a JD of 0.02 gives SCV around 14 mm/s. This is an approximation, not an exact equivalence across all corner geometries; the Klipper Kinematics doc has the precise derivation. But it's close enough to translate test results between firmwares. Junction Deviation (mm) At accel 2000 mm/s² At accel 5000 mm/s² At accel 10000 mm/s² 0.005SCV ≈ 4.5 mm/sSCV ≈ 7.1 mm/sSCV ≈ 10.0 mm/s 0.013 (Marlin default)SCV ≈ 7.2 mm/sSCV ≈ 11.4 mm/sSCV ≈ 16.1 mm/s 0.02SCV ≈ 8.9 mm/sSCV ≈ 14.1 mm/sSCV ≈ 20.0 mm/s 0.04SCV ≈ 12.6 mm/sSCV ≈ 20.0 mm/sSCV ≈ 28.3 mm/s 0.08SCV ≈ 17.9 mm/sSCV ≈ 28.3 mm/sSCV ≈ 40.0 mm/s You'll notice the same JD value produces a much higher SCV as acceleration climbs. That's not magic. Cornering speed at a given junction depends on both the deviation tolerance and the available acceleration to recover. If you're on a Voron pushing 10000 mm/s² with input shaper tuned, a JD of 0.04 corresponds to roughly 28 mm/s SCV, which is well above Klipper's 5 mm/s default. That's where the input shaper warning becomes important; we'll get to that. OrcaSlicer's Cornering test, end to end This is the actual procedure. OrcaSlicer added a unified Cornering calibration in the 2.3 release line that auto-detects your firmware and sweeps the right parameter. I want to walk through what it does and the prerequisites it enforces, because skipping any of them invalidates the result. OrcaSlicer's release notes from the 2.3 line announcing the Junction Deviation Test for Marlin2. The Calibration menu now has an Advanced submenu with Cornering alongside Input Shaper and VFA. Menu path Open OrcaSlicer with your printer profile loaded. The menu path is Calibration → Advanced Calibrations → Cornering. The in-app label is "Cornering (Jerk & Junction Deviation)" in the [truncated] --- ## OrcaSlicer Max Volumetric Speed Test: Find Your Hot End Limit (2026) URL: https://orcaslicer.net/orcaslicer-max-volumetric-speed/ I bought a Bondtech CHT nozzle expecting to print PLA twice as fast and ate three failed Benchies before I figured out the problem wasn't the nozzle. It was that my filament profile still had max volumetric speed set to the stock 14 mm³/s. The CHT could do 22 mm³/s easily, but the slicer was throttling me down to the V6 ceiling. One number changed and the prints came out clean. That's the trap with max volumetric speed (MVS). It's a silent governor. You can crank print speed to 300 mm/s in your profile and OrcaSlicer will quietly cap the toolhead at whatever the filament's MVS allows, never warning you, never highlighting anything red. The only clue is a print time estimate that doesn't match the speed you typed, and prints that come out the same regardless of how aggressive your speed settings get. I've watched people buy faster printers and high-flow hot ends chasing speed gains they already had on paper, blocked by a single field buried in filament advanced settings. This guide walks through OrcaSlicer's built-in MVS calibration test end to end. I'll show you exactly what the test prints, how to read the failure point, what number to enter in your filament profile, and where every hot end on the market actually lands compared to its marketing spec. The honest version, not the brochure version. If you've been chasing under-extrusion ghosts on a fast printer, this is probably your fix. Table of contents What max volumetric speed actually means What to calibrate before the MVS test Running the OrcaSlicer max volumetric speed test Reading the printed test and doing the math Where to enter the result in OrcaSlicer Real hot end numbers: manufacturer claim vs measured How filament type changes your ceiling Temperature is the cheat code (within limits) Test gotchas and known OrcaSlicer bugs How MVS interacts with other calibrations FAQ What max volumetric speed actually means Max volumetric speed is the cubic millimetres per second of molten plastic your hot end can melt and push through the nozzle without falling behind. The unit is mm cubed per second, written mm³/s. I'll use that shorter form from here on. It's the single number that caps how fast you can print a given filament on a given hot end at a given temperature, and it doesn't care what speed you typed in your slicer. The formula every slicer uses is the same: volumetric demand (mm³/s) = layer height (mm) × extrusion width (mm) × print speed (mm/s) If I'm printing at 0.2 mm layer height with a 0.45 mm line width at 250 mm/s, the math comes out to 0.2 × 0.45 × 250 = 22.5 mm³/s of melt demand. If my hot end is rated for 18 mm³/s on that filament, the slicer silently drops my effective speed during slicing to around 200 mm/s so the demand stays at or below 18. You won't see a warning. You'll see a longer print time estimate than you expected, and that's the only tell. I think of it as the thermal ceiling of the hot end. Past that ceiling, filament enters the melt zone faster than the heater can deliver joules to fully melt it. Partially molten filament reaches the nozzle, the extruder gear grub-slips on the cold core inside the melt zone, and you get under-extrusion no matter how aggressive your speed settings are. The hot end isn't broken. It just can't melt that volume per second at that temperature. One detail that catches people out: MVS is per filament, not per printer. The same Revo HF hot end will give you 23 mm³/s on PLA, around 8 mm³/s on PETG, and maybe 2 mm³/s on soft TPU. The slicer enforces the cap from the filament profile, which is why every filament you use seriously is worth its own MVS measurement. OrcaSlicer's max volumetric speed test sits under the Calibration menu. The bed loads with the test object as soon as you confirm the start, end, and step values. What to calibrate before the MVS test I've seen people skip straight to the MVS test and then complain it gave them a useless number. MVS isn't a fix-everything calibration. It's the ceiling test that runs after the rest of your printer is dialled in. If pressure advance is wrong, the test will fail at corners and you'll think it's a flow ceiling. If your filament is wet, you'll measure the moisture limit, not the hot end limit. If you've got a partial clog, you'll measure your clog. Here's what I run first, in order, before I bother with MVS. The full sequence lives in the OrcaSlicer calibration guide if you want the pillar overview, but the short version is: Flow rate (extrusion multiplier). Run the flow rate calibration so your extrusion multiplier is right. This is different from MVS. Flow rate corrects how much filament you're pushing per unit length. MVS sets the upper ceiling. If your flow is wrong, the MVS test reads a hot end ceiling that's not actually your hot end's fault. Pressure advance. Run pressure advance first. A wrongly-tuned PA fakes under-extrusion at corners during the MVS test and you'll bail out 5 mm³/s below your real ceiling. The MVS print is a continuous corkscrew of cornering, so PA matters more here than on a simple cube. Temperature tower. Pick the temperature you'll actually print at. Higher temp buys more MVS (more on that below), but you need to lock the temp first so the MVS number you measure matches the conditions you'll print at. Run the temp tower at the speed you intend to print and pick the highest temp where surface finish and overhangs still pass. Dry the filament. PETG, ABS, ASA, PA, TPU, and PC all soak moisture. A wet filament pops in the melt zone and the bubbles look like flow failure. If you haven't dried, you're measuring a damp filament's ceiling, not your hot end's. Check the nozzle. A partial clog drops your measured MVS by 30-50% silently. Cold pull, swap to a known-good nozzle, or do a quick atomic pull before the test. I'll be blunt: the MVS test only tells the truth if everything upstream is honest. Treat it as the last step in a calibration chain, not the first. Running the OrcaSlicer max volumetric speed test The test lives at Calibration > Max Volumetric Speed in the top menu. The exact submenu placement has moved between OrcaSlicer versions, so if it's not where I described, check the Calibration menu for "Flow rate" or "Max Flowrate" depending on your build. Once you click it, OrcaSlicer drops a dialog with three values and loads the test object onto the bed. The defaults you'll see (verify these on your build, the project moves them occasionally): Start volumetric speed: 5 mm³/s End volumetric speed: 20 mm³/s Step: 0.5 mm³/s That gives you a 30 mm tall tower (15 mm³/s range divided by 0.5 step) where each millimetre of Z adds 0.5 mm³/s of demand. Honest reality: those defaults are tuned for a stock E3D V6 class hot end. They'll waste your time on anything modern. Here's how I set the dialog for the hot end I'm testing: Hot end classStartEndStepTower height Stock V6, basic Ender / Prusa MK3 class5200.530 mm Bambu X1C / P1S stock, E3D Volcano5301.025 mm Revo standard, Mosquito5250.540 mm Revo HF, Rapido HF, Bondtech CHT, Mosquito Magnum10401.030 mm Rapido UHF, Bambu high-flow upgrade15501.035 mm I always use a 1 mm³/s step on high-flow hot ends and a 0.5 step on stock hot ends. The 0.5 step gives more resolution but doubles the print time, and on a high-flow hot end the failure point is dramatic enough that 1 mm³/s steps are plenty granular. On a stock V6 you need the 0.5 step because the failure point is subtler. What the test object is, in plain English The print is a tall, slowly-rotating single-wall tower. Single perimeter, no infill, no top or bottom layers. As Z increases, the slicer ramps the requested volumetric flow linearly from start to end. The twist is there so any direction-dependent failure (which happens more than you'd think) shows up as a banding pattern around the print rather than only on one side. Walls are a single perimeter on purpose. The moment the hot end can't keep up with melt demand, the wall develops visible gaps. There's no second [truncated] --- ## OrcaSlicer Input Shaper Calibration for Klipper (2026 Guide) URL: https://orcaslicer.net/orcaslicer-input-shaper/ Every Voron owner has a story about ringing. Mine was a 200mm Benchy I sent to a freshly-built V2.4 at 300mm/s, the slice preview looked clean, and the print came off the bed with what looked like four overlapping echoes of the bow trailing behind every corner. Three months later I'd run the OrcaSlicer ringing tower twice, paired it with a SHAPER_CALIBRATE pass off an ADXL345, and I was back to 300mm/s with no visible artifacts. Input shaping is the difference between a printer that looks fast and a printer that prints fast. This guide walks the OrcaSlicer Input Shaping calibration end to end for Klipper machines, then covers how to push the numbers into printer.cfg, how the test compares to a real accelerometer pass, and where the tower method is honestly good enough versus where you should just wire up an ADXL345 and be done with it. I'll also touch on Marlin's M593 path and RepRapFirmware for completeness, since some of you are running mixed fleets. If you want the wider context, our OrcaSlicer calibration guide is the parent for this whole cluster. Table of contents What ringing actually is, fast Before you calibrate: the mechanical readiness check The OrcaSlicer Input Shaping calibration, step by step How to read the printed tower Saving results for Klipper Saving results for Marlin (M593) Saving results for RepRapFirmware Shaper algorithms compared Typical ballparks per printer family Tower vs ADXL345: when each one wins Troubleshooting common results When to recalibrate FAQ What ringing actually is, fast A printer frame isn't a perfectly rigid block. It's a stack of extrusions and bolted joints that flex under load, and that flex has a natural resonant frequency, typically 25 to 100 Hz on desktop machines. Bed slingers ring lower because the bed mass is heavy; CoreXY toolheads ring higher because the moving mass is lighter. When the toolhead accelerates hard into a corner, it shoves a broadband impulse into the frame, the frequencies near resonance keep ringing after the move is done, and those leftover oscillations get etched into the wall as faint horizontal ridges. Hobbyists call the artifact ringing, ghosting, or echoing. Same thing. Echo spacing on the wall is print speed divided by resonance frequency. A 60 Hz resonance at 200mm/s prints visible echoes about 3.3mm apart, which is why ringing gets dramatically worse when you crank speed without dealing with the resonance first. Input shaping is a firmware-side pre-filter on planned motion. It splits each move into two or more sub-impulses spaced one half-period of the resonance apart, so the second impulse arrives when the first's residual vibration is at the bottom of its cycle. The two cancel. The trade-off is that the move is spread across the impulse window, which blurs corners slightly. That blurring is called smoothing, and shapers tolerant of frequency error (EI, 2HUMP_EI, 3HUMP_EI) smooth more than the narrow ones (ZV, MZV). A slicer can't do this on its own; the math has to run on the actual stepper pulse train at millisecond resolution inside Klipper, Marlin, or RepRapFirmware. OrcaSlicer's calibration only prints a test pattern that excites resonance and helps you pick numbers to plug into firmware. Before you calibrate: the mechanical readiness check I'll save you a wasted afternoon. If your printer rings below about 20 to 25 Hz, the Klipper docs recommend stiffening the frame or reducing moving mass before chasing a tune. Fix the machine first. Frame-foot bolts snug, extrusion corners squared and torqued, belts tight enough to give a clean musical pitch when plucked (without overtensioning into the pulleys), no slop when you rock the toolhead by hand. On a Voron, check that panels aren't buzzing against the frame, since that adds a parasitic resonance that input shaping won't touch. Two things you'll thank yourself for. First, use a shiny, opaque filament; the OrcaSlicer wiki specifically calls out high-gloss filament because the ringing bands are subtle and matte black hides them. Light gray PLA or pastel PETG works. Second, double-check the basics from our OrcaSlicer Klipper setup guide so Moonraker is connected and your printer profile matches the machine. One honesty note: if you own a Bambu X1C, P1S, or A1, this article isn't your path. Bambu firmware runs its own internal IMU-based input shaping that calibrates automatically and doesn't accept M593. Same with Prusa MK4, MK4S, MK3.9, and CORE One; Buddy firmware ships pre-calibrated and recalibration happens through the printer's own menu (with the optional accelerometer on the xBuddy board). If that's you, use the printer's built-in tool. The OrcaSlicer Input Shaping calibration, step by step The feature lives at Calibration → Input Shaping in the top menu bar, the same submenu that holds Flow Rate, Pressure Advance, and the Temperature Tower. You need OrcaSlicer 2.3.2 or newer (the wiki specifically says "Releases greater than 2.3.2," and at time of writing the current stable line has shipped it). If you don't see the menu, you're on an older build and need to update before going further. OrcaSlicer 2.3.x: Calibration menu → Input Shaping → Frequency Test. The dialog exposes shaper type, frequency range, and the test model selector. The calibration is a two-pass tower test. Pass one measures the dominant resonance frequency per axis. Pass two refines the damping ratio once you know the frequency. You run them in that order, because the damping test needs the frequency locked in. Pass 1: Frequency test setup Open the Frequency dialog. The fields: Test Model: Ringing Tower (recommended) or Fast Tower (shorter, for printers with severe ringing). Input Shaper Type: ZV, MZV, ZVD, ZVDD, ZVDDD, EI, 2HUMP_EI / EI2, 3HUMP_EI / EI3, DAA, FT_MOTION, or "Default." Frequency range: defaults 15 Hz to 110 Hz, the wiki recommendation. Damping value: 0.0 omits the parameter and lets firmware default fill it in (Klipper falls back to 0.1, Marlin to 0.15). Or enter 0.1 to 0.2 as a manual starting point per the wiki. Pick the shaper for the first pass by firmware target. Klipper users start on MZV (Klipper's default, where most CoreXY tunes land). Marlin users pick ZV (only stable choice). On RepRapFirmware 3.4+, OrcaSlicer's "Default" selection maps to MZV (older RRF below 3.4 maps to DAA). Delta and uncertain users should start with EI, which tolerates frequency error at the cost of more smoothing. Leave the range at 15 to 110 Hz; narrow it only if a first run lands in a tight zone. Leave damping at 0.0 for the first pass. Pass 1: Slicing and printing the tower Click Calibrate and Orca pushes you into slice view with a pre-loaded ringing tower. The slicer interpolates the frequency range across the tower's Z height, so each band on the wall corresponds to a specific test frequency. The tower runs at deliberately aggressive motion settings; if the print doesn't excite resonance, banding won't appear and the test is useless. The wiki recommends roughly 20,000 mm/s² acceleration and 200 mm/s speed. Warning: those numbers are test-only. The wiki explicitly says, "Restore your 3D Printer settings to avoid keep using high acceleration and jerk values" once you're done. Don't leave 20,000 mm/s² in your normal profile or you'll snap belts and lose steps. Ringing tower in OrcaSlicer's slice preview. Each Z band maps to a frequency from the 15 to 110 Hz sweep. Slice, upload to Moonraker, print. First-layer rules still apply. Total print time on a default tower is 25 to 45 minutes depending on size. Pass 2: The damping test After the frequency tower comes off the bed and you've identified X and Y frequencies (next section), return to Calibration → Input Shaping → Damping. The dialog locks the frequencies you measured and sweeps damping. Wiki suggests starting in 0.1 to 0.2. Print, read the smoothest band, log the damping values. Typical answers land 0.05 to 0.20, with 0.1 being the Klipper default. The Klipper [truncated] --- ## OrcaSlicer Retraction Test: Solving Stringing in 30 Minutes URL: https://orcaslicer.net/orcaslicer-retraction-test/ The first PETG print I pulled off my P1S looked like a spider had gotten loose in the chamber. Hairy whiskers between every part on the plate, plus a glossy blob the size of a grain of rice on the seam of a 30 mm cube. Thirty minutes with the OrcaSlicer Retraction Test fixed 90 percent of it; the rest came down to drying the spool overnight. If you're staring at a hairy benchy or a webbed multi-part plate, you're in the right place. This walkthrough covers OrcaSlicer 2.x from menu click to saved filament profile. I'll cite the OrcaSlicer wiki and Ellis's Print Tuning Guide as we go, plus GitHub issues that bite tuners. Retraction won't save a wet spool, a worn nozzle, or a hotend running 15 degrees hot, so we'll cover that distinction too. The OrcaSlicer calibration guide sets out the full sequence for context. What's in this guide What retraction actually does (and why too much hurts) The 30-second version Direct drive vs Bowden, why the numbers differ Step-by-step walkthrough, every click How to read the towers Per-material retraction characteristics Per-printer notes When retraction can't save you Z-hop, wipe, and retract-before-wipe Common mistakes Known bugs and quirks After retraction, what to calibrate next Frequently asked questions What retraction actually does (and why too much hurts) Retraction is the cold-end gear pulling filament backward by a small amount before the toolhead travels across open air. The goal is to drop pressure inside the melt zone so molten plastic doesn't weep out during the move. CNC Kitchen has the classic primer if you want the physics, but the short version: the nozzle holds a tiny reservoir of molten plastic, and you need to pull the column above it back hard enough that surface tension holds it in place during travel. Distance and speed both matter. Pull back too little and the plug keeps oozing. Pull back too much and you create a void that has to refill before the next perimeter starts, which shows up as gaps at the start of every line. Worse, on direct drive, over-retraction yanks molten filament past the heatbreak where cooling falls off, and that's how heat creep clogs happen 30 minutes in. So the sweet spot is the smallest distance that gives you clean travel. The retraction tower finds that band by changing the retraction length per Z range and printing the same travel pattern so you can read which band stopped stringing. The 30-second version One-paragraph recipe: Open OrcaSlicer, confirm your printer and filament profile match what you're about to print. Click Calibration in the top menu bar, then Retraction Test. Leave the defaults (Start 0, End 2, Step 0.1 for direct drive; Start 1, End 6, Step 0.2 for Bowden). Click OK, slice, print. Read the tower bottom to top. Find the lowest band with no whiskers between the two posts. Plug that number into the filament profile's Retraction Length field. Save under a duplicated name. Done. The rest of this guide is about what goes sideways, per-material differences, and bugs that send people in circles. Direct drive vs Bowden, why the numbers differ The biggest mistake I see on Reddit is people copying retraction values between machines with wildly different filament paths. A Bambu A1 (direct drive) and an Ender 3 V2 (Bowden) live in separate retraction universes. On direct drive, the gear sits right above the hotend, with maybe 30 to 50 mm of rigid filament between gear teeth and nozzle. Retract 0.8 mm and nozzle pressure drops within milliseconds because there's almost no compliance. That's why direct drive numbers stay well under 1 mm. On Bowden, a long PTFE tube (around 400 mm) connects gear to hotend. PTFE compresses and stretches under load, so the first 0.5 to 1 mm of any retraction goes into "unloading" tube spring before nozzle pressure starts dropping. That's why Bowden lands at 3 to 7 mm. Polymaker's wiki on travel and retraction explains the physics without slicer-specific marketing. OrcaSlicer's defaults reflect the split: 0 to 2 mm in 0.1 mm steps for direct drive (21 bands), 1 to 6 mm in 0.2 mm steps for Bowden (26 bands). Speed matters more than people credit. Retract too slowly (say 15 mm/s) and the nozzle keeps oozing during the retraction itself, making the tower look under-retracted at any distance. Ellis recommends starting at 30 mm/s for both retract and unretract. Practical window: 25 to 60 mm/s; direct drive 35 to 50, Bowden 30 to 45. Step-by-step walkthrough, every click Sanity check first. Update OrcaSlicer to a recent 2.x build. The 2.1.x line had a bug (issue #5896) where "Retract amount before wipe" didn't apply correctly, causing phantom stringing during calibration. Fixed in later 2.x builds. Pick the printer profile that matches your machine, then pick (or duplicate) the filament profile for the spool you're calibrating. Confirm the nozzle temperature in that profile matches what you actually print at. Calibrating at 240 C and printing at 220 C invalidates the test, because hot filament strings more. If you're unsure of the temp, run the temperature tower test first. Click Calibration in the top menu bar (next to File, Edit, View). Click Retraction Test. A dialog appears with three fields: Start, End, Step. For a fresh direct-drive calibration, leave defaults: Start 0, End 2, Step 0.1. For Bowden: Start 1, End 6, Step 0.2. If you already have a tuned value and want to refine, narrow the range. Running 0.8 mm and want a tighter look? Try Start 0.4, End 1.4, Step 0.05. Click OK. OrcaSlicer auto-generates the two-tower model with band metadata baked in. The Retraction Test loads two thin towers on the plate so the toolhead has to bridge open air on every layer, which is exactly where stringing reveals itself. Slice and check the preview. Scrub through layers and confirm travel moves are visible as dashed lines crossing between the posts. Band labels are visible in the layer scrubber's metadata pane in OrcaSlicer 2.x. The sliced preview shows the travel moves crossing between towers on every layer, which is the exact stress condition you want when reading whiskers. Print. A direct-drive tower with the default range takes 25 to 35 minutes on most modern machines; Bowden runs closer to 45 because the range is wider. Walk away. Come back when it's done. Pull the print, read it bottom to top. The bottom is your Start value, the top is End. Look at the gap between posts on each band. The first band where whiskers stop or become so thin you can pinch them off is your winner. Pick the lowest clean distance. Open Filament Settings, find the Retraction tab on the override panel, enter the value. Save under a new name like "Polymaker PETG (calibrated 2026-05)" so you keep stock as a reset point. The filament settings reference walks through the override pane in detail. OrcaSlicer pins retraction length on the filament profile by default, which means your tuned PETG number follows the spool, not the printer. How to read the towers Reading the tower is half the test. Here's what each pattern means. Clean towers, no whiskers, no holes at the bottom of the next band. Done. Save the value. Hair-thin whiskers you can wipe off. Acceptable on PETG. Save the cleanest band. Tiny blobs at the start of each tower segment. Slight under-extrusion from too-aggressive retraction. Drop 0.1 mm. Glossy "snot" between towers. Massively under-retracted. If at the highest band, rerun with a wider range. Hollow gaps at the bottom of upper bands. Significantly over-retracted. Drop 0.2 to 0.3 mm. Clicking from the extruder during the test. Over-retracting or pulling too fast. Slow speed first (try 30 mm/s), then shorten distance. Stringing identical at every band. Not a retraction problem. Skip to heat creep, wet filament, worn nozzles. That last one's the most important. If the tower looks identical top to bottom, the test is telling you retraction can't fix this print. Issue #4062 on the OrcaSlicer GitHub is a long thread of people who learned that the hard way. Per [truncated] --- ## OrcaSlicer Pressure Advance Calibration: Line, Pattern, and Tower Methods URL: https://orcaslicer.net/orcaslicer-pressure-advance/ Last weekend I pulled a Benchy off the plate that looked like it had been kicked in the bow. Tiny blob on every corner, a smear at the back of the cabin, that classic "soft" appearance where the walls bulge a hair past where they should land. I knew what was wrong before I checked the file, because I'd just swapped to a new spool of PolyTerra and never re-ran pressure advance. If you've seen blobby corners on a Benchy, or pulled-in inside corners on a calibration cube that look like the printer flinched, that's pressure advance. It's the calibration that fixes "my prints look 80% there but the corners are off," and OrcaSlicer ships three ways to dial it in. The trick is knowing which one to use, how to read the result, and how to get the value into the right field so it survives the next print. If you haven't done the prerequisites yet, our OrcaSlicer calibration guide covers the order: temperature, flow, then PA. What's in this guide What pressure advance actually does The 90-second version Klipper PA vs Marlin K vs Bambu Flow Dynamics Line method walkthrough Pattern method walkthrough Tower method walkthrough How to apply the result Per-firmware deep dive Common mistakes Known bugs to watch for After PA, what to calibrate next Frequently asked questions What pressure advance actually does Here's the physical thing PA is fighting. Your extruder pushes filament into the hot end. The melt zone has a tiny bit of compressibility (the molten polymer acts like a spring), and the path from gear to nozzle has friction. When the toolhead accelerates, the extruder needs to push harder to keep flow constant, because some of that motion goes into pressurizing the melt instead of squirting plastic out. When the toolhead decelerates near a corner, that pressure is still there, and it keeps oozing for a fraction of a second after the gear stops feeding. That's your corner blob. Pressure advance compensates by reducing extrusion ahead of a deceleration and adding extrusion ahead of an acceleration. Klipper invented the modern implementation under that exact name, and the official docs explain it clearly (https://www.klipper3d.org/Pressure_Advance.html). Marlin shipped a similar feature called Linear Advance, with a unitless K factor (https://marlinfw.org/docs/features/lin_advance.html). Bambu firmware does its own version called Flow Dynamics. The names differ, the math differs, the units don't translate. What goes wrong without PA: outside corner bulges (the nozzle hesitates, the melt keeps flowing), pulled-in inside corners (pressure hasn't built up by the time the toolhead is accelerating away), inconsistent extrusion width on small features like logos and text, and that "soft" looking corner on calibration cubes where the part measures correct on calipers but the edges aren't crisp. If your prints suffer from any of those, the fix isn't more retraction or hotter temps. It's PA. The 90-second version For people who just want the recipe and want it now: Open OrcaSlicer. Top menu, Calibration, Pressure Advance, Pattern. Pick Direct Drive or Bowden in the dialog. OrcaSlicer's default test sweep populates (DDE: 0 to 0.1, step 0.002. Bowden: 0 to 1.0, step 0.02). That's the slicer's bracket range, wide enough to cover almost any printer. Confirm the filament and process profile on the right. Use the spool you actually print with. Click OK. Print the test. 12 to 20 minutes. Read the row with the sharpest corners across all four columns. The PA value is labeled at the left. Filament Settings, Advanced, tick "Enable pressure advance," type the value, save. That's it. Everything below this section is the long version: which method to use when, how to read each test, and the firmware-specific gotchas that bite people who skip the docs. The Enable pressure advance toggle lives in Filament Settings under Advanced. This is where the calibrated value goes. Klipper PA vs Marlin K vs Bambu Flow Dynamics This is the section that confuses everyone. Three firmwares, three implementations, three different number ranges, and the values are not interchangeable. Klipper pressure advance is measured in seconds. It's roughly "how many seconds of extrusion to advance ahead of the move." Klipper's docs cite typical PA values between 0.050 and 1.000 (the high end usually only on Bowden), with a starting FACTOR of 0.005 for direct drive and 0.020 for Bowden (https://www.klipper3d.org/Pressure_Advance.html). OrcaSlicer's default DDE test sweep runs 0 to 0.1 in 0.002 steps, which is wide enough to bracket nearly every printer. Many 2026 direct-drive printers (K1, Voron, klipperized Sovol) land in the 0.030 to 0.060 zone, which is why Orca's range starts at 0. Bowden Klipper lives around 0.20 to 0.60. Marlin Linear Advance uses a unitless K factor, set per print with M900 (https://marlinfw.org/docs/gcode/M900.html). LIN_ADVANCE has to be enabled at firmware compile time, and most stock Creality Marlin builds don't ship with it on. If M900 errors in the console, your firmware doesn't support it and you'll need a custom build (Jyers, Professional Firmware, or hand-rolled Marlin). Once enabled, K values for direct drive on a stock Ender 3 S1 with a Bondtech land around 0.04 to 0.10. Bowden Ender 3 stock, more like 0.20 to 0.80. Bambu Flow Dynamics is a different beast. The X1C, P1S, P1P, A1, and A1 mini all run an automatic Flow Dynamics calibration before each print by default, and on stock Bambu filaments it's accurate enough that most people never need to touch PA. The catch: if you calibrate PA in OrcaSlicer for a third-party spool and forget to disable Flow Dynamics in the print dialog, the printer's auto-cal overwrites your value (https://forum.bambulab.com/t/pressure-advance-and-k-factor-with-orcaslicer/22626). If you're choosing between slicers, our OrcaSlicer vs Bambu Studio piece compares the PA workflows directly. Quick reference table: FirmwareFeature nameUnitsTypical DDE rangeTypical Bowden range KlipperPressure AdvanceSeconds (approx)0.030 to 0.060 (community 2026)0.20 to 0.60 Marlin (LA enabled)Linear Advance / K factorUnitless0.04 to 0.100.20 to 0.80 Bambu firmwareFlow DynamicsInternal (M900-flavored)~0.020 to 0.030 (auto)n/a (no Bowden Bambus) One number 0.04 in Klipper PA is not the same physical compensation as 0.04 in Marlin K. They happen to land in similar zones for some hardware combos but that's coincidence. Always recalibrate when moving between firmwares. Line method walkthrough The Line method is the original PA test. Orca prints parallel lines at increasing PA values, and you eyeball which line is most uniform end to end. Fast (5 to 8 minutes), easiest to set up, most fragile if your first layer isn't dialed. The Line method prints parallel runs at increasing PA values. The cleanest line, no thinning at the start and no blob at the end, is your value. Setup, click by click: Calibration menu, Pressure Advance, PA Line. Pick Direct Drive or Bowden. The dialog populates OrcaSlicer's default test sweep: DDE is 0 start, 0.1 end, 0.002 step. Bowden is 0 start, 1.0 end, 0.02 step. That's the slicer's bracket range, not a typical PA value range. Confirm filament profile and process profile on the right side of the slicer. Use your daily-driver filament, not whatever was loaded last. Bed clean. PEI wiped with IPA, glue stick or hairspray if your surface needs it. Bed mesh on. If you're on a Bambu, the print dialog will offer to run Flow Dynamics Calibration before printing. Uncheck it. We're testing PA manually, the auto-cal will fight us. Print. Now the read. You want to look at each line under direct light, ideally with a phone macro lens or a 5x loupe. Run your eye along each line: Thin start, full middle: PA too low. The extruder needed to push harder at the start of the line and didn't. Full middle, blobby tail: PA too high. Pressure built up but didn't get released in time, so it kept oozing. Uniform width front to back: winner. Note the PA value labeled next to that li [truncated] --- ## OrcaSlicer Flow Rate Calibration: Pass 1, Pass 2, and YOLO Mode Explained URL: https://orcaslicer.net/orcaslicer-flow-rate-calibration/ Last winter I had a spool of orange PETG that wouldn't stop blobbing on the top surface. The temp tower came back clean at 240C, retraction was dialed, but every solid infill cap looked greasy, like wet pavement, with a tiny ridge where the lines met. I'd been running flow ratio 1.0 because that's what the stock profile said, and I assumed "stock" meant "calibrated." It didn't. After a YOLO Recommended run I landed at 0.94, and the next print came off the bed matte and clean. That 6% difference is what flow rate calibration fixes. If you've already done a temperature tower and your prints still aren't right, flow rate is almost always the next thing to chase. OrcaSlicer gives you four ways: YOLO Recommended, YOLO Perfectionist, legacy Pass 1, and legacy Pass 2. They all answer the same question (how much plastic should the slicer push through the nozzle), but the math, print time, and failure modes are different for each. This is the long version. If you want the quick win, scroll to the 60-second section. If you want to understand why YOLO landed in PR #6479 in August 2024, why Pass 2 only refines downward, and why your Bambu A1 might silently break the test, keep reading. What's in this guide Why flow rate matters more than you think The 60-second version How flow ratio works in OrcaSlicer The two methods inside Orca YOLO mode walkthrough 2-Pass legacy walkthrough When YOLO wins, when 2-Pass wins Per-material defaults and what to expect Per-printer notes Common mistakes Known bugs to watch for After flow rate, what to calibrate next Frequently asked questions Why flow rate matters more than you think Flow ratio is the multiplier OrcaSlicer applies to every line of plastic it tells your extruder to push. Too high, the nozzle deposits more material than the slicer planned for. Too low, you get gaps. The symptoms get blamed on temperature or speed all the time, which sends people down the wrong rabbit hole. Three places it shows up. Top surfaces are the loudest tell. A correctly calibrated flow ratio gives a flat, satin-matte top with no line ridges and no over-extruded shine. If the top looks "wet" or has tiny waves, you're high by a percent or three. If you see daylight through gaps under raking light, you're low. Second is dimensional accuracy. A 20mm calibration cube printed at flow ratio 1.05 will measure roughly 20.4mm on the X and Y faces because every perimeter is laying down 5% extra material. People chase this with "horizontal expansion" settings instead, which is a band-aid over the actual problem. Third is perimeter wall thickness. Slice a part with two perimeters at 0.42mm line width and the wall should measure 0.84mm. If it's reading 0.92mm, you're 9% over and your part doesn't fit the bearing it was designed for. I've seen this kill print-in-place hinges over and over. If you haven't worked through the rest of the calibration ladder yet, my OrcaSlicer calibration master guide covers the order. Flow comes after temperature and either before or after pressure advance. Both orderings work, just don't change temperature after you flow-cal. The 60-second version For people who just want the answer: open OrcaSlicer, go to the Calibration menu in the top bar, hover Flow Rate, click YOLO Recommended. A new project loads with 11 small blocks. Set your filament, slice it, print it, look at the result under raking light, pick the block whose top surface looks the smoothest. Read off the modifier (something like +0.02 or -0.03), add it to your current flow ratio, save the new value in your filament profile. The Flow ratio field lives in Filament Settings. This is the one you update with your calibrated value. Total time on a Bambu P1S with stock PLA Basic is about 18 minutes, closer to 30 on a slower Ender 3. One print, one decision, one number saved. The rest of this article is why that's enough most of the time, when it isn't, and what the legacy 2-Pass workflow does differently. How flow ratio works in OrcaSlicer Flow ratio is a multiplier on the volume of plastic the slicer plans to extrude per unit of distance. Default is 1.0 for most generic profiles, 0.98 for Bambu-tuned profiles. Useful real-world range is 0.95 to 1.05, with most well-behaved filaments landing between 0.94 and 1.00. Cross-slicer confusion: PrusaSlicer calls this same setting extrusion multiplier. Cura calls it flow as a percentage, so flow ratio 0.97 in Orca equals 97% in Cura. Marlin firmware exposes it at runtime as M221. They're the same number, just labeled differently. If you've calibrated extrusion multiplier in PrusaSlicer to 0.96 for a specific spool, that translates directly to 0.96 flow ratio in OrcaSlicer. Bambu Studio's flow rate calibration is conceptually the same, but Bambu Studio's print prep also has an on-printer flow cal step that fights OrcaSlicer's test. My OrcaSlicer vs Bambu Studio comparison covers how these flows differ in practice. Subtle gotcha: Orca has both print_flow_ratio (print preset) and filament_flow_ratio (filament preset). They multiply at slice time. The cal tool historically wrote to print_flow_ratio while docs said filament_flow_ratio, which is issue #9261. Marked closed, but make sure your saved value lands in the filament profile or you'll be hunting your own ghost. The two methods inside Orca Before I get into YOLO and Pass 1/Pass 2, there's a more fundamental split: how do you decide which test square is "best"? OrcaSlicer's built-in tools default to the top-surface method, but the wider 3D printing community is split. The top-surface method is what Orca builds around. Each test block prints with several top layers (YOLO uses 5 top layers on a 2mm-thick block). The slicer varies the flow ratio across blocks, you look at the tops under raking light and pick whichever looks cleanest. No tools needed. The pattern OrcaSlicer uses is "Archimedean Chords," meaning the inner spiral prints last, so any over-extrusion shows up as a visible bump where the spiral meets the outer arcs. Clever visual cue once you spot it. The wall-thickness method works differently. You print a single-wall hollow object (a 30x30x10mm cube with one perimeter, no infill, no top), measure with calipers at multiple points, and compute: new_flow = current_flow * (expected_wall_width / measured_wall_width). If line width is 0.42mm and calipers read 0.45mm at flow ratio 1.0, your new flow ratio is 1.0 * (0.42 / 0.45) = 0.933. Six points averaged, away from the seam, light pressure. The Ellis Print Tuning Guide is famously skeptical of the wall method on the grounds that wall thickness is affected by pressure dynamics, not just flow. Real argument there. On stringy materials like PETG, though, top-surface gets noisy because filament fuzz hides the gaps. I use top-surface for PLA, wall method for PETG and TPU. Run both on a new printer the first time as a sanity check. YOLO mode walkthrough YOLO mode is genuinely named like a meme, because it is one. Stop running a coarse pass plus a fine pass, bet you're already in the ballpark, run one tight test. SoftFever merged it as PR #6479 on August 22, 2024 with input from igiannakas and jeremytodd1 in the discussion. It's been the recommended path since 2.2.x. Two YOLO variants: YOLO Recommended: 11 blocks, modifier range -0.05 to +0.05, step size 0.01. YOLO Perfectionist: 11 blocks, modifier range -0.04 to +0.035, step size 0.005. The math is additive, not multiplicative: NewFlowRatio = OldFlowRatio + selected_modifier. Start at 0.98, +0.02 block wins, new flow ratio is 1.00. No rounding errors, no compounding mistakes from a previous pass. To run it, click Calibration -> Flow Rate -> YOLO Recommended. A dialog opens asking for your starting flow ratio (defaults to whatever your filament profile already has). Confirm, and a new project loads with all 11 blocks laid out. The YOLO Recommended plate after slicing. Eleven blocks, each labeled with the flow modifier it's printing at, all on one build. Slice and print. The plate [truncated] --- ## OrcaSlicer Temperature Tower Tutorial: Find Your Sweet Spot in One Print URL: https://orcaslicer.net/orcaslicer-temperature-tower/ I had a roll of Sunlu Silk PETG that was throwing webs across every print like a haunted house. Two hours and one temperature tower later, I was running it at 235 instead of 245 and the strings vanished. That's the whole pitch for this article. One 90-minute print can save you a week of "why is this filament so bad" frustration. The thing is, most people either skip the temp tower entirely or do it wrong. They eyeball the surface, pick whatever looks shiniest, and move on. Then they're confused when their layer adhesion is brittle six months later. I've been there. The temp tower isn't about picking the prettiest block. It's about picking the lowest temperature that still passes a snap test, and OrcaSlicer's built-in wizard makes that almost trivial once you know what you're looking at. This is a sibling piece to the full Orca calibration guide, so if you haven't tuned anything yet, start there. If you're new and don't have the slicer installed, you can download OrcaSlicer first. Already running 2.3.x? Good, let's go. What's in this guide Why temperature matters more than people think The 30-second version How OrcaSlicer's temp tower works under the hood Step-by-step walkthrough How to read your tower Per-material recommended ranges Per-printer notes Common mistakes I see in the forums When the wizard breaks After the tower, what to calibrate next Frequently asked questions Why temperature matters more than people think Plastic isn't binary. It doesn't go solid to liquid at one neat number. Every thermoplastic has a glass transition temperature (Tg) and a melt zone above that. PLA's Tg sits around 60°C, its melt zone runs roughly 180 to 220°C. PETG's Tg is closer to 80°C and it melts cleaner around 230 to 250°C. Layer adhesion is a chemistry problem, not a "make it hotter" problem. Bottom of the melt zone, polymer chains barely entangle across the layer boundary. You get a part that looks fine but cracks along a layer line when you flex it. Push too hot and you cook the polymer, get heavy stringing, and on Bowden setups you risk heatcreep. The sweet spot is the lowest temperature that still gives you full layer bonding. The temp tower is just a fast visual way to find it without a tensile tester. The 30-second version For people who already know the drill and just want the recipe: The two clicks that start everything. Calibration sits right next to Help on the top menu in OrcaSlicer 2.3.x. Open OrcaSlicer. Click Calibration > Temperature. Pick your filament family from the dropdown. Keep the default start and end temps unless you have a specific reason. Click OK. A new project loads with the tower model on the plate. Slice. Print. Roughly 60 to 90 minutes on a 0.4 mm nozzle. Snap a block off with pliers. The lowest temperature that still tears across multiple layers (instead of clean-breaking along a layer line) is your answer. Save the result to a custom filament profile so the next Orca update doesn't overwrite it. That's it. The rest of this guide is about doing each step correctly, because every single one of those bullets has a way to go wrong, and I've seen all of them on the forums. How OrcaSlicer's temp tower works under the hood The tower model itself is a remix of gaaZolee's "Smart compact temperature calibration tower" from Thingiverse 2729076. OrcaSlicer ships its own version with embossed temperature labels on each block. Each band is roughly 10 mm tall at a 0.20 mm layer height, which works out to about 50 layers per band. The default tower has 8 bands, so you're looking at 80 mm of vertical print, give or take. The clever bit is how the temperature actually changes. OrcaSlicer doesn't pause the print and reheat. It injects an inline M104 S<temp> command at the boundary of each band. M104 is the non-blocking "set hotend temperature" gcode, so the print head keeps moving while the heater chases the new setpoint. That's intentional. If Orca used M109 (the blocking version that waits for the temp to stabilize), the print head would freeze for 10 to 30 seconds at every band boundary and you'd get a fat heat zit on the seam every 10 mm. Useless. Two practical consequences of this. First, the temperature lags the band by maybe 5 to 10 layers as the heater swings. The transition isn't instant. That's fine, you've got 50 layers per band, you're reading the middle 30 anyway. Second, if your firmware ignores inline M104 commands (this comes up on some heavily-customized Klipper configs and weirdly-flashed Marlin builds), the tower won't change temperature at all. The fix is to export the sliced gcode, search for "M104" in a text editor, and confirm those commands are actually in there. If they are and the printer's still ignoring them, your firmware is the problem. Scrubbing the layer slider near a band boundary should reveal the inline M104 command. If it's missing, regenerate the project before printing. The 5°C step is locked. You can't change it to 3°C or 10°C, the wizard won't let you. You can only set start and end. That's a design choice (5°C is a meaningful difference for most filaments), not a missing feature. Step-by-step walkthrough (every click) This works on Windows, macOS, and Linux. The UI is identical. I'm running 2.3.x on a Bambu P1S as the test rig, but the steps are universal. Pick the right printer profile. Top right corner, click the printer dropdown. Make sure the nozzle size matches your physical hotend. Default profiles are 0.4 mm, and if you've swapped to a 0.6 mm or 0.8 mm hardened nozzle and forgot to switch the profile, your extrusion volumes are wrong and the tower will lie to you. There's a long forum thread on this exact mistake with PETG on a 0.6 mm P1S, it's worth a skim if you've got a non-default nozzle. Pick the right filament profile. Right-side filament slot. Use the actual brand profile if it ships in Orca's library (Bambu PLA Basic, Polymaker PolyTerra, eSun PLA+, and so on). If your brand isn't there, pick "Generic PLA" or "Generic PETG" and we'll override the temp from the tower result later. The bed temp, fan curve, and retraction settings all come from this profile, so don't run a TPU tower with a PLA profile loaded. Open the Calibration menu. Top menu bar, click Calibration. On most builds it sits between Configuration Wizard and Help. Click Temperature. The dialog opens. You'll see a filament family dropdown, start temp, end temp, and the fixed 5°C step. Set start and end. Use the table in the next section, or just keep defaults. For the cleanest read, keep the total range at 40°C. That gives you 8 bands, which is enough resolution to spot the inflection point. Going narrower (say 220 to 240) sometimes doesn't print the embossed labels because of issue #10041. Click OK. A new untitled project loads with the tower model already on the build plate. Verify the slice. Click "Slice plate." Open the preview, drag the layer slider. You should see the gcode preview pane show M104 commands at each band boundary. If you don't see them, regenerate the project, your .3mf might be corrupted. Send to printer. Bambu users can use Send over LAN or cloud. OctoPrint folks use the OrcaSlicer plugin. Everyone else exports the sliced file to SD or USB. Print. Run time on a 0.4 mm nozzle is 60 to 90 minutes depending on range. PA CF and PC towers run longer because layer cooling is more conservative. Read the tower. See the next section, this is where most people botch it. Apply the result. File > New project. Open filament profile settings for the filament you tested. Set "Nozzle temperature initial layer" to your winning temp + 5, and "Nozzle temperature other layers" to your winning temp. Save the profile under a custom name (something like "Polymaker PolyTerra Tuned"). Built-in profiles get overwritten on Orca updates, custom ones don't. The wizard dialog. The 5°C step is locked, you only get to set the bookends. What you should see after clicking OK. Each band gets an embossed temperature label on the fr [truncated] --- ## The Complete OrcaSlicer Calibration Guide (Order Matters) URL: https://orcaslicer.net/orcaslicer-calibration-guide/ I spent six weeks blaming my Voron 2.4 for prints that came out fine, just not great. The corners had little bulges. The top surface looked slightly pillowed. Stringing on PETG no matter how I tuned retraction. I rebuilt the toolhead, replaced the hotend, even swapped the motherboard. None of it fixed anything. The fix was a flow ratio of 0.96 instead of 1.00, a pressure advance of 0.038 instead of the printer-default 0.05, and a max volumetric speed cap that I'd never set so OrcaSlicer was overrunning my hotend by about 30%. Three calibrations. About four hours of test prints. The "bad printer" turned out to be perfectly fine. This is the guide I wish someone had handed me before week one. It covers every calibration in OrcaSlicer's Calibration menu, in the order I actually run them, with the values you'll probably land on for the common filaments. Written against OrcaSlicer v2.3.2, the current stable build (13.7k GitHub stars at the time of writing). Why calibrate at all? OrcaSlicer ships with sensible defaults for most printers. If you just want to print PLA Benchies, the defaults usually work. So why does the rabbit hole exist? Because every printer has small mechanical tolerances. Every filament has different rheology. Every hotend has a slightly different thermistor offset. The defaults assume "average everything", and most prints end up averagely-okay. Calibration replaces those averages with the actual numbers for your rig and this spool. What you get from a few hours of calibration: Stringing drops from "have to clean every print" to "occasional whiskers" Top surfaces stop pillowing and start looking smooth like injection-moulded parts Sharp corners stop bulging or losing material Dimensional accuracy good enough to print bearings, gears, and threaded parts that fit on the first try Print speed safely 30 to 50% higher because you know your hotend's real flow ceiling Settings you can copy-paste between filament brands without re-discovering them The calibration suite is the single biggest reason power users choose OrcaSlicer over Bambu Studio. Bambu Studio ships with two calibrations (Flow Dynamics and Flow Rate). OrcaSlicer ships with at least nine. That's not a marketing number, it's the actual menu count when you open Calibration in v2.3.2. The order debate (and what to actually do) Here's the thing nobody tells you up front: the OrcaSlicer wiki and the most-shared online guides give different orders. This is not a small disagreement. The Calibration menu inside the app lists the tests in this sequence: Temperature, Flow Rate, Pressure Advance, Max Volumetric Speed, Retraction, Tolerance, Cornering, Input Shaping, VFA. The official wiki Calibration Guide page suggests almost the same. But search "OrcaSlicer calibration order" on Reddit or any of the dozen tutorial blogs and you'll see this instead: Temperature, Flow Ratio, Pressure Advance, Retraction. Volumetric Speed and the rest get tacked on at the end as "advanced". I've gone through GitHub Discussion #2476 where a user asked SoftFever directly which order to follow. The maintainer never posted a definitive answer. The top-voted community reply admits "the tests likely interact, you might want to re-run after each major change." That's the honest version. Here's what I do, and what most successful Voron and Bambu users I know also do. Treat calibration in three stages, not as a flat list: Stage 0 (hardware, not OrcaSlicer): bed level, Z-offset, e-steps or rotation_distance. If these are wrong, every other calibration will lie to you. Stage 1 (per filament): Temperature Tower, Max Volumetric Speed, Flow Ratio, Pressure Advance, Retraction. Run these every time you switch to a new spool brand or color. Maybe twenty minutes of slicing time, an hour of print time. Stage 2 (per filament, optional): Tolerance Test. Only matters if you're printing functional parts that need to fit. Skip for cosmetic prints. Stage 3 (per printer, once or after mechanical changes): Cornering, Input Shaping, VFA. These tune the printer's motion system. They don't change with filament. If you only have time for three calibrations, do Temperature, Flow Ratio, and Pressure Advance. That'll cover ninety percent of "my prints look bad" complaints. Everything else is polish. Stage 0: the hardware basics OrcaSlicer can't fix Don't skip these. I've watched people run six rounds of flow rate calibration trying to figure out why nothing converges, when the real problem was their extruder's rotation_distance was off by 5%. Bed level: print a flat 100mm square at 0.2mm layer height. If any corner shows a different first-layer squish than the others, your bed isn't level. Re-mesh. Modern Bambu, Creality K-series, Prusa MK4, and Voron with KAMP all auto-mesh on each print, so this is mostly handled. On older machines, run a manual mesh. Z-offset: the gap between nozzle and bed on layer one. Too high and your first layer doesn't stick. Too low and the nozzle scrapes plastic into the bed. The "paper test" works fine for getting close (slide paper under the nozzle, adjust until it drags slightly). Live-tune from there during the first layer of a calibration print. E-steps (Marlin) or rotation_distance (Klipper): tell the firmware exactly how much filament one motor revolution pushes. Set extruder to push 100mm of filament with the hot end at temperature, measure how much actually pushed (mark the filament before, measure with calipers after), and adjust the firmware variable proportionally. This is upstream of OrcaSlicer's flow ratio. If e-steps are wrong, flow ratio just compensates badly. Calibration: Temperature Tower Path: Calibration → Temperature What it fixes: the optimal hotend temperature for this exact filament. Brand and color both matter. Black PLA from Brand A often runs five to ten degrees different from white PLA from Brand B because of pigment additives. The defaults in the filament profile are reasonable starting points, not destinations. OrcaSlicer pre-fills a sensible range when you pick a filament: 180 to 230°C for PLA, 220 to 250°C for PETG, 230 to 270°C for ABS, 220 to 245°C for TPU. Each tower segment is 5°C. Print the whole thing, let it cool for ten minutes, then inspect. What you're looking for, top to bottom: Stringing between the pillars (less stringing = better) Surface finish on the flat faces (smoother = better) Layer adhesion (try snapping a segment with pliers, the layer that breaks should look like a clean fracture, not delaminate cleanly) Bridge sag on the overhang feature (less sag = better) Any temperature-related defects like blobs, bubbling, or oozing Pick the lowest temperature where layer adhesion is still solid and stringing looks acceptable. Going lower reduces oozing and improves overhangs. Going higher improves layer bonding but worsens stringing. Tied results, pick the lower number. The OrcaSlicer temperature tower drops the hotend temp 5°C every floor. Inspect each segment for stringing, surface finish, and bridge sag, then pick the lowest temperature where layer adhesion still feels solid. Common landing zones for a 0.4mm nozzle: PLA: 200 to 215°C (Bambu PLA Basic typically 210, Polymaker PolyTerra 205) PETG: 235 to 245°C (240 is the most common landing) ABS: 240 to 250°C ASA: 245 to 255°C TPU 95A: 220 to 235°C Gotcha: the default 5°C step is too coarse for high-speed printing where 2 or 3°C swings genuinely change quality. Override start, end, and step manually for fine work. Second gotcha: if your hotend has a poor PID tune, every block prints inconsistently and the tower lies. Run a PID tune on the printer firmware first if you suspect drift. Calibration: Flow Rate (use YOLO, not the old two-pass method) Path: Calibration → Flow Rate → YOLO (Recommended) or Pass 1 / Pass 2 Flow rate is the extrusion multiplier. Set it wrong and your entire print is over- or under-extruded by a few percent, which shows as either pillowy ridges on top surfaces (over) or pinhole gaps and weak wa [truncated] --- # Section: Filaments ## OrcaSlicer Filament Drying Guide: When and How to Dry It URL: https://orcaslicer.net/orcaslicer-filament-drying/ I once cracked open a fresh roll of PA6-CF, slapped it on the spool holder, and watched my nozzle pop like microwave popcorn within ninety seconds. I'd let the bag sit half-open on the workbench for two days, and that was enough to wreck the print, the part, and most of my afternoon. If you've ever heard that crackling sound or pulled a vase off the bed that looks like it was sandblasted, you're reading the right guide. This is the practical drying playbook I wish I had when I started running nylon and PETG through my Bambu rig. We'll cover what wet filament actually sounds and looks like, the per-material drying temps straight from vendor TDSs, how the AMS HT and AMS 2 Pro compare for in-printer drying, what standalone dryers are worth the bench space, and why your kitchen oven is a last resort and not a first choice. I'll also flag where OrcaSlicer can mask wet-filament symptoms (the Cooling tab), and where it categorically can't help (drying itself). Table of contents Why your filament is making popping sounds Which filaments are hygroscopic, and which barely care Per-material drying temperature and time table In-printer drying: AMS HT vs AMS 2 Pro vs original AMS vs AMS Lite Standalone filament dryers compared Can you use a kitchen oven? The OrcaSlicer angle: Cooling tab tuning Storing filament after drying Common gotchas FAQ Why your filament is making popping sounds The single most diagnostic symptom of wet filament is audible. You'll hear popping, crackling, or hissing right at the nozzle, and that's not the hotend, that's trapped water flashing to steam as the filament enters the melt zone. Sovol's writeup describes it as a violent little explosion inside the nozzle, and that's exactly what's happening, just at a microscopic scale. The visual tells stack up fast once you know what to look for. You'll see wisps of steam coming off the nozzle in good light. The print surface goes from glossy to foamy or pitted. Stringing gets dramatically worse on PETG and TPU, even with retraction settings that worked fine yesterday. Inter-layer bonding weakens, and parts snap along layer lines under loads they used to handle. Wet PLA stringing on a vase print versus dry PLA on the same model side by side Here's the part most beginners miss: drying removes water, but it doesn't undo hydrolysis. Once water molecules have been sitting at melt temperatures inside the polymer, they act like tiny chemical scissors, severing the long polymer chains. That damage is permanent. A long-wet spool of PETG might print after drying, but it'll often still be brittle because the polymer backbone has been chopped up. So drying isn't a fix for filament you've been printing wet for weeks; it's a fix for filament that's just sat in humid air. Don't get hung up on chasing a specific moisture percentage. Vendor TDSs don't publish a "below X%" threshold for printable, and pretending there's one is internet folklore. Use the symptoms instead: popping, visible steam, fuzzy surface, sudden stringing on a profile that used to work. If two of those three show up, dry the spool before you change a single slicer setting. Which filaments are hygroscopic, and which barely care Not every filament drinks water at the same rate. Roughly ordered worst to best for absorption: nylon (PA, PA-CF, PA-GF) is the worst offender, followed by PVA, polycarbonate (PC), PETG, ABS, ASA, then PLA at the milder end. TPU sits somewhere in the middle and varies wildly between brands and shore hardness. Nylon can become functionally unprintable within a humid day. PVA goes squishy in a few hours of high-humidity exposure. PLA can sit on an open shelf for weeks in a dry climate and still print fine, which is why so many makers skip drying entirely. The trap is that habit catches up the moment you switch to PETG or nylon. Absorption is a function of humidity, temperature, and time, not just the material. The same spool of PETG behaves differently in Phoenix in August than in London in February. I keep a cheap hygrometer on the workbench, and anything above 45% relative humidity means spools come out of dry-boxes only when I'm loading the printer. Per-material drying temperature and time table This is the table I check before every drying cycle. Every row pulls from the actual vendor TDS, not forum guesses. If your filament brand isn't Polymaker or eSUN, check your specific TDS, but these values are a sane starting point for most common formulations. MaterialTempTimeSource PLA (Polymaker PolyLite / PolyMax / PolyTerra)55°C6 hPolymaker TDS HT-PLA (Polymaker)60°C6 hPolymaker wiki PETG (PolyLite PETG)65°C6 hPolymaker TDS ABS (PolyLite ABS)70°C6 hPolymaker TDS V5.5 ASA (PolyLite ASA)70°C6 hPolymaker TDS V5.3 TPU 95A (PolyFlex TPU95)65°C8 hPolymaker TDS V5.1 TPU 95A-HF70°C8 hPolymaker wiki Nylon PA6-CF (PolyMide PA6-CF)80°C12 h (oven)Polymaker TDS V5.2 Polycarbonate (PolyLite PC)75°C12 hPolymaker TDS V5.3 PVA (Polymaker PolyDissolve S1)80°C12 hPolymaker wiki PVA (eSUN ePVA)45°C>10 heSUN TDS V4.0 BVOH (Verbatim / Ultrafuse)Follow vendor TDS for your specific BVOH grade; published Ultrafuse and Verbatim datasheets cover storage only.Ultrafuse / Verbatim datasheets Per-material drying temperature and time chart graphic A note on PVA between the two main brands: Polymaker's PolyDissolve S1 wants 80°C for 12 hours, while eSUN's ePVA wants only 45°C for more than 10 hours. That's a huge spread, and it's because the polymer chemistry isn't identical between brands. Don't average them. Read your specific TDS. Bambu's own wiki suggests a different convention for the "dry while printing" use case. It sets PLA to 45°C and PETG to 55°C when drying happens alongside a print, on the logic that the heated chamber acts as a long, gentle dry. That's a perfectly valid maintenance mode, but it's not a replacement for the higher-temp recovery cycle the TDS calls for when a spool has actually gone wet. In-printer drying: AMS HT vs AMS 2 Pro vs original AMS vs AMS Lite If you're on the Bambu ecosystem, your in-printer drying options vary wildly across the AMS lineup. They aren't interchangeable, and three of the four can't actually dry filament at all in the active-heat sense. AMS HT is the new high-temperature unit and it's the only Bambu AMS that hits truly useful drying temperatures. Per the Bambu wiki, it has a "built-in heating and drying module with a maximum drying temperature of 85°C," and the spool rotates 30° every five minutes for uniform drying. Two important caveats: it holds a single spool, and "each AMS HT requires an external 220V power supply for drying," meaning you can't draw drying power from the printer. The wiki is also explicit that "the AMS HT cannot perform standalone drying without a connection to the printer." Per Bambu Lab specifications it pulls roughly 170W during drying cycles, though I'd treat that figure as a ballpark rather than a guaranteed continuous draw. Bambu Lab AMS HT mounted on H2D printer with spool inside AMS 2 Pro hits 65°C max via "two drying modules, consisting of heaters, fans, and temperature sensors," per the official drying-function page. There's a gotcha buried in the same doc: "the maximum temperature is only achievable when the ambient temperature is ≥25°C." If your garage is at 18°C in winter, the AMS 2 Pro will dissipate heat faster than it can produce it and your actual chamber temp won't reach 65°C. That makes it a decent in-place dryer for PLA, PETG, and ABS in a climate-controlled room, but a poor fit for nylon, PC, or PVA, which need higher temps. Bambu themselves point users to the AMS HT for those materials. Original AMS (the unit shipping with X1 and P1 series before AMS 2 Pro) has no active heating at all. It's desiccant-only, with a humidity sensor and an orange-indicator desiccant pack that changes color when spent. You replace the pack when it goes dark. That's passive storage, not drying, and you cannot use it to recover a wet spool. AMS Lite (A1 and A1 mini) is open-fr [truncated] --- ## How to Use Bambu Lab AMS Filaments with OrcaSlicer (2026 Guide) URL: https://orcaslicer.net/orcaslicer-bambu-ams-filaments/ The first time I watched my X1C burn through a 14-color print using an AMS 2 Pro stacked with an AMS HT, I sat there with coffee getting cold because the H2D in the next room was doing the same trick with its own HT chain. I'd spent a weekend wiring up profiles in OrcaSlicer, and once the RFID hits and the slots light up with the right Bambu PLA Basic colors, the whole thing feels almost too easy. Then I plugged in a third-party eSun PLA spool, and the illusion broke. That's the AMS family in 2026. When it works, it's the slickest filament-handling stack in consumer FDM. When it doesn't, you're poking at the Device tab pencil icon, wondering why your custom preset won't stick, or staring at a printer that won't talk to OrcaSlicer because the January 2025 firmware update flipped a switch nobody told you about. This guide covers how to use Bambu Lab AMS filaments with OrcaSlicer, every variant, every material, every gotcha. Table of contents The four AMS variants in May 2026 Which printer takes which AMS Material compatibility, variant by variant RFID auto-detection for Bambu spools Manual slot mapping for third-party spools OrcaSlicer Device tab walkthrough LAN Mode plus Developer Mode after firmware 01.08.03.00 Multi-color setup and the Type 2 wipe tower Dry-while-print on HT and 2 Pro Common gotchas Spoolman, inventory, and what it does for Bambu owners FAQ The four AMS variants in May 2026 If you've come to OrcaSlicer in the last six months, you've probably noticed that "AMS" is now a family, not a product. There are four distinct units shipping as of May 2026, and they don't all do the same job. Two of them aren't even the same shape. Four Bambu AMS units side by side, original AMS, AMS 2 Pro, AMS HT single-spool, AMS Lite. Here's the rundown, with the specs you actually need to feed into OrcaSlicer's filament configuration. VariantSlotsActive dryingMax drying tempNotes Original AMS4No, desiccant compartments onlyn/aSealed enclosure, RFID-equipped AMS 2 Pro4Yes65 °C (only at ambient ≥25 °C)Dynamic air ventilation, dry-while-print AMS HT1 (single-spool)Yes85 °C170W heater, requires external 220V supply AMS Lite4Non/aOpen structure, A1 and A1 mini only The detail that catches people out: AMS HT is a single-spool unit, not a 4-slot like the others. The product page over at the Bambu Lab AMS HT store listing confirms this, and so does every retailer carrying it. You don't buy "an HT" and load four spools in it. You buy one HT per spool of PA-CF or PC you want hot-dried, and you chain multiple units together if you need more than one engineering filament loaded at once. That chaining is how Bambu gets to the "up to 24 filaments in one print job" headline. Per Bambu Wiki, you can run up to 8 AMS HT single-spool units alongside 4 AMS 2 Pro modules on supported printers, totaling 24 filaments addressable. I've run 2 HT plus 1 AMS 2 Pro on an X1C and OrcaSlicer handled the slot mapping cleanly. The AMS Lite is the odd one out. It's open-frame, not sealed, and exclusive to the A1 series. Per the AMS Lite FAQ on Bambu Wiki, it "is not compatible with X1 and P1 series printers" and "cannot be combined with other AMS models." You can't bolt a Lite onto an A1 setup that already has another AMS. Which printer takes which AMS Here's the cleanest mental model I've found for which AMS works with which printer. A1 and A1 mini: AMS Lite only. One unit, 4 colors max. P1P, P1S, X1C, X1E: original AMS, AMS 2 Pro, AMS HT. Up to 4 AMS HT can chain on X1/P1 series per the Bambu Wiki AMS HT connection guide. H2D: original AMS, AMS 2 Pro, AMS HT, and supports the heaviest chaining for that 24-filament number. H2D also gets a couple of OrcaSlicer features the other printers don't, including the filament-grouping button (more on that in the multi-color section). There is no scenario in May 2026 where the AMS Lite plays nicely with an H2D or an X1. I know it looks similar from across the room, but the feeder mechanism, the data protocol, and the printer's filament-routing assumptions are all different. Material compatibility, variant by variant This is where most of my own troubleshooting time has gone. Bambu's general AMS material guide says the sealed AMS family (original AMS, 2 Pro, HT) supports PLA, PETG, ABS, ASA, PET, PA, PC, PP, POM, HIPS, plus Bambu's PLA-CF, PAHT-CF, and PETG-CF. The wiki is explicit that TPE, generic TPU, PVA, and most third-party carbon or glass fiber filaments are not compatible across the board. There's one nuance: Bambu's own "TPU for AMS" SKU is formulated specifically to feed through the sealed AMS variants. Generic flexible TPU still isn't safe. Material compatibility matrix table MaterialOriginal AMSAMS 2 ProAMS HTAMS Lite PLAYesYesYes (overkill)Yes PETGYesYesYesYes ABS / ASAYesYesYesYes (open-structure caveat) PA (Nylon)Feeds, can't dryFeeds, dries to 65 °C (incomplete)Yes, dries to 85 °CNot recommended PCFeeds, can't dryFeeds, drying incompleteYes, dries to 85 °CNot recommended PA-CF / PAHT-CFYes (Bambu)Yes (Bambu)Best choiceNo, CF damages channel TPU 95A (Bambu "for AMS" SKU)YesYesCheck Bambu wiki per SKUNo Generic TPU / TPENoNoNoNo PVAYes (sealed)Yes (sealed, can dry-print)YesNot recommended (open structure) BVOHYesYesYesNot recommended The PA caveat bit me hardest. I bought an AMS 2 Pro thinking 65 °C was enough for nylon, and it isn't. Per Bambu's AMS 2 Pro/HT FAQ, PA and PC require higher drying temperature and the 2 Pro can't dry them completely, so Bambu recommends the HT for these filaments. The 2 Pro will feed PA and dry it partially, but you'll still get stringing and the surface finish of a damp spool. PVA in the AMS Lite is the other gotcha. The Lite is open-frame, and per Bambu's PVA guide, the open structure means PVA continues absorbing moisture during the print. Bambu's own recommendation for PVA in the Lite is to keep prints under one hour. For long multi-day prints with PVA supports, you want a sealed AMS, ideally the 2 Pro so you can dry-print. Our printing PVA and BVOH supports guide goes deeper. RFID auto-detection for Bambu spools This is the magic part. Every Bambu-branded spool has RFID tags on both sides of the cardboard, and each AMS unit (including the HT and 2 Pro) has RFID readers built into the slot. Drop a Bambu PLA Basic spool in, and within a few seconds the slot in OrcaSlicer's Device tab fills with the right filament profile, K-value, and color swatch. You don't pick anything from a dropdown. OrcaSlicer Device tab showing AMS panel with four loaded slots and one empty OrcaSlicer pulls this data straight from the printer over the local network. You get three modes to control the behavior, settable on the printer touchscreen: Read Filament on Insertion (reads when you slot a new spool), Read Filament on Startup (reads everything on power-on), and Manual Read (tap a slot to force a re-read). I leave mine on Read on Insertion because I'm constantly swapping colors. If you mostly leave the same four colors loaded for weeks, Read on Startup is fine and a bit gentler on the readers. If RFID isn't firing, check that the spool is seated flush against the back of the slot and try a manual read from the touchscreen. Manual slot mapping for third-party spools Now the part that costs me actual hours. Third-party spools don't have Bambu RFID. eSun, Polymaker, Overture, Sunlu, doesn't matter, you have to map each slot manually in OrcaSlicer's Device tab. The workflow looks like this: Open OrcaSlicer and click the Device tab (it's one of the four main tabs across the top). Find the AMS panel on the right side. Each slot has a little pencil icon in the corner. Click the pencil on the slot you want to edit. Choose the brand from the dropdown, then the material below it. Click Confirm. That's the official workflow. In practice, a few things go wrong often enough to know up front. OrcaSlicer issue #3874 documents the most common bug: custom filaments don't persist after you click Confirm, the slot reverts. Workaround: save y [truncated] --- ## OrcaSlicer PVA and BVOH Soluble Support Settings URL: https://orcaslicer.net/orcaslicer-pva-bvoh/ The first PVA spool I loaded into an MMU2S sat on a shelf for nine months before I tried it, and the first layer foamed like a soda can the second the nozzle hit temperature. I ran it through a 45 degree dryer for twelve hours, reloaded, and watched the same spool lay down a perfect dense interface against a PLA shell that came off the bed free of witness marks. That gap between wet brick and clean release is the whole story of soluble supports, and it's why I've spent more time tuning PVA and BVOH in OrcaSlicer than any other filament category. This guide pins every temperature, drying spec, and dissolution warning to the vendor datasheet it came from. PVA and BVOH are the materials where vendor specs differ the most, and a single "typical PVA range" recommendation will get someone's nozzle clogged. So we're going per-brand on every number. Table of contents When soluble supports earn their keep PVA vs BVOH chemistry and price OrcaSlicer's PVA and BVOH profiles Print temps, bed, and speeds per brand Drying specs and the AMS HT advantage Configuring OrcaSlicer's support settings Printer, AMS, and MMU compatibility Water-bath dissolution: temperature, time, agitation Common failure modes FAQ When soluble supports earn their keep I don't reach for PVA on every print. Tree supports in OrcaSlicer are good enough now that for 80% of overhangs I just toggle them on and accept a 30-second cleanup. But there are jobs where rigid supports physically cannot work: deep sealed cavities, lattice interiors, undercut hooks that wrap back on themselves, and any downward face where a witness mark would ruin the part. That's where soluble shines. You print the model in your usual material, print the interface layer (and optionally the body) in PVA or BVOH, then drop the whole thing in warm water and walk away. The catch is hardware. You can't do this on a single-extruder printer. You need an IDEX rig like the Snapmaker J1, a tool-changer like the Prusa XL, an MMU like the MMU2S or MMU3, or a Bambu AMS. The slicer needs to assign the soluble material to the support interface layers only, because PVA costs five to ten times what PLA does and you do not want a 200-gram support tower made of the expensive stuff. If you're still picking your approach, my tree supports guide and the normal vs tree comparison are the cheaper starting point. OrcaSlicer Support tab with support_interface_filament dropdown showing Generic PVA @System selected PVA vs BVOH chemistry and price PVA is polyvinyl alcohol, a water-soluble synthetic polymer with the idealized formula [CH2CH(OH)]n. Unlike most vinyl polymers, it's prepared by hydrolysis of polyvinyl acetate rather than direct monomer polymerization. That hydrolysis route matters because every spool of PVA you buy is a slightly different blend, which is why eSUN, Polymaker, and Bambu publish wildly different print temperatures for material that nominally has the same chemistry. BVOH is butenediol vinyl alcohol co-polymer. Verbatim's product page lists density at 1.14 g/cm3 and flags the spool as light and UV sensitive, which lines up with my experience: BVOH stored in clear bags by a window goes brittle faster than PVA does. Prusa's knowledge base is the cleanest comparison I've found. Quoting their water-soluble materials article directly: "The main difference between BVOH and PVA is in their printability and price. PVA is usually cheaper, but it's more prone to nozzle clogging" and "BVOH sticks a little bit better than PVA." They go further elsewhere: "PVA+ and BVOH are similar in many ways, but BVOH is absolutely the better (but more expensive) option. It is stiffer and will behave much better than PVA+ when printed." I don't love giving people a multiplier on how much faster BVOH dissolves, because no vendor publishes one. From running both, BVOH strings less, clogs less, and dissolves faster in the same water bath, but the spool is roughly double the price. If you print soluble once a month, get cheap eSUN PVA, dry it religiously, and accept the stringing. If you're printing weekly or running PETG models that push 240C, the BVOH premium pays for itself. Both materials are water-soluble, not biodegradable. PVA is sometimes mislabelled as "compostable" in marketing copy. It's not. It dissolves in water and that's the disposal route. OrcaSlicer's PVA and BVOH profiles Honest take: OrcaSlicer's out-of-box soluble coverage is thin. There are exactly two soluble JSONs in the OrcaFilamentLibrary, Generic PVA @System.json and Generic BVOH @System.json, and both are stubs. The PVA file declares "inherits": "fdm_filament_pva" with setting ID OGFSA04; the BVOH file declares "inherits": "fdm_filament_bvoh" with setting ID OGFSS97_00. The actual default temperatures live in compiled C++ inside PrintConfig.cpp, not the JSON, so you can't open the file and read what nozzle temp Generic PVA will use. Better to ignore the generic preset entirely and use the vendor datasheet for whatever spool you bought. There are no vendor-specific Polymaker, eSUN, or Verbatim soluble presets in the OrcaFilamentLibrary at the moment. Polymaker maintains its own profile import guide on their wiki. For everyone else the workflow is the same as any custom filament: copy the Generic preset, rename it, plug in vendor numbers from the table below, then flip the filament_soluble flag to true in Advanced mode. Once that flag is on, OrcaSlicer treats the filament as a support material in the support-tab dropdowns. My filament settings walkthrough covers the rest of the per-filament tuning that applies to PVA and BVOH just like it does PLA. Print temps, bed, and speeds per brand This is the table I wish I'd had on day one. Every number is from the vendor's own datasheet or product page, verbatim. Don't average these. eSUN's upper limit (230C) overlaps Bambu's lower limit (240C), and a 215C setting that works for PolyDissolve will under-extrude eSUN at the same throughput. SpoolNozzleBedFanSpeed eSUN PVA (TDS V4.0)180-230°C45-60°C100%20-50 mm/s Verbatim BVOH (product page)210±10°C60°Cn/an/a Polymaker PolyDissolve S1 (TDS V5.1)215-225°C25-60°CON30-40 mm/s Bambu Lab PVA (Bambu wiki)240-250°Cvariesn/an/a A few things worth calling out. The eSUN PVA TDS V4.0 recommends "set the distance between the support and the model to 0, slow down and turn off the fan for the first layer to bond well with the main material." That maps onto OrcaSlicer's support Z-distance set to 0 and a slowed first layer. eSUN also rates density at 1.25 g/cm3, heavier than BVOH and PolyDissolve, which matters when estimating spool consumption. The Polymaker PolyDissolve S1 TDS V5.1 gives the tightest spec: 215-225C nozzle, 25-60C bed, 30-40 mm/s, retraction at 1mm/20mm/s. Polymaker also recommends running the spool out of a PolyBox or sealed bag during printing, because equilibrium water absorption hits 10.21% at 70%RH / 23C. Of the three datasheets, the Polymaker one is the most usable starting point if you're cloning a profile. Bambu's own PVA spool runs hot. Their wiki recommends 240-250C, well into PETG territory. That's deliberate; their formulation is tuned for X1C/P1S hot ends and pairs with PETG model materials where eSUN PVA would risk softening from radiated heat. If you're running a Bambu printer with a Bambu PVA spool, use Bambu's numbers from the Bambu PVA printing guide, not the generic preset. Comparison table of PVA vs BVOH vs PolyDissolve S1 print temperatures and drying times Drying specs and the AMS HT advantage If you read nothing else in this article, read this section. Wet PVA does not just print badly. It actively damages your nozzle, because the moisture flashes to steam inside the melt z [truncated] --- ## OrcaSlicer Glow and Sparkle PLA: Print Speed and Nozzle Wear URL: https://orcaslicer.net/orcaslicer-glow-sparkle-pla/ The first time I printed glow-in-the-dark PLA, I left a freshly sliced dragon on my desk under a window for an afternoon, killed the lights at midnight, and watched the thing glow lime green like a movie prop. I've since run glow and sparkle PLA on three printers, swapped out two brass nozzles I should've replaced earlier, and learned the hard way that OrcaSlicer doesn't have a magic "glow mode" button. What it does have is the right vendor profiles bundled, plus enough manual settings to get a clean print without melting your nozzle. This piece walks through what's inside these filaments, which OrcaSlicer profiles ship out of the box, the print settings vendors recommend, and the nozzle wear math you need before committing a kilo to your AMS. Table of contents What's actually inside glow and sparkle PLA OrcaSlicer's built-in profile coverage Recommended print settings for glow PLA Recommended print settings for sparkle PLA The nozzle wear warning you can't skip Charging and glow duration in practice Common gotchas FAQ What's actually inside glow and sparkle PLA Glow-in-the-dark PLA is regular polylactic acid blended with a phosphorescent pigment. The pigment of choice across pretty much every modern spool is strontium aluminate (SrAl2O4), usually doped with europium and dysprosium. Wikipedia's strontium aluminate page documents the chemistry, noting "SrAl2O4 is the primary composition" with europium-dysprosium dopants giving "peak emission wavelengths are 520 nm." That 520nm number is why the glow looks lime green rather than the sickly yellow zinc sulfide produced in older toys. The same article notes strontium aluminate provides "phosphorescence approximately 10 times longer" than zinc sulfide, which is why a charged print can stay visible for hours instead of minutes. It's also non-toxic and non-radioactive. I mention that because every time I post a glow print on Reddit somebody asks about radium. There's none in any of this. Wikipedia describes strontium aluminate as "non-toxic" and "biologically and chemically inert." The catch is hardness. Chaostrade's pigment FAQ measures it at "relatively high hardness 7,6 in Moh's" scale, close to quartz. Push hot polymer plus quartz-grade ceramic powder through a brass nozzle for a few hundred grams and you'll feel the consequences. Sparkle PLA is a different animal that gets lumped in because the visual hook is similar. Instead of a phosphor, you've got reflective particles suspended in the polymer. Bambu's product copy for PLA Sparkle says it has "metal particles included in its composition" to give that glittery look, which means Bambu's sparkle behaves like a mildly abrasive metal-loaded filament. Proto-Pasta's Stardust HTPLA is the counterexample. Their product page explicitly states "No abrasive fillers so expect normal wear with standard nozzles" and "little risk of clogging or wearing your nozzle." Same visual family, very different nozzle math. Don't assume "sparkle" means "abrasive" until you read the vendor sheet. Close-up of a glow-in-the-dark PLA print under daylight next to the same print glowing green in a dark room OrcaSlicer's built-in profile coverage Here's the part that surprised me when I first went looking: OrcaSlicer doesn't have a separate glow plugin or feature. There's no toggle, no special mode. What it has is filament profiles, and the Bambu profile pack that ships with the slicer already includes the ones you need. If you crack open BBL.json on the main branch, you'll find these entries in the filament list: Bambu PLA Glow @base Bambu PLA Sparkle @base Panchroma PLA Glow @base That third one is interesting because Panchroma is Polymaker's brand, not Bambu's, and yet the Polymaker glow profile got bundled into the Bambu vendor pack. So if you're running a Bambu printer profile in Orca, you can pick any of those three from the filament dropdown without downloading anything extra. For other Polymaker spools, their official presets site lists OrcaSlicer as a supported target you can download from directly. If you're running a brand that isn't in the bundle (eSUN, Overture, Sunlu glow), the easiest move is to duplicate the closest Bambu profile, rename it, and tweak temps. OrcaSlicer stores those as `.orca_filament` files (singular, not plural), which I cover in my walkthrough on OrcaSlicer filament settings. For a deeper dive on cloning and exporting, my notes on custom filament profiles cover the file format end to end. OrcaSlicer filament dropdown showing Bambu PLA Glow and Bambu PLA Sparkle profiles selected Recommended print settings for glow PLA The vendor specs are tighter than you'd expect, and they're worth respecting because the abrasive pigment punishes high speed. Polymaker's Panchroma Glow wiki entry lists nozzle temperature at 190 to 230 degrees Celsius, bed at 25 to 60 degrees Celsius, and print speed "up to 200mm/s." That 200mm/s number is the upper bound for the polymer, not a target. I run mine slower for nozzle longevity. Qidi's print lab guide gives the most practical breakdown I've found. They recommend "Nozzle temperature 215 to 230 degrees Celsius, 5 to 10 degrees Celsius higher than standard PLA," "Slower than standard PLA. Faster speeds increase nozzle wear and risk under-extrusion," and a layer height of "0.2 to 0.3mm. Thicker layers contain more particles and glow brighter." They also push 3 to 4 walls minimum, which translates to "1.2 to 1.6mm total wall thickness with a 0.4mm nozzle." The wall count matters because more pigment volume per square millimeter of surface equals more visible glow. Bambu's spec for PLA Glow is similarly strict on geometry. Their store page tells you "Printing with a 0.2 mm nozzle is not recommended; instead, it is advised to use a 0.4 mm nozzle for optimal results." So 0.4mm and bigger is the safe zone. For the baseline PLA reference I use as my starting point before tuning for glow, I've got a separate write-up on OrcaSlicer PLA settings. OrcaSlicer print settings panel showing reduced print speed and 0.28mm layer height for glow PLA My personal recipe: outer wall around 50mm/s, infill around 120mm/s, 0.28mm layer, 4 walls, nozzle at 220 degrees. The bundled Bambu PLA Glow profile gets you most of the way. I just drop the outer wall speed and bump walls to 4 if the model is small enough. Recommended print settings for sparkle PLA Sparkle splits into two camps, and the camp dictates everything. For Bambu PLA Sparkle, the "metal particles" wording on their store page means it's effectively a metal-loaded filament. Treat it as abrasive. I won't quote specific temps or speeds here because the Bambu PLA Sparkle TDS PDF is the authoritative reference and I'd rather you open it than trust a paraphrase. The bundled OrcaSlicer profile reads those values directly. For Proto-Pasta Stardust HTPLA, the picture is different. Their product page lists "190 to 230C nozzle" with "No heated bed required but up to 70C okay." Because they explicitly use non-abrasive glitter, you can run it through a brass nozzle without worrying about bore wear. One Proto-Pasta tip I lifted directly: "Glitter particles will lay flat with smaller layers, giving top/bottom surfaces more shine than sidewall." That means if you want the sparkle to pop, run thinner top and bottom layers, not thicker ones. The flake orientation matters more than the volume of glitter you push through. The nozzle wear warning you can't skip This is the section I wish I'd read before I burned through my first brass 0.4. OzFDM's knowledge base entry on glow filament puts a real number on it: "A standard brass nozzle will show measurable wear after printing as little as 200 to 400 grams." That's less than half a spool. They follow it with the blunt advice to "always use a hardened steel nozzle for glow-in-the-dark filament." Qidi's lab quantifies the upside. From their tips guide: "The abrasive particles erode the bore from 0.4mm toward 0.6mm or larger," and "Hardened steel is the pract [truncated] --- ## OrcaSlicer Wood-Filled PLA Settings and Nozzle Choice Guide URL: https://orcaslicer.net/orcaslicer-wood-pla/ The first time I ran ColorFabb woodFill through a 0.4mm brass nozzle, I got about ninety minutes of beautifully grainy printing and then a clog that took me two hours and a bent guitar string to clear. I've since burned through a stock brass nozzle, replaced it with hardened steel, and printed wood-filled spools from three different brands across two printers, so I've earned the right to say this loudly: wood-filled PLA is not brown PLA, and treating it like brown PLA inside OrcaSlicer is how you cook a hotend. This guide pulls together the settings I actually use, the nozzle gear that survives the abrasion, and the brand-by-brand gotchas that aren't obvious from the spool label. I'll flag the foaming-PLA trap, which catches more first-time wood-PLA buyers than any other issue, and I'll show you how to start from a Generic PLA profile because OrcaSlicer's wood coverage is still thin. Table of contents What wood-filled PLA actually is What's in your spool: composition by brand OrcaSlicer profile coverage for wood PLA Nozzle size and material: the most important call Nozzle pricing, verbatim from vendors Temperature, speed and retraction in OrcaSlicer Drying wood-PLA and clog prevention Sanding, staining and sealing Common gotchas I've hit personally FAQ What wood-filled PLA actually is A real wood-filled PLA is a composite. You've got a PLA matrix (sometimes blended with PHA), and that matrix carries milled wood fibre. The proportion sits roughly between 11% and 30% by weight depending on the brand, and that fibre is what gives the print the matte fibrous surface, the brown gradient, and the unmistakable smell of a wood shop while it's running. It's also what wrecks soft nozzles. The fibres are physically larger than a clean PLA pellet stream, so they snag inside a small bore, swell when heated, and grind down brass over hours of printing. Here's the trap I want to clear up front: not every spool branded "wood" contains wood. Polymaker is open about this. Their PolyWood and Polymaker Wood PLA are 100% PLA with a foaming additive, density around 0.8 g/cm³, and zero actual wood powder. The Polymaker product page calls PolyWood a "wood mimic filament without actual wood powder, which removes all risks of nozzle clogs." Foaming PLA looks like wood but doesn't sand or stain like wood, and it doesn't require the same nozzle precautions. If your spool's data sheet brags about foaming or low density, you're holding plain PLA dressed in a brown jacket. Confirmed wood-filled options I trust: ColorFabb woodFill at roughly 30% wood by weight, and Bambu Lab PLA Wood which the Bambu wiki describes as PLA mixed with "fine wood powder." Hatchbox Wood PLA is reported at around 11% wood fibre, but the Hatchbox product page returned a 403 when I tried to verify it directly, so treat that number as third-party reported rather than vendor confirmed. Wood PLA print next to a real wood block What's in your spool: composition by brand I keep a sticky note above the printer with this table because I've embarrassed myself in front of fellow makers by mixing up which spool is which. The composition genuinely changes how you slice, dry and finish the part, so it's worth committing to memory. ColorFabb woodFill: PLA/PHA matrix plus fine pinewood fibres, roughly 70/30 PLA to wood. Real fibre. Will sand and stain like wood. Bambu Lab PLA Wood: PLA with fine wood powder. Real fibre. Bambu's own wiki warns the powder can clog a 0.2mm nozzle. Hatchbox Wood PLA: reported as around 11% recycled wood fibre. I haven't run it personally and the vendor page wasn't reachable, so flag as third-party reported. Polymaker PolyWood: 100% PLA, foaming, density 0.8 g/cm³. Zero wood powder per Polymaker's own copy. Polymaker Wood PLA: also 100% PLA with foaming technology. Don't confuse the name for real wood content. eSUN: this one is messy. Their older line was a foaming PLA at 0.7 g/cm³ marketed for wood look. Their newer "PLA-Wood" product copy mentions "natural plant fibers." Don't generalise. Read the specific SKU's TDS before assuming. The practical takeaway: if you're holding a real wood-filled spool, you need a bigger and tougher nozzle, slightly different settings, and a drying routine. If you're holding a foaming pretender, you can run it like normal PLA but you lose the sand-and-stain magic. OrcaSlicer profile coverage for wood PLA I'll be honest with you about the state of profiles inside OrcaSlicer, because the official wood-PLA support is thinner than people expect. When I went looking for a generic Wood PLA filament JSON in the top-level resources/profiles/Generic/filament/ folder of the repo, I didn't see one in the rendered listing. There are vendor folders that may carry brand-specific profiles, but the generic Wood PLA slot isn't sitting there waiting for you. The community has been asking for this. GitHub Issue #7527 is an open request to "incorporate newly released Bambu filament types, specifically Bambu TPU For AMS and wood fill types," explicitly to "maintain compatibility with Bambuslicer profiles." A second ticket, Issue #8147 opened in January 2025, asks the maintainers to "add bambu pla wood" and was tagged as an enhancement. So you're not crazy if you can't find a built-in Wood PLA option. It's a known gap. What I do in practice: clone Generic PLA, rename the clone to something obvious like "Wood PLA (woodFill)" or "Bambu PLA Wood," and then tweak the temperature, retraction and speed against the brand's data sheet. OrcaSlicer saves user filament profiles with a .orca_filament extension (singular), which is the file you'll want to back up so a profile reinstall doesn't nuke your custom settings. If you've never built one of these from scratch, I've written a separate walkthrough on how to create a custom .orca_filament profile that goes step by step through the dialog. For the printer side, I keep my base machine profile untouched and store the wood-specific tweaks at the filament level only. That way switching back to plain PLA is one dropdown click and I don't carry low retraction settings into a non-abrasive print where they'd cause stringing. Nozzle size and material: the most important call If you take one thing from this article, take this. The single biggest predictor of whether your wood-PLA print finishes cleanly is the nozzle you put in front of the filament. Get this wrong and the rest of your settings don't matter, because you'll be unclogging at hour two. Start with size. ColorFabb's own how-to-print-with-woodFill blog states "0.4 mm nozzle or larger" as the floor, with the practical recommendation being to go bigger. I run 0.6mm hardened for almost all my wood-PLA work and it's been the right call. Bambu Lab's wiki is even more explicit: 0.4mm, 0.6mm and 0.8mm are all supported on their hotends, but the 0.2mm nozzle is specifically not recommended because "PLA Wood is made of high-quality PLA mixed with fine wood powder, which can clog the 0.2 mm nozzle." That's the manufacturer telling you straight out that small bores and wood powder don't mix. Now material. Brass is fine for a single test print so you can decide whether you even like the filament, but it's a wear part with wood-PLA. The fibres are abrasive, the bore widens with use, and once that bore starts to round out you'll see your extrusion width drift and your top surfaces get fuzzy. I got maybe four or five medium-size prints out of my last brass nozzle before retiring it. I'm not going to give you a hard "X hours to failure" number because I haven't seen credible data on wood-PLA brass wear, and the well-known CNC Kitchen nozzle wear study actually tested carbon-fibre PETG and glow-in-the-dark PLA, not wood. Anyone quoting a specific wood-PLA hours-to-failure figure is guessing. The materials that actually last: Hardened steel: the affordable durable choice and what I default to. ColorFabb's how-to specifically recommends "hardened steel or ruby nozzle" for woodFill. DLC-coated tool steel: E3 [truncated] --- ## OrcaSlicer Silk PLA Settings for That Glossy, Satin Finish URL: https://orcaslicer.net/orcaslicer-silk-pla/ I'll never forget the first silk PLA print I pulled off the bed and held up to a window expecting fireworks. What I got was a dull, slightly waxy benchy that looked like standard PLA wearing a cheap rain jacket. The spool wasn't bad. My settings were. I'd loaded a no-name silk roll, kept my standard PLA profile, and cranked outer-wall speed because "PLA is fast." That print taught me more about silk than any wiki could, and most of what I learned eventually mapped back to a handful of OrcaSlicer fields I'd never bothered to read. Since then I've put several kilos of silk through three printers, mostly Polymaker Panchroma Silk and Eryone Silk plus a few mystery rolls. OrcaSlicer ships a usable starting point for silk, and the rules are fairly consistent across brands. But vendor TDSs disagree with each other and sometimes with OrcaSlicer's bundled profile, so you can't grab one number and run with it. This guide walks through what's in the slicer, what the vendors say, where they fight each other, and how I'd dial in a silk profile today. Glossy silk PLA print catching directional light next to a matte standard PLA print of the same model Table of contents What silk PLA actually is (and isn't) What OrcaSlicer actually ships for silk Speeds: 40 to 60 mm/s, with caveats The fan-speed argument nobody wins Layer height: the 0.12 to 0.20 mm range The glossy-bottom trick (community wisdom) Multi-color silk in OrcaSlicer The Eryone profile quirk you should know Gotchas: stringing, brittleness, clogging A starter profile you can actually use FAQ What silk PLA actually is (and isn't) Let's clear something up before we touch a slicer field. There is no silk in silk PLA. The "silk" part is marketing for the satin sheen on the finished print. Spectrum's official PLA Silk technical data sheet calls it "a PLA-based material for additive manufacturing with colour concentrates which imparts the satin texture effect to printed items," and Polymaker, when rebranding their silk line as Panchroma in 2024, described it as "a new family of engineered polymer compounds containing PLA." So we're talking about a PLA backbone with additives mixed in. What are those additives? Honestly, the vendors don't tell you. Material-science write-ups speculate about fatty-acid ester lubricants combined with pigments, and some second-tier blogs claim there's a small amount of TPE or TPU mixed in. I've seen that last one repeated everywhere, but I couldn't find a single vendor TDS that confirms it, so I'm going to leave it alone. The honest framing is: PLA plus proprietary additives that change how the surface reflects light and how the polymer chains slide past each other in the melt zone. This matters because silk doesn't behave like standard PLA. It's typically run hotter (most TDSs sit between 200 and 230°C versus 190 to 210°C for plain PLA), the surface is more sensitive to layer time, and mechanical properties drop a bit. Spectrum's TDS records a notched Izod impact strength of 6 J/m, on the lower side for PLA, and Polymaker's PolyLite PLA Silk V5.1 TDS lists Charpy impact at 3.46 kJ/m². Silk PLA is a finish material first, a strength material a distant second. Treat it like decorative trim, not structural parts. What OrcaSlicer actually ships for silk If you go digging in OrcaSlicer's source tree, the silk story is more interesting than you'd expect. There isn't a single monolithic "silk" profile. There's a small handful of silk JSON files split across the Eryone vendor folder and the consolidated OrcaFilamentLibrary folder that the project introduced in 2024. Here's what's actually in there at the time of writing: Eryone Silk PLA. Lives at resources/profiles/Eryone/filament/Eryone Silk PLA.json. There's also a 0.2 nozzle variant in the same folder. Generic PLA Silk @System. Lives in OrcaFilamentLibrary/filament/. This is what you'll select for unbranded silk rolls. fdm_filament_pla_silk. The base silk profile in OrcaFilamentLibrary/filament/base/. Every silk variant ultimately inherits from this. Polymaker Panchroma PLA Silk @System and @base. In OrcaFilamentLibrary/filament/Polymaker/. PolyLite Dual PLA @System. The multi-color spool's profile, which doesn't actually apply silk-specific settings. We'll get into that. You won't find a dedicated eSUN eSilk profile in there. eSUN's folder inside OrcaFilamentLibrary has PLA+, PLA-Basic, PLA-Marble, PLA-Matte, and ePLA-LW, but no silk JSON. If you're running an eSUN silk roll, your cleanest move is to load Generic PLA Silk and override the nozzle temp. I do that all the time and it's fine. Now for the interesting bit. Open the base silk file, fdm_filament_pla_silk.json, and you'll see it's almost empty: { "type": "filament", "name": "fdm_filament_pla_silk", "inherits": "fdm_filament_pla", "from": "system", "filament_id": "OGFL96", "instantiation": "false", "description": "To make the prints get higher gloss, please dry the filament before use, and set the outer wall speed to be 40 to 60 mm/s when slicing.", "filament_flow_ratio": ["0.98"], "slow_down_layer_time": ["8"] } That's it. Inherits from the standard PLA base, drops flow ratio to 0.98, bumps slow_down_layer_time from 4 to 8 seconds, and tucks the dry-and-slow advice into the description field. Everything else (nozzle 220°C, bed 55°C, fan_min_speed 100, close_fan_the_first_x_layers 1) comes from the standard PLA base unchanged. OrcaSlicer's silk profile is really "standard PLA with longer cooling time per layer plus a friendly note." OrcaSlicer Filament tab showing the Generic PLA Silk profile selected with the Cooling sub-tab open If profile inheritance feels a bit mysterious, our walkthrough on OrcaSlicer filament settings covers how the @base, @System, and vendor JSON files stack up. Worth a read if you plan to make your own. There's one quirk worth flagging right away. Polymaker's Panchroma PLA Silk @base inherits from fdm_filament_pla, not from fdm_filament_pla_silk. So while the Generic PLA Silk profile gets the 8-second layer time, Panchroma Silk inherits the standard PLA 4-second layer time. Two profiles, both technically "silk," both shipping with OrcaSlicer, with different cooling behavior. Not a bug exactly, but it's the kind of thing that bites you if you assume "silk" means the same setup across brands. Speeds: 40 to 60 mm/s, with caveats The most quotable number is hiding in plain sight: OrcaSlicer's base silk JSON description tells you to "set the outer wall speed to be 40 to 60 mm/s when slicing." That's the bundled recommendation, written by the profile maintainers. If you do nothing else from this guide, do that. The problem is that if you go cross-check against vendor TDSs, you get a circus. Here's what the major silk brands tell you: OrcaSlicer base silk profile. 40 to 60 mm/s for the outer wall. Polymaker PolyLite PLA Silk TDS (V5.1). Printing speed 30 to 70 mm/s. Polymaker Panchroma Silk wiki. Claims usable up to 250 mm/s and "maintains rich, silky shine even at 150 mm/s." Eryone Silk PLA product page. 30 to 60 mm/s (see the Eryone Silk PLA listing). eSUN eSilk PLA. Less than 200 mm/s. Spectrum PLA Silk TDS. 40 to 110 mm/s. That's a 30 to 250 mm/s spread. The reality after enough botched prints to fill a parts bin: ignore the upper end of vendor claims when finish matters. Polymaker's 150 mm/s number is achievable on a stiff CoreXY, but the sheen falls off well before you get there. I've never beaten OrcaSlicer's 40 to 60 mm/s outer-wall window for actual gloss, even on a well-tuned Bambu X1C. Push inner walls and infill to 150 to 200 mm/s if your printer can handle the volumetric flow. Outer wall is what your eye reads as "silky," so don't gamble there. A Numakers community guide is stricter, recommending 30 to 45 mm/s and "no more than 50 mm/s" for excellent visual results. For day-to-day printing, the 40 to 60 mm/s OrcaSlicer window is the right balance. The fan-speed argument nobody wins This is the section that took me longest to write because I don' [truncated] --- ## How to Add a Custom Filament Profile to OrcaSlicer (2026 Step-by-Step) URL: https://orcaslicer.net/orcaslicer-custom-filament-profile/ Last week I bought a no-name silk PLA spool off Amazon for nine bucks, the kind with a hand-applied paper sticker and zero recommended print settings. I opened OrcaSlicer, stared at the filament dropdown, and realized I needed to either clone a profile, import one from a stranger on Printables, or build a fresh one from scratch. This guide walks through all three options the way I actually do them at my own bench, with the menu paths, the file extensions, and the gotchas that have bitten me on real prints. I'm going to be honest up front: cloning is faster than starting from scratch, and importing a community profile is faster than cloning, but none of those three options skip calibration. If you want a profile that prints clean parts at the speed your machine can actually handle, you'll run a temp tower, a flow rate test, and a pressure advance test no matter which path you pick. The click sequence is the easy part. The tuning is the work. For the big-picture context on how filament settings fit together, the complete OrcaSlicer filament settings guide is the pillar I keep coming back to, and this article is the practical "how do I get a new profile into the dropdown" companion. Table of contents The two ways to add a custom filament profile Path A: clone an existing profile in seven clicks Path B: import a community .orca_filament profile Settings every custom profile needs to set The calibration sequence that turns a clone into a profile Clone a brand profile or clone Generic? A decision rule Where OrcaSlicer stores your profile on disk Sharing your custom profile with Export Preset Bundle Keep your profiles tidy Common pitfalls and how to avoid each Frequently asked questions Where to go next The two ways to add a custom filament profile OrcaSlicer gives you two real paths to a custom filament profile, plus one half-supported drag-and-drop trick I'll mention in passing because it works for 3MFs only. Path A clones a profile that's already in your slicer. You pick the closest match in the dropdown, click the pencil icon to open Filament Settings, change the values you know are different, then save it as a new user preset. This is what I do for any spool I've physically unboxed myself, because I'd rather start from a tuned profile and adjust two numbers than start from Generic and adjust twelve. Path B imports a community profile that someone else already built. You download a .orca_filament file (or a .zip of .json files, or a full .orca_printer bundle), go to File > Import > Import Configs, and load it. This is faster if a trustworthy source has already done the tuning for your exact spool. It's slower in the long run if the profile they uploaded was tuned on a different machine, which most are. Here's the side-by-side I use when I'm deciding which one to pick. Decision factorPath A: clone existingPath B: import community Time to first print5 minutes2 minutes Best forAny new spool, branded or no-nameExact spool match on Printables / GitHub Calibration still requiredYesYes, every time Reliability of starting valuesHigh if cloning a brand profile, medium if cloning GenericVaries wildly with uploader Inheritance preservedYes, clone inherits from parentDepends on bundle File extension involvedNone, all in-app.orca_filament, .orca_printer, or .zip Before either path makes sense, you need to know where you're starting from. Here's my Prepare view with a Bambu Lab P1S, the 0.4 mm nozzle, Textured PEI plate, Bambu ABS in the filament slot, and a process preset I'd already cloned and renamed. My current setup before adding a custom filament profile. The Filament dropdown on the right is where Path A starts. The process preset is already a clone, which is why it has the "- Copy" suffix. Path A: clone an existing profile in seven clicks This is the path I recommend for almost everyone, especially anyone with a single no-name spool they want to print today. Cloning preserves the inheritance chain, which means any value you don't change is still tied to the parent profile. That's useful because Orca's defaults aren't random, they're the result of a lot of work the maintainers and the community have already done. Step 1: pick the closest base profile Open OrcaSlicer, stay on the Prepare tab, and click the Filament dropdown on the right panel. Don't pick "Generic PLA" reflexively. Scroll the list. If your spool is Polymaker PolyLite PLA, pick that. If it's Overture PETG, pick that. If you have a cheap unbranded silk PLA that smells suspiciously like Sunlu, pick the Sunlu silk variant. The closer your base profile is to the spool you actually own, the fewer numbers you'll have to change in step 3. If the dropdown is overwhelming, I sort by clicking through System Filaments and using the filament-type and vendor filters. Here's that dialog with both filters visible. The filament-type and vendor filters on the Filaments dialog. I use vendor first, then narrow by type. This is also where you'd come back later to tick "Compatible Printers" if a custom profile doesn't show up in your Prepare dropdown. Step 2: open Filament Settings via the pencil icon Once your closest base profile is selected in the dropdown, look for the small pencil icon right next to the filament name. Clicking it opens the Filament Settings dialog. This is where every editable property of the profile lives, organized into Filament, Cooling, Setting Overrides, Multimaterial, Notes, and a few other tabs depending on your version. Quick note on what not to click: there's a separate save button that just writes back to the current profile. If you hit that while the parent is a system preset, it'll either silently fail or behave unpredictably depending on your Orca version. We're going to use a different save flow in step 4. Step 3: edit your starting values Don't try to change everything at once. The values worth editing before your first test print are: Filament diameter. Measure your spool with calipers in three spots. If you get 1.71 to 1.76, use the average. Don't assume 1.75. Nozzle temperature, first layer and other layers. Pull from the spool sticker if there is one, otherwise use the middle of the type's range (210 C for PLA, 235 C for PETG, 250 C for ABS). Bed temperature. Same logic. Flow ratio. Leave at 1.0 for now, you'll tune this with the calibration sequence later. Max volumetric speed. Drop it to a conservative number (10 mm3/s for PLA, 8 for PETG, 6 for TPU) until you've run the MVS test. Everything else (pressure advance, retraction tweaks, cooling curve) gets dialed in during calibration. Don't waste an hour adjusting numbers you'll overwrite tomorrow. Step 4: Save as new user preset (the floppy disk dropdown) At the top of the Filament Settings dialog, there's a floppy disk icon. Click it. The important option is Save as new user preset. Pick that, not the plain save. This explicit "Save as" creates a new file in your user folder and points its inherits key at the parent profile you cloned from, which is exactly what you want. If you skip "Save as new user preset" and just hit the regular save, behavior depends on whether the parent is a system or user preset. For a system parent, Orca should refuse the write but some builds have been reported to silently lose the edit. Issue #4995 in the OrcaSlicer repo is a long-running thread about exactly this confusion. Use the explicit save flow and you avoid the whole class of problems. Step 5: name your profile cleanly I use the pattern [Vendor]-[Material]-[Color]-[Variant] @[Printer]-[Nozzle]. Example: Sunlu-PLAplus-Black @P1S-0.4. For true no-names where there's no vendor, I drop in the Amazon order date as a pseudo-vendor: Amazon20260308-Silk-Red @Ender3V2-0.4. It looks ugly, but a month later when I'm staring at thirty user profiles in a dropdown, I can tell exactly which roll the profile belongs to. Avoid special characters (slashes, colons, asterisks). Some Orca versions choke on them when writing the JSON filename. Step 6: [truncated] --- ## OrcaSlicer Carbon-Fiber Filament Settings: PETG-CF, PA-CF, PLA-CF (2026 Guide) URL: https://orcaslicer.net/orcaslicer-carbon-fiber-filaments/ I bought my first spool of PLA-CF expecting it to make my drone arms unbreakable. They snapped on the first crash, cleaner and faster than the cheap PLA+ arms they replaced. That was the moment I learned what "carbon fiber" actually means on a 1 kg spool, and what it absolutely does not mean. It's a lesson the marketing doesn't go out of its way to teach you. I've since printed PLA-CF, PETG-CF, PA-CF, and PC-CF across a Bambu X1C, a P1S, a Voron 2.4, and one stubborn old Ender 3 that should never have been asked. Every blend has its own gotchas, its own hardware tax, and its own honest answer to the question "is this actually worth it." This guide gives you the per-blend OrcaSlicer settings, the real cost of the hardware you need before you press print, and the truth about where CF wins and where plain nylon quietly beats it. Table of contents CF reality check (read this before you buy a spool) What CF blends actually are The four CF blends at a glance Hardware you need before printing CF Hardened nozzle costs, ranked honestly OrcaSlicer's built-in CF profiles PLA-CF settings in OrcaSlicer PETG-CF settings in OrcaSlicer PA-CF settings (the big one) PC-CF settings (boss tier) CF blends vs continuous carbon fiber composites Aesthetic vs structural, the honest section Common CF print failures and fixes Calibration order for any new CF spool FAQ CF reality check (read this before you buy a spool) If you bought carbon fiber filament because the name sounds strong, you bought it for the wrong reason. CF blends are mostly about stiffness, dimensional stability, and that matte graphite aesthetic. Real impact resistance still belongs to plain nylon. Real strength-per-dollar usually belongs to a good PLA+ blend. I'm going to repeat that several times in this guide because it's the single thing that resets a buyer's expectations correctly. Here's the short version of what's coming. Every CF filament wears brass nozzles down in roughly 3 to 5 hours of printing, so a hardened steel nozzle is non-negotiable before you press print. PLA-CF tunes like PLA with a wider fan and a slower max volumetric speed. PETG-CF wants less cooling than its plain counterpart. PA-CF and PC-CF demand a heated chamber, aggressive drying, and patience. And the speckled finish you're going to see on every print isn't a defect. It's the carbon fiber. This guide sits under the OrcaSlicer filament settings pillar, so if you came in cold and need the general filament workflow first, start there. Otherwise, let's get into what those chopped fibers actually do once they're in the melt. What CF blends actually are Carbon-fiber filament is not a carbon-fiber composite in the way the aerospace industry uses that phrase. The spool you buy is a base thermoplastic (PLA, PETG, PA, or PC) that's been compounded with roughly 5 to 20 percent chopped carbon-fiber strands. The fibers are short, typically 100 to 300 microns long, and they're randomly oriented inside the polymer matrix. They don't form a continuous reinforcement network. They sit in the plastic like very small rebar chunks suspended in concrete, except they're not aligned and they don't span the full length of the part. Bambu PLA-CF MSDS composition. The actual carbon fiber content is 5 to 10 percent. The other 90 to 95 percent is plain PLA. When you buy "carbon fiber" filament, this is what you're actually buying. The MSDS above is the honest answer to "how much carbon fiber is in carbon fiber filament." For Bambu's PLA-CF the answer is 5 to 10 percent. That's not unusual. Most hobby CF blends sit in the 10 to 15 percent range. Engineering-grade PA-CF can hit 20 percent, occasionally higher. The rest is the base polymer, which is going to dominate the mechanical behavior of the finished part more than the fibers do. What the fibers buy you: Higher stiffness, also called modulus. A PA-CF tensile bar resists bending more than a plain PA bar at the same wall count. Better dimensional stability. The fibers reduce shrinkage during cooling, which is why nylon CF blends warp less than plain nylon. A dark matte aesthetic. The chopped fibers scatter light and give that gritty graphite finish that's honestly half the reason people buy CF in the first place. Slightly higher continuous-use temperature for the polyamide and polycarbonate variants. What they don't buy you: Higher impact strength. In most lab tests, CF lowers impact resistance compared to the base polymer. Plain nylon takes a hammer better than PA-CF. Plain PLA bends further before snapping than PLA-CF. Strength in the Formula 1 sense. People hear "carbon fiber" and picture a monocoque chassis. Chopped short fibers in FDM filament add stiffness, not the tensile strength a continuous-weave laminate provides. Magic durability. PLA-CF is still PLA underneath. It still creeps under load and softens around 55 to 60 C. The cost premium is real. Plain PLA runs around 18 to 25 USD per kg. PLA-CF runs roughly 25 to 35 USD per kg. PETG-CF jumps to 45 to 65 USD per kg. PA-CF lands between 60 and 100 USD per kg for hobby brands, and 120 to 200 USD per kg for engineering grades like Polymaker Fiberon, Bambu PAHT-CF, or 3DXTech CarbonX. (Prices flagged VERIFY at publish time.) The four CF blends at a glance Blend Base polymer Difficulty Enclosure needed Best for PLA-CF PLA Easy No Aesthetic prints, jigs, stiff cosmetic parts PETG-CF PETG Medium Helpful, not required Outdoor parts, chemical resistance PA-CF Nylon (PA6 or PA12) Hard Yes Drone frames, gripper jaws, brackets that see load PC-CF Polycarbonate Hardest Yes, actively heated High-temp, high-load engineering applications PLA-CF is the gateway. If you've never run a CF filament, this is where you start. It tunes like PLA, prints with the part cooling fan blasting, and doesn't need a sealed chamber. It's the CF blend an A1 Mini or a Prusa MK4 can run without complaint, assuming you've swapped the brass nozzle out. PETG-CF is a marginal upgrade over plain PETG. It's harder, occasionally less stringy, and a bit more rigid. The trade-off: it loses the "bend before break" character that makes plain PETG forgiving. I've snapped PETG-CF brackets that plain PETG would have flexed through. PA-CF is where carbon fiber starts to pull its weight. Nylon CF blends are what people mean when they say "engineering grade." This is the material for drone frames, gripper jaws, jigs, and brackets that see real mechanical load over time. It's also the material that turns most non-enclosed printers into expensive paperweights. PC-CF is the top tier for amateurs. Brutal print temperatures, mandatory chamber heating, and the filament dries faster than it prints if you leave it out overnight. Save it for parts that actually need a 110 C-plus continuous service temperature. For most people, that's never. Hardware you need before printing CF There are four hardware rules that apply to every CF blend on this page. No exceptions, no "I'll just try it once." 1. Hardened steel nozzle is mandatory. Brass nozzles wear out in roughly 100 to 250 g of CF filament. Visible orifice rounding and length shortening show up in as little as 3 to 5 hours of printing. Hardened tool steel runs orders of magnitude longer than brass on CF blends per community testing. The fibers themselves are harder than brass, that's the whole problem. There's no "I'll be careful" workaround. 2. 0.4 mm minimum, 0.6 mm recommended. Chopped fibers occasionally clump in the melt. A 0.4 mm orifice is the bare minimum and will clog more often than a 0.6 mm setup. Bambu's own filament guide for PA-CF and PA6-CF recommends 0.6 mm hardened steel as the default. For drone frames and load-bearing parts, 0.6 mm is faster anyway because the wider extrusion stiffens the geometry further. 3. Direct drive is strongly preferred. Bowden tubes add friction and lag that work against stiff, slightly abrasive CF filament. Direct-drive extruders give cleaner retractions and fewer feed issues. Bambu X1C and P1S are direct drive. [truncated] --- ## OrcaSlicer Nylon (PA / PA-CF) Settings & Drying Tips (2026 Guide) URL: https://orcaslicer.net/orcaslicer-nylon-pa-settings/ My first serious nylon print failed in the most embarrassing way possible. The nozzle hissed like a tea kettle, the part looked like burnt popcorn, and I sat there staring at a 200 gram brick of stringy PA6 wondering what I had done wrong. I hadn't touched the slicer. I'd just opened a fresh spool and hit print. Twelve hours in a SUNLU E2 at 80°C later, the same model, the same OrcaSlicer profile, came off the plate clean. The settings hadn't changed. The water did. That's the punchline of this entire article, and I want to put it at the top so nobody scrolls past it. Nylon doesn't have a settings problem. It has a water problem. Every number I'm about to give you assumes a dry spool, and if your spool isn't dry, no profile on Earth will save the print. I've burned enough filament learning that lesson that I'll keep saying it until the captions stop being German. Table of contents What nylon actually is (and why the variant matters) OrcaSlicer's bundled PA profiles Drying: the real article OrcaSlicer settings, the actual numbers Bed adhesion (PA does not stick to PEI) PA-CF and PAHT-CF: hardware is not optional Failure modes and what they really mean Calibration order in Orca Buying guide, two honest paragraphs Wrap What nylon actually is (and why the variant matters) Nylon is the consumer name for polyamide, abbreviated PA. It isn't one material, it's a family, and the family members behave differently enough that you can't share a single profile across them. I learned this the hard way when I tried to print Polymaker CoPA with my eSun ePA settings and watched the bed lose adhesion at layer 30. Same word on the spool, completely different chemistry. Here's the cheat sheet I keep taped to the side of my dry box. It covers everything you'll actually see on a filament shelf in 2026, with the headline mechanical numbers that matter when you're deciding which variant to buy. VariantTensile (approx)Water absorptionMelt zoneHonest take PA6 (nylon 6)70 to 80 MPaUp to ~3% by weight220 to 240°CStrongest unfilled, soaks water aggressively, the diva of the family PA66 (nylon 6,6)~80 MPa~2.5%250 to 265°CRare unfilled, almost always sold as PA66-CF or PA66-GF PA12 (nylon 12)~50 MPa~0.5%240 to 270°CNear-immune to water, low warp, the nylon to start with PA612 (nylon 6,12)~55 MPa~1.5%250 to 280°CThe compromise: low warp, higher melt, Polymaker Fiberon PA612-CF lives here CoPA (copolymer)~60 MPa~1.5%250 to 270°CPolymaker's blended PA, marketed as warp-resistant nylon, easy to print PA-CF / PA6-CF90 to 120 MPa~1.5 to 2%260 to 300°CStiff, dimensionally stable, eats brass nozzles in under a kilo PAHT-CF (Bambu)~110 MPa~1% (their claim)280 to 300°CLess hygroscopic than PA-CF, the production engineering choice on X1C A few traits are shared across the whole family, and they're the reason nylon is both rewarding and punishing. Nylon is hygroscopic. PA6 picks up moisture out of the air within hours, and Bambu's own PAHT-CF product page says the selling point is that it absorbs about half as much as PA-CF. Nylon is also self-lubricating, which is why it's the default filament for gears, bushings, hinges, and any sliding part. And when it's printed dry, interlayer bonding (the polymer's "reptation" behavior, if you want to sound clever in a forum) is excellent. Wet PA6 tells a different story, Sovol's 2025 internal data showed wet samples losing up to 83% of stiffness and 42% of strength compared to dried equivalents. The honest framing: unfilled PA12 and CoPA are the easiest entry points. PA66-CF is the hardest. PA-CF and PAHT-CF sit in the middle, hard because of the hardware they demand, not because the slicer settings are exotic. If you're choosing between filament chemistries for an engineering project, you might also want to compare against my OrcaSlicer PC settings guide and the broader OrcaSlicer carbon fiber filaments overview, since the use cases overlap. OrcaSlicer's bundled PA profiles OrcaSlicer 2.x ships with a stack of nylon profiles that cover most of what you'll buy. I've verified the dropdown on the current build, here's what you'll see when you scroll the filament list. Generic profiles include Generic PA, Generic PA-CF, and Generic PA6-CF. These are intentionally conservative on volumetric flow and bed temperature so they boot up on any printer, including open-frame Bowden machines. I treat them as starting points, not finals. Bambu profiles include Bambu PA, Bambu PA-CF, Bambu PAHT-CF, and the newer Bambu PA6-GF. These are tuned for the X1C / P1S running the engineering plate. The defaults are noticeably more aggressive than the generic profiles, especially on MVS. If you're on a Bambu printer, see my dedicated OrcaSlicer profile guide for the X1C and P1S for printer-side gotchas that interact with these filaments. Polymaker profiles include PolyMide PA612-CF (Fiberon), PolyMide CoPA, and PolyMide PA6-CF (Fiberon PA6-CF20). Polymaker's defaults are the closest to "use this as published" of any vendor I've tried, especially the PA612-CF profile. Other vendors bundled include Fiberlogy Nylon PA12, PA12-GF15, and PA12-CF15, plus eSun ePA and ePA-CF, and Overture Easy Nylon. If you're on a less common engineering printer like the QIDI X-Max 3 or Plus4, my OrcaSlicer guide for the QIDI X series covers the chamber-temperature overrides that make a real difference for nylon on those machines. What Orca exposes that the stock Bambu slicer hides is the part that actually matters for PA. You get a fixed pressure advance value per filament (which moves a lot when nylon picks up water), a per-filament max volumetric speed override, per-filament min and max layer time, and a per-filament fan curve. For nylon, the PA-value tunability alone is worth switching slicers, because the right value at 0.3% moisture and the right value at 0.8% moisture are not the same number. OrcaSlicer's Prepare view in German with a bearing assembly loaded. "Aktuelle Platte slicen" = "Slice current plate" and "Projekt" = "Project". The workflow is identical in English, the buttons just sit in the same screen positions regardless of language. Drying: the real article If you only read one section, read this one. I want to be honest about how much of my own nylon trouble traces back to ignoring this exact paragraph. Drywise's published measurements show that PA6 1.75 mm filament going from 1.0% moisture down to 0.5% moisture takes at least 24 hours at 80°C. The same paper says over 55% of neat nylon spools they tested arrived from the factory above the 0.75% threshold where visible defects start. Factory sealed. Vacuum bagged. Above the threshold. That's the reality, and it's why I treat every new spool of PA as wet until proven otherwise. The practical moisture threshold is 0.5% by weight. Above that, you'll see hissing or popping at the nozzle (water flashing to steam), visible bubbles or vapor at the extrusion, stringing despite well-tuned retraction, rough or fuzzy dimpled surfaces, layer delamination where parts snap clean on the Z axis, filament so brittle it breaks in the spool path, and brown discoloration when you bump the nozzle temperature to compensate. If you're seeing more than two of those at the same time, stop, dry, restart. Drying temperatures and times The vendor recommendations have converged pretty tightly. Here's what I cross-checked across Bambu Lab's wiki, Polymaker's product pages, Fiberlogy's spec sheets, ColorFabb, and Drywise. FilamentDrying tempTime (fresh spool)Time (soaked spool)Source Bambu PA / PA-CF / PAHT-CF80°C12 hours24 hoursBambu Lab wiki Polymaker PolyMide PA612-CF80°C oven, 75°C PolyDryer12 hours oven18 hours PolyDryerPolymaker spec Polymaker PolyMide CoPA80°C12 hours24 hoursPolymaker spec Fiberlogy PA1270 to 80°C8 to 12 hours16 to 24 hoursFiberlogy ColorFabb PA-CF70°C4 to 6 hours12 hoursColorFabb (their low-end recommendation) Generic PA6 (any unknown vendor)80°C24 hours24 to 48 hoursDryw [truncated] --- ## OrcaSlicer Polycarbonate (PC) Settings - Strong Functional Parts (2026 Guide) URL: https://orcaslicer.net/orcaslicer-pc-settings/ I lost three prints in a row to PC before I figured out the problem wasn't my settings, it was my filament sitting open on a shelf for two weeks. The fourth one, after a proper 8-hour dry cycle at 80C and a chamber held near 55C, came off the plate clean, matte-black, and tough enough that I couldn't snap it across my knee. That's the polycarbonate experience in a sentence: it punishes laziness and rewards process, and once you've dialed it in inside OrcaSlicer it'll print parts that genuinely outlast anything you've made in PLA or PETG. This guide walks through every setting that matters for PC in OrcaSlicer in 2026, based on the actual values shipped in OrcaSlicer's source profiles (verified against the BBL fdm_filament_pc.json on the main branch) and cross-referenced with vendor data sheets from Bambu, Polymaker, and Prusa. I'll show you what works, what fails, and where the bundled profiles are too optimistic for your machine. I'm assuming you're running OrcaSlicer 2.x on a printer that can actually handle PC, and if you're not sure whether yours qualifies, there's an honest list of suited and not-suited machines below. Table of contents What polycarbonate is and why it earns the hassle Is your printer actually suited for PC? Build plate and bed adhesion Drying PC: not optional What ships in OrcaSlicer for PC Verified defaults from the source profiles Recommended starting settings Slicing and sending a PC print in Orca Calibrating PC in OrcaSlicer PC-CF and PC-GF variants Failure modes and fixes PC vs the alternatives Printer-specific notes Wrap-up What polycarbonate is and why it earns the hassle Polycarbonate is an amorphous thermoplastic built from bisphenol-A (BPA) carbonate ester chains. The amorphous structure is what makes it so impact-resistant: instead of cleaving along crystal planes the way nylon or PLA can, the chains absorb energy and yield. Three numbers matter when you're slicing it. Glass transition (Tg): about 145 to 150C for neat PC. Polymaker quotes 150C for PolyMax PC, 3DXTech quotes 147C for CarbonX PC+CF, and OrcaSlicer's BBL base profile uses a conservative temperature_vitrification of 120C. The Tg matters because it's the rough ceiling for what your part can do before it goes soft. Heat deflection temperature (HDT): roughly 115 to 130C depending on filler and load. Compare that to PLA at about 55C and you can see why people use PC for parts that live near motors, in sunny cars, or anywhere hot air collects. Notched Izod impact: Polymaker publishes 65 kJ/m² for PolyMax PC. Bambu's @base profile records an impact_strength_z of 9.0 kJ/m² for Z. ABS sits around 20 to 25 kJ/m² unnotched and shatters at -20C. PC stays tough even cold. What PC is actually good for: drone arms, RC parts, motor mounts, tool handles, jigs, automotive interior brackets, anything that has to take repeated impact or live somewhere hot. What it's bad for: anything cosmetic (it tends to dull and yellow under UV), anything with critical surface finish, and anything you need fast or cheap. The downside that lives in every PC print: because PC is amorphous, it shrinks more or less linearly across the entire glass transition range as it cools. That continuous shrinkage is exactly why corners pop off the plate, why mid-print layer cracks sound like a gunshot, and why the brim tears the skin off your bed if you're not careful. The cure is to slow that cooling down, which is what the enclosure is for. We'll come back to that. Is your printer actually suited for PC? I want to get this out of the way before anyone wastes a spool. PC needs three things from your printer: a hot end that can sit at 280C for hours without a soft throat, a heated bed that holds 100 to 110C, and a chamber that stays warm enough to keep the print from cooling unevenly. If any of those three is missing, you're going to fail prints. Here's Bambu's own compatibility table for PC across the X1C, P1S, A1, and AMS combinations: Notice that Bambu's nozzle range for their own PC is 260 to 290C, which lines up exactly with the nozzle_temperature_range_low/high values in fdm_filament_pc.json. Notice also that PC is rated for 0.4, 0.6, and 0.8mm nozzles, but with a stainless or hardened insert. Brass works for neat PC, but you do not want to feed PC-CF through a brass nozzle for more than a few hundred grams. Suited machines (honest list) PrinterEnclosureChamber strategyNotes Bambu X1CPassiveReaches 45 to 55C with bed at 110CWorkhorse for hobbyist PC. AMS works. Bambu X1EActive heater60C set point defaultPurpose-built for engineering filaments. Bambu P1SPassiveSimilar to X1C, slightly coolerOK for small/medium parts. Big tall PC is borderline. Bambu H2D, H2S, P2SPassiveSimilar envelope to X1CTurn off aux fans for PC. Qidi X-Plus 3, X-Max 3, Plus 4Active heaterUp to 65CStrongest non-Bambu option for PC. Voron 2.4 / TridentFull panels + heater or Nevermore50 to 70C achievableExcellent if you've already done the chamber work. Prusa CORE OnePassive enclosureReaches 40 to 45CDecent for PC-CF, marginal for neat PC. Creality K1C / K2 PlusPassive40 to 50CWorks for small parts. Stock hot end is rated. Not suited (the honest answer is: don't) If you're on an Ender 3, an Anycubic Kobra 2 or 3, an Elegoo Neptune 4, a stock Bambu A1 or A1 mini, a Prusa MK4S or MK3S+ without the enclosure, or a stock Bambu P1P, you will not get reliable PC results. The bed slingers especially are a no-go: PC needs a still warm chamber, and a Y-axis bed flinging around generates a thermal turbulence that wrecks layer adhesion. The honest recommendation is to use PETG-CF, PETG-HF, or Polymaker's PolyMax PC-FR (which has a much lower shrinkage rate than neat PC) instead. Or, if you've got a P1P, fit the official side panels and you've basically got a P1S, and you can come back to this guide. Hot end and nozzle requirements All-metal hot end, rated for 300C continuous. PTFE-lined throats melt at PC temperatures, and once they melt they wick filament backward and clog. For neat PC, a brass nozzle is fine short-term but a plated copper or hardened steel nozzle is the safer call. For PC-CF and PC-GF (and any reinforced PC), hardened steel, ruby, or tungsten carbide is mandatory. I've watched a brass nozzle round out from a sharp 0.4 to a sloppy 0.45 in under 800g of PC-CF, and the print quality drops well before that. High-flow hot ends (Bambu's CHT clone, E3D Revo HF, Phaetus Rapido HF) help PC because the longer melt zone lets you push more material at a lower nozzle temperature, which cuts down on oozing and stringing. The full Bambu X1C and P1S OrcaSlicer setup guide covers the X1C and P1S thermal envelopes in more detail. Build plate and bed adhesion Bambu's heatbed temperature table makes the bed strategy easy: PC sits at 100 to 120C with a glue stick or PVP-based adhesion aid. The OrcaSlicer BBL base profile sets hot_plate_temp, hot_plate_temp_initial_layer, eng_plate_temp, and textured_plate_temp all to 110C, with cool_plate_temp and supertack_plate_temp disabled (0). That tells you everything you need to know: PC wants the engineering plate or the textured PEI plate, period. Cool plate isn't rated for 110C and Super Tack doesn't survive the temperature either. Bambu Engineering Plate: the go-to. A thin smear of glue stick (Bambu's own, or a generic Elmer's, or Magigoo PC) and PC sticks like it's bolted on. Cools down before you try to release the part. Textured PEI: works without adhesive if it's clean, but Magigoo PC makes the experience much more forgiving on tall or thin-walled parts. Garolite (G10/FR4): a community favourite for PC and nylon. Almost zero warp because Garolite stays dimensionally stable at 120C. Downside: you can't easily swap to other materials. Smooth PEI: works for small parts only. Anything taller than a credit card warps off. Brim strategy: PC essentially requires a brim. Set 5 to 10mm in OrcaSlicer's Other tab, and use mouse-ear brims on the corners of rectangular parts. The Orca community rule is [truncated] --- ## OrcaSlicer TPU Settings: Soft, Medium & Hard Shore Hardness (2026) URL: https://orcaslicer.net/orcaslicer-tpu-settings/ I tried to print NinjaFlex 85A on a Bowden Ender 3 about two years ago and the filament just buckled in the extruder, wrapping around the gear like spaghetti. Twenty minutes of disassembly later I learned the rule: soft TPU is a direct-drive-only material. The Shore hardness on the spool matters more than the brand, and OrcaSlicer's Generic TPU defaults assume 95A. Here's what each hardness actually wants. I've since printed TPU at 60A, 85A, 95A, 98A, and 63D across a Voron, a Bambu X1C, an A1 Mini, and that same Ender 3 (after I converted it to direct drive). Every Shore band has its own personality, and the slicer settings that nail one will jam the next. This guide hands you per-hardness numbers, the real story on Bambu AMS compatibility (it's not what you think), and the two OrcaSlicer bugs that actively fight you during calibration. Table of contents TPU in 60 seconds (why it's different) Shore hardness explained The AMS-TPU compatibility myth (read this first) OrcaSlicer's built-in TPU profile (what's actually in it) TPU 95A starting settings (the baseline) Per-Shore-hardness settings Extruder reality check: direct drive vs Bowden vs AMS Bed adhesion and first layer for TPU Stringing, oozing, and Z-hop for TPU Common TPU failures and fixes Brand notes (what's different about popular TPUs) Calibration order for TPU FAQ TPU in 60 seconds (why it's different) TPU stands for thermoplastic polyurethane. It's an elastomer, so it acts like soft plastic when cool and a sticky, taffy-like melt when hot. It doesn't warp like ABS, which means you don't need an enclosure. It is, however, hygroscopic. Wet TPU pops, hisses, prints fuzzy, and produces weak layers. Dry the spool at 65 to 70 C for 8 to 12 hours before you print anything serious. The thing that makes TPU genuinely different from PLA or PETG isn't temperature. It's that the filament itself is springy. Your extruder gear can push faster than the filament can transmit force down the path to the melt zone. The filament compresses, then buckles, then kinks, then jams. That's the entire physics of every TPU failure I've ever had. The slicer setting that controls this is max volumetric speed, and for soft TPU it's by far the most important number on the panel. If you only learn one TPU lesson from this guide, learn that the max volumetric speed setting is what stops your filament from buckling. The second rule is the extruder. Direct drive minimizes the distance between the gear and the melt. Bowden setups put a long PTFE tube in between, and inside that tube is exactly where soft TPU buckles. I'll cover this in detail below, but the short version: anything 90A or softer needs direct drive, full stop. Shore hardness explained Shore A measures soft, rubbery materials. Shore D measures harder, plastic-leaning materials. The scales overlap at the firm end (roughly 95A equals about 45D, and 63D equals something past 100A). The number on the spool tells you how the filament behaves both in your hand and in your extruder. Here's the honest version of the scale, with what each band is good for and what it'll do to your printer. Shore band Feel Typical use cases Print difficulty 60A to 70A Squishy, almost gummy Vibration dampers, soft prosthetic liners, foam-like cosplay Very high. Direct drive only. Loose idler tension required. 75A to 85A True flexible, bends easily Shoes, watch bands, soft phone cases, gaskets High. NinjaFlex 85A sits here. Direct drive mandatory. 90A to 95A Medium-firm, the "default" TPU Protective cases, drone feet, belts, plugs Medium. Direct drive preferred. Bowden possible at 95A with tuning. 98A Very firm, rubbery rebound Bumpers, gears, semi-rigid clips Low. Bowden compatible. Easiest TPU to print. 63D (Shore D) Semi-rigid, still has flex Industrial parts, rigid tread wheels, hinges Low. Prints more like soft PETG. The single thing I wish someone had told me earlier: temperature that's perfect at 95A will buckle and jam at 70A. Shore hardness changes everything about how TPU prints, not just how it feels in your fingers. The AMS-TPU compatibility myth (read this first) This is the section I'd want pinned to the top of every Bambu forum thread, so I'm putting it before the slicer settings. The common belief is that AMS Lite handles flexible TPU because it has a shorter path than standard AMS. That's wrong. No Bambu AMS variant auto-feeds flexible TPU. Not AMS, not AMS Lite, not AMS 2 Pro, not AMS HT. "Bambu TPU for AMS" exists, and it's the source of the confusion. It is not the same as TPU 95A. Bambu reformulated it at roughly Shore 68D, which is significantly harder than 95A and closer to a tough, rubbery PETG. They made it firm on purpose so it could survive the AMS feed path without buckling. If you're shopping for soft, bendy TPU and you grab a roll of "Bambu TPU for AMS" thinking it's the same as the rest, you'll end up with a near-rigid filament you can't bend in your hand. Here is the actual compatibility story across every Bambu AMS variant. I've checked this against Bambu's own filament guide table, multiple forum confirmations, and the OrcaSlicer source profile for "Bambu TPU for AMS" which lists the filament type as TPU-AMS rather than just TPU. AMS variant Flexible TPU (95A or softer) "Bambu TPU for AMS" (~68D) Workaround for soft TPU AMS (standard, X1C/P1S) Not supported. TPU 95A HF officially blocked from auto-feed. Works (this is what it's designed for). External spool, fed directly into the toolhead. AMS Lite (A1 / A1 mini) Not supported. The longest filament path of any Bambu setup. Works. External spool holder, route directly to the feeder. AMS 2 Pro Not supported for true flexible TPU. Works. External spool. AMS HT Functions as a sealed dry box only when printing TPU. Manual feed through the dedicated back outlet. Works through standard feed. The back-outlet manual feed is the workaround. I know this is unwelcome news if you bought an A1 expecting AMS Lite to feed you NinjaFlex. The reality is Bambu's official recommendation for any TPU below 95A on the A1 series is to bypass the AMS Lite entirely and use the external spool holder routed directly into the toolhead's PTFE inlet. There's more context on what each AMS variant can and can't do in the Bambu AMS compatibility guide. Red TPU loaded on a Bambu printer via the Device tab. Notice the filament path: when you're running flexible TPU, this is the only routing you should trust, no matter which AMS sits on top of your printer. OrcaSlicer's built-in TPU profile (what's actually in it) OrcaSlicer ships a "Generic TPU @System" profile that inherits from a base file called fdm_filament_tpu.json in the OrcaFilamentLibrary. I pulled the actual values from the source so you know exactly what the slicer is doing when you pick that dropdown option. Setting OrcaSlicer Generic TPU default Honest note Nozzle temperature 240 C High. Most 95A brands publish 220 to 235 C. Dial down to 225 to 230 once you have a working baseline. Nozzle initial layer 240 C Same. Nozzle range (low/high) 200 / 250 C The legal window. Textured plate temp 35 C Correct. Don't go above 60 C or TPU will slip on the plate. Cool plate temp 30 C Use with glue stick as release agent. Max volumetric speed (MVS) 3.2 mm³/s Low on purpose. Generic PLA is 15. Don't raise this for soft TPU. Retraction length 0.4 mm Way lower than PLA's 0.8 to 1 mm. Correct. More retraction equals more jam risk on TPU. Fan min speed 100% Aggressive cooling. Drop to 50 to 70% if you see layer separation. Close fan first X layers 1 Fan off for layer 1 only. Fine. Filament density 1.24 g/cm³ Standard for TPU. I want to flag the 240 C number specifically. It's high compared to what Polymaker, Overture, Siraya Tech, and NinjaTek publish for their TPU lines. OrcaSlicer biases hot on purpose because lower-viscosity melt extrudes more easily, which reduces jam risk more than it hurts surface quality on TPU. Once you've printed a calibration cube successfully, run a temperatur [truncated] --- ## OrcaSlicer ABS & ASA Settings: Why You Need an Enclosure (2026) URL: https://orcaslicer.net/orcaslicer-abs-asa-settings/ I printed a Voron part in ABS on an open-frame Ender 3 once. Watched it warp off the bed at layer 60 like the plastic was trying to escape the printer. ABS without an enclosure is genuinely not worth attempting unless you've cobbled one together out of cardboard and a space heater. Here's the OrcaSlicer settings recipe for ABS and ASA, plus the honest take on whether your printer can actually handle the material. I've run ABS through a P1S, an X1C, a Voron 2.4, and (briefly, regrettably) that open-frame Ender. The pattern is consistent. About 90% of the "OrcaSlicer ABS not working" threads I read on the Bambu and Voron forums trace back to one of three things: no enclosure or a poorly sealed one, a cooling fan setting that nobody bothered to drop to zero, or a cold draft from an open window or AC vent ruining the chamber temp mid-print. The other 10% is wet filament, but we'll get to that. This guide is for the people actually trying to print functional ABS or ASA parts in OrcaSlicer in 2026. If you're on a Bambu X1C, P1S, X1E, or Voron 2.4, you're in the right spot. If you're on a Bambu A1, P1P (stock), MK4 (stock), or open-frame Ender, I'll tell you straight: you should be running PETG instead, or building an enclosure first. Table of contents ABS and ASA in plain English The enclosure question (no, you can't skip it) OrcaSlicer's built-in ABS and ASA profiles Recommended starting settings: Generic ABS Settings for ASA: what changes vs ABS The anti-warping recipe Fume safety (don't skip this) Common failure modes and fixes ABS variants: ABS-GF, ABS+, PC-ABS FAQ ABS and ASA in plain English ABS is acrylonitrile butadiene styrene. It's a three-monomer terpolymer, roughly 15-35% acrylonitrile, 5-30% butadiene, and 40-60% styrene by weight. The acrylonitrile and styrene give the material its stiffness and shape, and the butadiene rubber phase gives the impact resistance that makes ABS the standard for LEGO bricks, automotive trim, and 3D-printed enclosures that need to survive being dropped. ABS is an amorphous polymer, which means it doesn't have a true melting point. What it has is a glass transition temperature (Tg) of roughly 105 C. Below 105 C, ABS is a hard solid. Above 105 C, it softens. That's why ABS parts can sit in a hot car or near a 3D printer's heated bed without sagging, while PLA turns into modeling clay at 60 C. It's also why we can heat-soak the chamber to 50-60 C without the print itself going soft. What ASA is and why it exists ASA is acrylate styrene acrylonitrile. It's the same family as ABS, but the butadiene rubber phase is swapped for an acrylate rubber. That sounds boring, and chemically it kind of is, except for one thing: butadiene has carbon-carbon double bonds that UV light loves to attack. Those double bonds break under sunlight, the rubber phase oxidizes, and you get yellowing and surface cracking after a few months outdoors. Acrylate rubber doesn't have those bonds. ASA parts hold color and mechanical integrity in direct sunlight for years. That's the whole pitch. ASA is "ABS with UV resistance." It prints essentially identically: same warping tendencies, same need for chamber heat, same fume profile (slightly less smelly per most users), same hygroscopic behavior. The OrcaSlicer profile for ASA runs about 5-10 C hotter at the nozzle than the ABS profile, and that's pretty much the only meaningful settings difference. When to pick ABS vs ASA vs going back to PETG Use caseBest pickWhy Indoor functional part, no UV exposureABSCheaper, more brand options, mechanically equivalent Outdoor part, mounted in sunlightASAUV stability over years, not months Automotive interior trim, dashboardASACabin temperatures plus UV through windshield Drone arms, enclosure brackets, RC partsABS or ABS-GFImpact resistance and stiffness Acetone vapor smoothing plannedABSASA is less responsive to acetone smoothing Open-frame printer, no enclosurePETGStop fighting physics, switch materials Cosmetic display piecePLABetter surface finish, easier print, no fume issue If you're choosing ABS for outdoor parts and don't need acetone smoothing, you're using the wrong material. Go ASA. If you're choosing ABS because you read it's "stronger than PLA" and you have no enclosure, you're about to learn an expensive lesson in warping. The OrcaSlicer filament settings pillar covers the full material matrix if you're still deciding. The enclosure question (no, you can't skip it) I'll keep this section short and honest. ABS and ASA warp because the polymer shrinks significantly as it cools from print temperature (245 C molten) down past Tg (105 C) and onward to room temperature. The bottom of the print cools first because it's bonded to the bed and the cooled layers beneath it. The top is still being deposited hot. Differential cooling means differential shrinkage, and differential shrinkage pulls the corners off the bed. An enclosure fixes this two ways. It keeps the ambient air around the print warm (40-60 C) so the print cools slowly and uniformly. And it blocks drafts, which is what causes those audible mid-print "pop/crack" sounds when a tall ABS print delaminates along a layer line. Printer compatibility, honest version PrinterABS/ASA ready?Chamber typeRealistic chamber temp Bambu X1EYes, best in classActive heated50-60 C controlled Bambu X1CYesPassive (bed-soak)45-55 C with door shut Bambu P1SYes, with caveatsPassive (bed-soak)40-45 C max Bambu H2D / H2SYesActive heated50-60 C Voron 2.4 / TridentYes, with bed fansPassive plus DIY bed fans50-60 C achievable Qidi X-Max 3, X-Plus 3YesActive heated55-65 C Creality K1 Max, K2 PlusYesEnclosed, partial heating40-50 C Snapmaker Artisan, J1YesEnclosed40-50 C Prusa XL with enclosureYesPassive40-50 C Bambu A1 / A1 miniNo, don't tryOpen frame, no kitAmbient room Bambu P1P stockNo, upgrade to P1S kit firstOpenAmbient room Prusa MK4S, MK4 stockNo, buy the enclosureOpenAmbient room Ender 3 family, Sovol SV06/SV07No, build an enclosureOpenAmbient room The Bambu A1 and A1 mini are excellent printers, but they're open-frame by design and Bambu doesn't sell an enclosure kit for them. Trying to print ABS on an A1 is a "don't" not a "should I." If you have one and you need ABS-grade parts, the answer is: buy a P1S, or build a cardboard-and-foam enclosure as a stopgap, or stick to PETG. Can I get away without an enclosure? For very small parts under about 40 mm tall and 80 mm in any horizontal dimension, sometimes yes. The thermal mass is low, the surface area for differential cooling is low, and you might get a clean print in a draft-free room. But the part will still be weaker than the same geometry printed in PLA at the same speed, because the layer adhesion is poor without ambient heat. So you're using a fume-heavy, harder-to-print material to get a worse result than PLA. That's not a win. If you're committed to printing ABS without an enclosure, the cardboard-and-tape DIY enclosure is real. A large cardboard box with the front cut open as a flap, taped around the printer's frame, plus a piece of clear acrylic or shower curtain as a viewing window, will hold 35-40 C around a Bambu P1P or MK4. Not pretty, not safe long-term, but functional for a few prints while you decide whether to commit to ABS on real hardware. Just make sure nothing flammable touches the bed or heatbreak. OrcaSlicer's built-in ABS and ASA profiles OrcaSlicer ships with a fairly deep library of ABS and ASA presets out of the box. You'll find them by clicking the filament dropdown in the top-left of the prepare panel, then selecting your printer profile, then scrolling the filament list. The generic profiles are your fallback when your specific spool brand isn't listed. Profile typeWhat's availableWhen to use GenericGeneric ABS, Generic ASA, Generic ABS-GF, Generic PC-ABS, Generic PCTGUnknown brand, off-brand spool, or starting point for tuning BambuBambu ABS, Bambu ASA, Bambu ASA-Aero, Bambu ABS-GF, Bambu PCUsing Bambu's own filament, especially on X1C [truncated] --- ## OrcaSlicer PETG Settings: The Anti-Stringing Profile (2026) URL: https://orcaslicer.net/orcaslicer-petg-settings/ I spent three weekends fighting PETG stringing on a Bambu A1 before I finally accepted that the PLA retraction recipe doesn't translate. PETG is a different polymer with different physics. The strings aren't a slicer bug or a printer flaw, they're a chemistry problem with a chemistry answer. Here's the anti-stringing recipe that actually works in OrcaSlicer. If you've landed here, you've probably tried bumping retraction to 6 mm on a direct drive, slammed the fan to 100 percent, dropped the nozzle 10 degrees, and watched your part come off the bed with even more hair than before. I've done all of that. So has every PETG user on the Bambu forum, the Prusa forum, and the OrcaSlicer issue tracker. The good news is that PETG stringing is solvable. It just needs a different mental model than PLA, and it needs you to do one thing most articles bury at the bottom: dry the spool. This guide is a 2026 update for OrcaSlicer 2.x. It works on Bambu A1, P1S, X1C, Voron, Prusa MK4, Ender 3 family, Sovol, and any other printer that Orca runs. I'll give you the recipe up front, then explain why each setting is the way it is, then walk through per-variant tweaks for PETG-CF, PETG HF, Translucent, and Tough. There's also a troubleshooting matrix at the end that I wish I'd had three years ago. Table of Contents The Anti-Stringing Recipe (TL;DR) Why PETG Strings (The Physics, Briefly) OrcaSlicer's Bundled PETG Profiles Tuning PETG Step by Step Setting-by-Setting Breakdown Bed Adhesion (Don't Kill Your Plate) PETG Variants: When to Deviate PETG Troubleshooting Matrix PETG vs PLA: Quick Decision Guide Advanced: PETG as Support Interface for PLA FAQ Conclusion The Anti-Stringing Recipe (TL;DR) If you only do three things, do these: dry the filament at 65 C for 6 to 8 hours, lower your retraction speed below your PLA value, and enable Z-hop at 0.2 mm. That alone fixes the majority of PETG stringing complaints I've seen on forums. Everything else in this guide is refinement on top of those three. Here are the starting values I run on a stock 0.4 mm hot end. Treat these as calibration baselines, not gospel. Per-brand tuning still matters, and your spool's temperature label is a hint, not a command. SettingDirect DriveBowdenNotes Nozzle temperature235 to 240 C235 to 245 C240 C first layer, 235 to 240 C after Bed temperature75 to 80 C75 to 80 CConservative guides cite 50 to 70 C; community runs hotter Part cooling fan30 to 50%30 to 50%0% on layers 1 and 2 Outer wall speed40 to 60 mm/s40 to 60 mm/sSlower hides recovery artifacts Inner wall / infill80 to 150 mm/s80 to 120 mm/sBound by MVS, not speed Max volumetric speed10 to 13 mm3/s10 to 13 mm3/sHF variants and Bambu hot ends can push 18 to 22 Retraction length0.8 to 1.5 mm4 to 6 mmLonger rarely helps past these Retraction speed25 to 40 mm/s25 to 35 mm/sSlower than PLA on the same machine Z-hop0.2 to 0.4 mm0.2 to 0.4 mmType Auto or Normal Wipe before retractOn, 2 mmOn, 2 mmHides seam blobs Pressure Advance0.040 to 0.0600.20 to 0.60 (Klipper)Calibrate per filament Travel speed200 to 300 mm/s200 to 250 mm/sFaster travels = less drool time First layer speed20 to 30 mm/s20 to 30 mm/sSlow first layer, hot first layer Brim5 mm if small footprint5 mm if small footprintPETG warps less than ABS but still warps The full anti-stringing setup is a 30-minute job if you've never done it. It's worth pairing with the broader OrcaSlicer filament settings guide for context on where each option lives in the UI. Then walk the calibration path: temperature tower first, retraction test second, pressure advance third. One thing to flag before you start: do not use a smooth PEI plate without a release agent. PETG bonds chemically to PEI when both are hot, and on cooldown it can rip the PEI coating right off your build plate. It's a permanent plate-killer. I cover this in detail in the bed adhesion section, but if that's the only thing you remember, I'll have done my job. Three parts from one calibrated PETG profile. None of them are perfect, but the strings are gone and the layers tear instead of snap. Why PETG Strings (The Physics, Briefly) I'm not going to lecture, but two minutes of polymer chemistry will save you hours of slider-dragging. PETG is polyethylene terephthalate glycol-modified. The glycol modification disrupts the regular PET crystal structure, which is what gives PETG its transparency and printability. It's also what gives it a long, sticky melt window. PETG's glass transition temperature (Tg) sits around 80 C, with sources placing it anywhere from 75 to 85 C depending on formulation. PLA's Tg is roughly 60 C. That higher Tg matters because PETG stays soft and tacky for longer above the nozzle exit. The molten strand has more time to drag, more time to stick to itself, and more time to form a hair on a travel move. The melting range matters too. PETG doesn't have a sharp melting point the way crystalline polymers do. It becomes viscous around 230 C and is fully fluid by 250 to 260 C. Most printing happens inside that "still viscous, not quite liquid" window. Flow is sticky. Pressure inside the nozzle builds and releases slowly. Where PLA snaps cleanly on a retraction, PETG stretches like cheese and leaves a tail. Then there's moisture. PETG is aggressively hygroscopic. A spool left out overnight in a humid garage can already show stringing symptoms. Inside the hot end, absorbed water flashes to steam, which disrupts the melt, weakens layer bonds, and pushes extra material out the nozzle during travels. This is the single biggest cause of "I've tried every setting and it still strings." If you've ever heard a faint popping or hissing during a PETG print, that's water boiling out of your filament in real time. There's a knock-on effect on cooling. PETG cools more slowly than PLA across a print. That's good for layer adhesion because the previous layer stays warm enough to fuse properly. It's bad for stringing because the molten strand has more dwell time during travels. You can see why this material punishes lazy retraction settings. The short version: the PLA retraction recipe (short distance, fast speed) doesn't work because PETG's melt physics are different. You need slightly more retraction distance, noticeably slower retraction speed, and Z-hop to physically separate the nozzle from the print during travels. If you came here from my OrcaSlicer PLA settings guide, this is where the recipes diverge. OrcaSlicer's Bundled PETG Profiles OrcaSlicer ships with a Generic PETG, Generic PETG HF, and presets for popular brands including Bambu, Polymaker, Overture, eSUN, Hatchbox, Sunlu, Prusament, and ELEGOO. Coverage expanded a lot in 2024 to 2025, so the exact list depends on your installed version. If your spool's brand isn't bundled, the Generic PETG profile plus a calibration pass is fine. ProfileTypical MVS DefaultBest ForNotes Generic PETG10 to 12 mm3/s (varies)Any PETG when no brand profile existsConservative; see GitHub issue 12303 Generic PETG HF14 to 22 mm3/sHigh-flow formulationsPolymaker PolyLite PETG HF, etc. Bambu PETG Basic~12 mm3/sBambu X1/P1/A1 stockTuned for textured PEI plate Bambu PETG HF~18 mm3/sFast Bambu printsRecalibrate if importing into Orca from Studio Bambu PETG Translucent~10 mm3/sOptical clarityLower fan, lower temp, thin layers Bambu PETG-CF~14 mm3/sStiff functional partsHardened steel nozzle required, ~255 C Polymaker / Overture / eSUN / Hatchbox10 to 14 mm3/sBrand-specific tuningStill verify against your spool One honest caveat. OrcaSlicer's bundled Generic PETG MVS default has been wrong for some printer profiles. GitHub issue #12303 documents the Generic PETG MVS being set to 20 mm3/s on at least one printer profile (Sovol SV06), which is well above what a stock 0.4 mm hot end can actually deliver. The user's own testing pointed to about 12 to 13 mm3/s as the real ceiling. This is exactly why I tell people to run their own max volumetric speed test rather than trusting any bundled number, no matter how off [truncated] --- ## OrcaSlicer PLA Settings: Temp, Speed & Cooling for Sharp Prints (2026) URL: https://orcaslicer.net/orcaslicer-pla-settings/ I've burned through 47 spools of PLA across 6 different printers in the last two years and I can tell you with confidence that OrcaSlicer's Generic PLA defaults aren't wrong, they're just conservative. The slicer ships with 220°C nozzle, 55°C bed, 100% fan, and a 12 mm³/s max volumetric speed because those numbers work on basically everything from a stock Ender 3 to a Bambu X1C. They're also leaving roughly 30% of your print speed on the table, and on some brands they're 10°C hotter than the sweet spot. Here's the path I use to go from defaults to dialed in, with the actual numbers pulled from OrcaSlicer's source profiles on GitHub (verified May 12, 2026). I'll show you what loads when you click "Generic PLA," why each value is set where it is, what to change when you swap brands, and how to read the calibration tests so you stop second-guessing every print. Table of contents PLA in 60 seconds, and why it prints the way it does OrcaSlicer's PLA profiles, and why Generic PLA is conservative Recommended starting values for PLA on a 0.4 mm nozzle Temperature, find the right number for your spool Speed and cooling are the same dial Max volumetric speed, the ceiling you didn't know you hit PLA variants, when Generic doesn't cut it First layer and bed adhesion by plate type Retraction and pressure advance for PLA Common PLA problems and what actually fixes them PETG, ABS, and TPU sit on the same shelf FAQ PLA in 60 seconds, and why it prints the way it does I'll start with the chemistry briefly because a few slicer settings make more sense once you know what's inside the filament. PLA is polylactic acid, a thermoplastic polyester polymerized from lactide, usually derived from corn starch. I think of it as the easiest filament on the shelf because it doesn't warp, doesn't stink, and sticks to almost any bed surface at 55°C. I also remind people that ease has a price. Its glass transition temperature is around 60 to 65°C, so a part sitting in a sun-baked car cabin or going through a dishwasher cycle will droop. I've watched a phone holder in my dashboard turn into modern art on a hot July afternoon. OrcaSlicer's source profile sets the temperature_vitrification field to 45°C as a safety floor. I keep hammering this with new printer owners. PLA isn't an engineering plastic. It's stiffer and harder than PETG, ABS, or TPU, but it's also more brittle. I find Bambu's own source profile data useful here: 13.8 MPa impact strength for PLA Basic versus 4.6 MPa for Silk and 7.8 MPa for PLA-CF, so even within "PLA" the numbers swing hard. For a wider map of how PLA fits next to other materials, the OrcaSlicer filament settings pillar is the place to start. I wish someone had told me one more thing on day one: PLA is mildly hygroscopic. It absorbs roughly 0.5 to 1% water by weight in 24 hours at 50% relative humidity. That's less than PETG and far less than nylon, but it's more than the "PLA doesn't need drying" crowd will admit. Wet PLA pops at the nozzle, strings like spider web, and snaps when you bend a thin wall. I've solved more "my prints suddenly look bad" tickets with a filament dryer than with any slicer change. Dry the spool at 45 to 50°C for 4 to 8 hours. Never go above 55°C because the spool can fuse to itself. OrcaSlicer's PLA profiles, and why Generic PLA is conservative I'd open OrcaSlicer alongside this section if you can. The Filament dropdown in the Prepare tab is alphabetical. The base profile that everything else inherits from is Generic PLA @System, and the file it loads (fdm_filament_pla.json) is the canonical default. I pulled the values below directly from that file on May 12, 2026. The Filament dropdown in OrcaSlicer with Generic PLA loaded. Everything in this article maps back to fields on this panel. Generic PLA verified defaults (pulled from source on 2026-05-12) ParameterValueWhat it does nozzle_temperature220°CPrint temp after the first layer nozzle_temperature_initial_layer220°CSame temp on layer 1 nozzle_temperature_range190 to 240°CAllowed slider range hot_plate_temp55°CBed temp for textured PEI cool_plate_temp35°CBed temp for Bambu Cool Plate eng_plate_temp0°CEngineering plate, not recommended for PLA filament_max_volumetric_speed12 mm³/sThe real ceiling on print speed filament_flow_ratio0.98Extrusion multiplier baseline filament_density1.24 g/cm³Used for weight estimates fan_min_speed100%Part cooling fan at full after layer 1 close_fan_the_first_x_layers1Fan off for layer 1 only overhang_fan_threshold50%Overhang ratio that forces full fan slow_down_layer_time4 sForce minimum 4-second layers on small parts slow_down_min_speed10 mm/sFloor speed when slowing for cooling fan_cooling_layer_time100 sRamp fan up if layer takes longer additional_cooling_fan_speed70%Aux fan speed if your printer has one temperature_vitrification45°COrca's "part deforms above this" floor I think the most important thing to understand about these numbers is that they're conservative on purpose. The 220°C is on the high end of practical, the 12 mm³/s MVS leaves a stock 0.4 mm hot end coasting, and the 100% fan from layer 2 is correct for surface detail but slightly hurts layer adhesion. I compare it with Bambu's own Generic PLA Basic profile, which sets MVS to 21, almost double. I don't read that as "the filament is different." I read it as Bambu knowing their hot end can push more and tuning the profile to it. Generic PLA variant overrides I find the Generic PLA variants useful because each one only overrides the values that differ from the base, which keeps things tidy. OrcaSlicer ships several of them. VariantMVSRetractionFlow ratioNotes Generic PLA @System12 mm³/sbase0.98The catch-all baseline Generic PLA High Speed18 mm³/sbase0.98For HF chemistry spools Generic PLA Matte11 mm³/s0.8 mm0.98Filler-loaded, slightly slower Generic PLA Silk7.5 mm³/s0.5 mm0.98Lower fan, longer min layer time (8 s) Generic PLA-CFbasebase0.95Hardened nozzle required Bambu Lab variants worth knowing VariantNozzleMVSFlowNotes Bambu PLA Basic220°C210.98Impact 13.8 MPa, the "standard" PLA Bambu PLA Matte220°C~150.98Impact 6.6 MPa, more brittle Bambu PLA Tough+245°C210.98Range 220 to 250, longer min layer (8 s) Bambu PLA Aero (LW-PLA)250°C typ.60.6Foaming, fan_min 30% Bambu PLA Silk230°Clower0.98Outer wall 40 to 60 mm/s Bambu PLA-CFbasebase0.98Required nozzle HRC 40 I scroll further and find Polymaker (PolyLite, PolyMax, PolyTerra), eSUN PLA+, Hatchbox, Prusament, Overture, Atomic, Sunlu, Anycubic, Elegoo, Inland, and Creality presets. They're available even on a non-Bambu printer. I treat them as starting points because every spool of "Hatchbox PLA" can differ from the next based on which factory pressed it. Recommended starting values for PLA on a 0.4 mm nozzle I'll split this into two columns. The first is what to set when you've got a stock printer and a fresh spool. The second is what to set when you've tuned the printer and you know the filament. I keep the first column intentionally cautious. I run most of my prints with the second column values. SettingStock printer, first spoolTuned printer, known filament Nozzle temp210 to 215°C205 to 220°C (from your tower) First layer nozzle+5°C of print tempsame as print temp Bed temp (textured PEI)60°C55°C Bed temp (Cool Plate)45°C35°C Part cooling fan100% from layer 2100% from layer 2 (lower for silk) MVS12 mm³/s (Orca default)15 to 22 mm³/s (tower it) Flow ratio0.980.95 to 1.02 (from your test) Retraction (direct drive)0.8 mm at 30 mm/s0.4 to 1.0 mm Retraction (Bowden)4 mm at 35 mm/s3 to 6 mm Pressure advanceleave default, then towerfrom your PA tower I want to be honest about how I treat that left column. It's a starting point that gets you a usable first print without disasters. It's not the answer. I'll insist on this: every printer plus brand plus color combination wants its [truncated] --- ## Best OrcaSlicer Filament Settings: All Materials Cheatsheet URL: https://orcaslicer.net/orcaslicer-filament-settings/ I keep a notebook on the workbench. Two pages of it are nothing but per-spool calibration values, written in pencil because they keep changing. Bambu PLA Matte black, flow 0.97. Polymaker PolyTerra grey, flow 0.99. eSun PETG translucent, flow 0.94, retract 1.1mm at 28 mm/s. The notebook exists because the OrcaSlicer defaults are sensible starting points, not destinations. This guide is the digital version of that notebook. Per-material settings for every common FDM filament: PLA, PETG, ABS, TPU, PC, Nylon, the carbon-fiber blends, soluble supports, and the exotic stuff. Real numbers from manufacturer datasheets, the OrcaSlicer wiki, and roughly two years of personal printing. Updated against OrcaSlicer v2.3.2 (current stable build). You'll also find: how to add a custom filament profile properly, where the AMS RFID auto-detect actually works (and where it doesn't), and the single biggest hygroscopic mistake people make. The master cheatsheet Use this as your starting-point reference. The values are ranges because every spool varies, but landing inside the range will get you a printable result on a generic 0.4mm direct-drive setup. Bowden retraction values are roughly 4 to 6 times higher than the direct-drive numbers shown. Filament Nozzle Bed Cooling Retraction (DDE) Speed (outer) Max vol PA (Klipper) Enclosure? Drying PLA Basic200-215°C55-65°C100% after L20.4-0.8mm @ 35-5050-80 mm/s14-18 mm³/s0.025-0.040No50°C × 6h (optional) High-speed PLA205-230°C55-65°C100%0.4-0.8mm @ 40-50250-500 mm/s*22-32 mm³/s*0.020-0.035No50°C × 6h Silk PLA215-230°C55-65°C30-50%1-2mm @ 30-4540-60 mm/s8-12 mm³/s0.030-0.045No50°C × 6h (always) PLA+205-225°C55-65°C80-100%0.4-0.8mm @ 35-5050-100 mm/s12-15 mm³/s0.030-0.045No50°C × 6h (eSun ships wet) PETG235-245°C75-85°C30-50% (off L1-2)0.6-1.2mm @ 25-4040-60 mm/s7-10 mm³/s0.045-0.070Optional65°C × 8h (always) PETG-CF240-260°C75-90°C30-40%1-2mm @ 25-4040-100 mm/s6-10 mm³/s0.040-0.060Recommended70°C × 8h (always) ABS240-250°C100-110°C0-10% (off L1-3)0.4-0.8mm @ 35-5040-80 mm/s8-12 mm³/s0.030-0.050Required75°C × 4h (optional) ASA245-255°C100-110°C0% off L10.4-0.8mm @ 35-5040-70 mm/s8-11 mm³/s0.030-0.050Required80°C × 4h TPU 95A220-235°C40-60°C30-50%0.5-1.5mm @ 20-3020-40 mm/s4-8 mm³/s0.05-0.08No55°C × 6h (always) TPU 85A210-230°C40-50°C30-40%0mm or as low as possible15-25 mm/s1-2 mm³/s0.06-0.10No45°C × 6-8h PC blend250-270°C100-110°C0-10%1-2mm @ 30-4530-60 mm/s6-10 mm³/s0.025-0.045Required85°C × 4-8h (mandatory) Pure PC270-310°C110-120°C0%1-3mm @ 30-4530-50 mm/s5-8 mm³/s0.025-0.045Required120°C × 4-8h (mandatory) PA (Nylon)240-280°C90-110°C0-30%1-3mm @ 25-4030-50 mm/s6-10 mm³/s0.030-0.045Required80°C × 12h (mandatory) PA-CF270-290°C90-110°C0-20%1-3mm @ 25-3530-80 mm/s8-12 mm³/s0.025-0.040Required80°C × 12h (mandatory) PVA200-215°C50-60°C30-50%1-2mm @ 30-4025-40 mm/s3-5 mm³/s0.04-0.06No45°C × 12h (mandatory) BVOH205-220°C50-65°C30-50%1-2mm @ 30-4025-40 mm/s3-5 mm³/s0.04-0.06No60°C × 4h (mandatory) Wood-fill PLA190-215°C50-60°C80-100%0.6-1.2mm @ 35-5030-50 mm/s6-10 mm³/s0.030-0.050No50°C × 6h Metal-fill PLA195-220°C50-60°C100%0.4-0.8mm @ 35-5030-50 mm/s5-8 mm³/s0.030-0.045No50°C × 6h Glow PLA205-220°C55-65°C100%0.4-0.8mm @ 35-5030-50 mm/s5-8 mm³/s0.030-0.045No50°C × 6h * High-speed PLA needs a high-flow hotend (Bambu, Revo HF, CHT clones, Volcano) to actually achieve the rated speed. Stock V6 hotends will under-extrude past about 18 mm³/s regardless of what you set in the slicer. Hardened nozzle warning: every carbon-fiber, glass-fiber, glow, sparkle, marble, wood, magnetic, and metal-filled filament needs a hardened nozzle. 0.4mm minimum, 0.6mm preferable. Brass nozzles wear visibly within 500g. OrcaSlicer's filament profiles surface this via the required_nozzle_HRC field, which warns you if the hardness setting doesn't meet the filament's needs. The Filament settings dialog is where every per-spool value lives in OrcaSlicer. Type, vendor, diameter, density, flow ratio, pressure advance, recommended temps. Open it from the Filament panel's edit icon next to the loaded filament name. The hygroscopic ranking (and the mistake everyone makes) Every filament absorbs moisture from the air. Wet filament prints with popping, stringing, blobs, milky surfaces, and weak layer adhesion. The single most common AI-generated guide claim, "TPU is more hygroscopic than PETG", is wrong. Real-world ranking, fastest to slowest absorption rate: Nylon (PA, PA-CF, PA6-GF): absorbs up to 10% of its weight in 12 hours of room-air exposure. Print directly from a heated dry box if you want consistent results. Without drying, prints look fuzzy and pop continuously at the nozzle. PETG: absorbs 0.5 to 1% in 24 hours at 50% relative humidity. Wet PETG strings catastrophically and the layer surface looks slightly milky. PVA / BVOH: ridiculously hygroscopic. Will turn into spaghetti within hours of being out of a dry box if humidity is above 30%. PLA / ABS: roughly equivalent rates. Slower than PETG. Will absorb meaningfully over weeks but not hours. TPU: absorbs less than 1%, but matters more because TPU is already soft and any moisture-driven defects compound quickly. PEEK / engineering high-temp materials: least hygroscopic among common FDM filaments. Drying temperatures from manufacturer datasheets (Bambu, Polymaker, eSun): PLA: 50°C × 6 hours PETG: 65°C × 8 hours ABS: 75°C × 4 hours ASA: 80°C × 4-6 hours TPU 95A: 55°C × 6 hours; TPU 85A: 45°C × 6 hours PC blends: 85°C × 4 hours; pure PC: 110-120°C × 4-8 hours Nylon: 80-95°C × 8-12 hours PVA: 45-50°C × 8-12 hours The mistake everyone makes: assuming a sealed bag of new filament is dry. Most factory packaging is humidity-permeable enough that filament shipped during a rainy month from a non-AC warehouse arrives partially wet. eSun and Polymaker have both been called out on Reddit and on the Prusa forum for shipping wet PLA+. Always dry on receipt for hygroscopic materials. PLA: standard, high-speed, silk, and the +/Pro tier PLA Basic is the default first stop. OrcaSlicer ships a "Generic PLA" profile and brand-specific PLA presets for Bambu, Polymaker, eSun, Overture, Hatchbox, and Sunlu. Wiki temperature range 180-220°C. Practical printing range much narrower at 200-215°C. Bambu PLA Basic typically lands at 210, Polymaker PolyTerra at 205. Bed 60°C is fine for textured PEI. PLA wants aggressive cooling. It's the only common material where you genuinely run 100% fan from layer 2 onwards. Bridge sag, overhang quality, and surface detail all improve linearly with cooling. The exception is the first three layers, where you want fan low or off for adhesion. Common gotcha: stringing on PLA usually means the spool is wet, not that retraction is undertuned. PLA absorbs moisture more slowly than PETG but it absolutely absorbs. Dry first, retest, then tune retraction. Bambu PLA Matte vs PLA Basic: Matte runs about 5°C hotter (210-230 vs 200-220) and lower max volumetric speed (10 vs 18 mm³/s) because the matting filler (essentially calcium carbonate) increases viscosity. Same retraction, same adhesion, similar PA. Don't use the PLA Basic profile for PLA Matte; the speed will under-extrude. High-speed PLA (PLA-HF) Bambu invented this category with the X1 / A1 launch. The chemistry: lower-molecular-weight PLA blends, plasticisers (PEG or citrate esters), and nucleating agents that crystallise fast on the way out of the nozzle. Result: viscosity drops dramatically. The same hotend can push 24-32 mm³/s instead of 12-18. Practical settings: 200-230°C nozzle (hotter for higher speed; H2D forum threads suggest 230°C for 600 mm/s inner walls), 60°C bed, 100% fan. Outer wall 250-300 mm/s, inner walls 400-600 mm/s on high-flow hotends, infill 600+ mm/s. The catch: high-speed PLA needs a high-flow hotend. Bambu hotend, Revo HF, CHT clones, Volcano. Stock V6 won't hit the rated speed; you'll under-extrude. Pressure advance often runs slightly lower than standard PLA (0.020-0.035) because it flows so easily. Brands: Bambu PLA Basic [truncated] --- # Section: Troubleshooting ## OrcaSlicer Settings Not Saving: Diagnose the Profile Reset Bug URL: https://orcaslicer.net/orcaslicer-settings-not-saving/ I spent three weeks dialing in a custom Voron filament profile, then swapped the printer dropdown to a friend's Prusa for a test print and watched my 235C nozzle temp, my retraction tweaks, and a per-object color assignment evaporate in the time the UI took to redraw. My first instinct was the one every angry GitHub issue starts with: OrcaSlicer ate my settings. It hadn't. The profile was sitting safely in user/<number>/ on disk. The printer switch had just hidden it from the dropdown because the new machine's compatibility filter didn't match my filament's printer condition. That moment, plus a weekend reading every "settings not saving" issue on the tracker, convinced me of one thing. Almost every complaint with that title lands in one of four buckets, and only the fourth is an actual bug. The other three are the three-tier preset system working exactly as designed, in ways the UI doesn't always make obvious. If you're here because something you saved looks gone, this guide gets you to the right diagnosis in under sixty seconds. Table of contents The four-bucket diagnostic flow System vs user presets: the read-only rule The save procedure that actually works The "unsaved changes" dialog and Transfer vs Discard Switching printers wipes filament and process picks Compatibility conditions and the disappearing preset trap The known issues: #2278, #2705, and the historical #980 The Flatpak app ID migration trap in v2.3.2 Where presets actually live on disk Backups, recovery, and why there's no Reset button FAQ The four-bucket diagnostic flow Before you reformat anything or start reinstalling, run through this short decision tree. I've watched it close out 90 percent of the "settings not saving" panic I've personally hit, and it maps almost one for one to what the maintainers see on the issue tracker. Question 1. When you hit save on the modified value, did a small name-entry dialog pop up that you closed or cancelled? If yes, you were editing a read-only system preset. The dialog was Orca asking you to name a new user preset, because the bundled vendor one cannot be overwritten. Hitting cancel discards the edit. This is bucket one and it's by far the most common cause. Question 2. Did the symptom appear right after you changed the machine in the printer dropdown? If yes, you hit Issue #2705. Orca re-evaluated which filament and process presets are compatible with the new printer, and yours got swapped out. Your data is fine. You just need to mark the relevant filament and process presets as compatible with both printers before switching, or save them as user presets first. Question 3. Did a specific custom filament or process simply vanish from its dropdown, even though the file is still on disk? That's a compatibility condition filter doing its job, and in some cases it's the documented JSON typo bug in Issue #12193 where Orca writes the key compatible_prints instead of compatible_printers and the preset becomes invalid. Question 4. Did a genuinely-saved user preset disappear after a clean restart, with no printer switch and no missed dialog? That's the small, real bug bucket: Issue #980 from the 1.6.2-beta era, the AppData-leftover workaround documented in Discussion #1395, and the unresolved edge case in Issue #7009 where the save flow stores the diff against the parent instead of as a new preset. Workarounds exist for all three. The floppy disk save icon next to a modified preset name System vs user presets: the read-only rule If you read nothing else here, read this. The biggest source of "settings not saving" complaints isn't a bug. OrcaSlicer inherits a three-tier preset model from PrusaSlicer (which inherited it from Slic3r), and one tier is bundled, read-only, and vendor-provided. You cannot write to it. Ever. The DeepWiki presets page spells it out: "System presets: Read-only, vendor-provided, Always visible if compatible" and "User presets: Always visible". The underlying source flags every preset with an is_system boolean in Preset.hpp, and the save path branches on it. If is_system is true, Orca won't let you overwrite. Instead, the moment you modify any value and click the floppy disk save icon, it opens a name dialog and stores your edit as a fresh user preset in user/<number>/ on disk. The Obico profile management guide describes the same flow: click the "Save current filament" button, then "Save as new user preset". If you've ever hit cancel on that prompt because you "just wanted to change one number," you've encountered the bug that isn't a bug. Your edit lived only in memory and dies the moment Orca closes. For a deeper read on how the three tiers fit together, including which level overrides which when settings conflict, I'd point you at our companion piece on the three-tier preset model explained. The short version: Process sits at the top, Filament sits above Process in the merge order for any per-filament keys, and Printer sits underneath as the base. The save procedure that actually works Here's the flow I use now, after losing the same Voron profile twice in a week. It's boring on purpose. Change the value you want to change in the right-hand settings panel. Look at the preset name in the dropdown. You should see a small modified indicator next to it. Click the floppy disk save icon next to the preset name (it's the small disk-shaped button right next to the dropdown). When the name dialog appears, type a clear, version-tagged name. I use things like PETG-235-v3 so I can tell my iterations apart later. Confirm. The preset now appears in the dropdown as a user preset. Exit Orca completely. Relaunch. Verify the preset is still in the dropdown before you do any more work. Step 6 sounds paranoid and it is. It's the workaround for Issue #7009, where on certain custom-printer setups the save flow has been observed to drop the new preset and instead store the modifications as "unsaved changes" against the parent. Reopen catches it before you spend another hour on a profile that wasn't actually written. Save as new user preset name dialog The "unsaved changes" dialog and Transfer vs Discard Sooner or later you'll hit a second dialog, and this one trips up everybody, including the person who filed Issue #4422. It pops when you try to switch to a different preset while the current one has uncommitted modifications. The text reads: "Would you like to keep these changed settings (new value) after switching presets?" The two buttons are labeled Transfer and Discard. The reporter of #4422 says they're confusing and suggests "Use New" and "Keep Old" would be clearer. The issue is closed without a label change, so you have to memorize them. Transfer carries your in-flight modifications over to the destination preset. Discard abandons them. If you don't know which to pick, hit Discard and re-edit on the destination preset deliberately. Transfer can quietly stack changes onto a preset you didn't mean to modify, and you'll spend ten minutes later wondering why your PLA profile suddenly has a 60mm/s outer wall speed that belongs to your PETG settings. This dialog also fires any time you switch printers and the compatible-filament list changes. The unsaved changes dialog showing Transfer and Discard buttons Switching printers wipes filament and process picks: Issue #2705 This is the one that ate my Voron tuning. Issue #2705, titled "Switching Machine Preset Changes Filament and Process Presets," documents the exact reproduction: "Create and save a custom machine preset" "Start a new project" "Create a custom process preset" "Select multiple color filament presets" "Switch to a different machine preset" What you observe, quoted from the issue: "When you select a different machine preset, the process preset and the filament presets will change, erasing any part level multi-color selections." The report was filed against v1.8.0-rc2 on Windows 10, labeled bug, and the related PR #4930 is closed. The behavior still happens because it's not [truncated] --- ## Elephant Foot in OrcaSlicer: Compensation Done Right (2026 Guide) URL: https://orcaslicer.net/orcaslicer-elephant-foot/ The first time I really noticed it, I was holding a black ABS enclosure I'd printed overnight, and the base flared out a clean half-millimeter past the design. The model was 40.00 mm in Fusion, but my calipers read 40.6 at the base and 39.98 ten layers up. After a few weekends of changing one variable at a time, I finally figured out why OrcaSlicer's compensation setting fixed it on PLA but barely dented it on PETG. If you're here, you've probably seen the same thing: a print that looks fine from chest height but obviously wrong under a desk lamp, with a soft outward swell on the first one to three layers. This guide walks through what's actually happening, the OrcaSlicer settings that matter, per-filament starting values, and two non-slicer fixes that beat compensation on the parts they suit. Table of contents What elephant foot actually is (and what it isn't) Why it happens, over-squish meets bed heat The OrcaSlicer setting: elefant_foot_compensation Companion settings: compensation layers and layer density Per-filament starting values (community ranges) First-layer flow vs walls: the bottom_solid_infill_flow_ratio trap Z-offset and ABL: why ABL is not a fix The CAD chamfer-bottom alternative Why brim, raft, and skirt don't fix it Troubleshooting checklist and FAQ What elephant foot actually is (and what it isn't) Elephant foot is a visible outward bulge on the first one to three layers of a print. The base comes out wider than the rest of the geometry, and from the side the silhouette looks like a flared boot. It's a first-layer geometry defect, and the giveaway is that calipers read oversize only near the build plate while upper layers measure correctly. I want to draw a line between this and plain old first-layer over-squish, because the two get blamed for each other constantly. Over-squish is when the entire first layer is mashed flat by a nozzle set too low, and you'll feel it as a glossy surface on the bottom of the part. Elephant foot is the geometric consequence of that over-squish (and a hot bed), but they aren't the same symptom. You can have a perfectly textured first layer and still get elephant foot on PETG because of bed heat alone. I've seen it on a freshly leveled Bambu P1S. Side-by-side photo of a 3D-printed cube with visible elephant-foot bulge at the base vs. a clean-edged cube after compensation There are three legitimate ways to address it, and they don't all live inside the slicer. You can shrink the first layer in software (OrcaSlicer's compensation setting), you can chamfer the bottom of the model in CAD so the bulge fills the chamfer rather than protruding, and you can dial in your Z-offset and bed temperature so the bulge never forms. A few popular suggestions (brim, raft, auto bed leveling) don't actually touch the problem. I'll get to why later. Why it happens, over-squish meets bed heat There are two physical causes, and they multiply each other. The mechanical cause is straightforward: if your nozzle is closer to the bed than the configured first-layer height, every extruded line gets squished wider than its set extrusion width. A 0.42 mm line on a 0.4 mm nozzle can spread to 0.55 mm or more if the gap is wrong, and the part comes out measurably oversize at the bottom. The thermal cause is what makes PETG and ABS so much worse than PLA. A hot bed keeps the bottom layers semi-molten for longer than it takes to print the next few layers above, and the weight of all that subsequent plastic squeezes the soft base outward. Siraya Tech sums it up cleanly in their elephant foot writeup: too much bed heat, a nozzle that's too close to the build plate, the weight of the print pressing down, and poor cooling at the bottom. Their standard bed temperature ranges line up neatly with how often each material flares: FilamentBed temperature range PLA50-60 C PETG70-80 C ABS80-100 C TPU40-50 C PLA cools fast enough at 55-60 C that the base sets before the weight of the upper layers can do real damage. PETG and ABS stay tacky for much longer at 70-100 C, so even a slightly low nozzle turns into a visible bulge. That's why your PLA prints can look perfect on the same printer where your ABS bottoms flare. It isn't a calibration regression. It's the material. Knowing this two-cause model matters because it changes what you reach for first. On PLA, dial Z-offset. On PETG and ABS, you'll likely need both Z-offset and slicer compensation, because even a perfect nozzle gap can't fight that much heat soak. The OrcaSlicer calibration guide is the right place to start before touching the compensation slider. The OrcaSlicer setting: elefant_foot_compensation OrcaSlicer's primary lever for this is a setting called Elephant foot compensation, which lives in the Quality tab under the Precision section. The UI label uses the correct English spelling. The internal config key, however, is elefant_foot_compensation, spelled with one fewer "e" than the English word. That isn't a typo in this article. It's a heritage quirk inherited from PrusaSlicer's source code, and you'll see it preserved verbatim in OrcaSlicer's PrintConfig.cpp. If you're editing JSON profiles or grepping logs, search for "elefant" or you'll come up empty. Here are the verbatim particulars from the source: Label: Elephant foot compensation Default value: 0.0 mm Minimum: 0 Tooltip: "Shrinks the first layer on build plate to compensate for elephant foot effect." What it does is shrink the first-layer perimeter inward by the value you set. If you enter 0.15 mm, every outer wall on the first layer is offset 0.15 mm toward the inside, so when that 0.15 mm of bulge forms during printing, the bulged edge lands roughly where the design wall was supposed to be. The 0.0 mm default means no compensation, which makes sense for a generic profile that can't assume your bed temperature, plate type, or filament. Screenshot of OrcaSlicer Quality tab with the Elephant foot compensation field highlighted Two important caveats. First, OrcaSlicer didn't invent this feature. The setting is inherited wholesale from PrusaSlicer, which is why the spelling quirk persists. PrusaSlicer's knowledge base notes that values around 0.2 mm usually work well for the default 0.4 mm nozzle, and that official Prusa profiles have this setting turned on by default. OrcaSlicer's default is 0.0 mm; Bambu's stock profiles override that for some printers, which I'll get to in the per-filament section. Second, this setting has a known limitation flagged in OrcaSlicer Issue #6871 (open as of writing). If the bottom of your part has a chamfer or radius so the second layer overhangs the first, aggressive compensation can shrink the first layer enough that the second layer's outer line gets printed in the air with nothing under it. That shows up as a stringy, droopy second-layer edge. For flat-bottomed parts this never bites, but on something like a chess pawn with a curved foot, it's a real trade-off. Companion settings: compensation layers and layer density There are two other settings that sit next to the main compensation field, and most beginners ignore them. They control how the compensation tapers out and what happens to the solid infill on the affected layers. The first is elefant_foot_compensation_layers. Its defaults from the source: Label: Elephant foot compensation layers Default value: 1 Minimum: 1 Tooltip: "The number of layers on which the elephant foot compensation will be active. The first layer will be shrunk by the elephant foot compensation value, then the next layers will be linearly shrunk less, up to the layer indicated by this value." If you set compensation to 0.15 mm and layers to 4, the first layer shrinks by 0.15 mm, the second by 0.11 mm, the third by 0.08 mm, the fourth by 0.04 mm, and from layer five onward the geometry matches the design exactly. That linear taper smooths the transition from the inset first layer back to full size, which matters because a sharp single-layer step can be visible on shin [truncated] --- ## OrcaSlicer Won't Open or Won't Launch: Windows, Mac, Linux Fixes URL: https://orcaslicer.net/orcaslicer-wont-open/ I double-clicked the OrcaSlicer shortcut on my Windows 11 box last month, watched the splash flicker for half a second, and got nothing. No window, no error, no Task Manager process. I've now hit a launch failure on every desktop OS I own, and I've learned that "OrcaSlicer won't open" is almost never one bug. It's a small family of bugs that look identical from the outside but need very different fixes underneath. The good news: most of the time you can get back into the slicer in under fifteen minutes once you know which symptom matches which cause. This guide walks the three OSes in order of how loud the issues are in the SoftFever/OrcaSlicer repo right now, and it leans hard on the actual GitHub issue threads instead of recycled forum guesses. If you're stuck at a damaged-app dialog on a Mac or a vanishing icon on Windows, you're in the right place. Table of contents What's happening when OrcaSlicer won't open Windows: the Defender false positive on v2.3.2 Windows: missing Visual C++ runtime fix Windows 11 slow launch and blank home screen (Issue #13239) Windows: corrupted profile reset done safely macOS: the "App is damaged" Gatekeeper fix macOS: Apple Silicon, Rosetta, and the right build Linux: AppImage permissions and FUSE mismatches Linux: when the Flatpak build saves you OS version minimums and graphics requirements Crash logs per OS and FAQ What's happening when OrcaSlicer won't open Launch failures fall into three buckets, and figuring out which one you're in saves you from chasing the wrong fix. The first bucket is the splash hang, where you see the OrcaSlicer logo for anywhere from a few seconds to ninety seconds and then either nothing renders or a blank home tab appears. The second is the vanishing window, where the app shows in Task Manager (or Activity Monitor) for five to ten seconds, then disappears with no dialog at all. The third is the explicit error: a "damaged" warning on macOS, a missing DLL message on Windows, or a shared-library error on Linux. Match your symptom to one of those buckets and skip straight to the relevant section. If you've got a hard error message on screen, I'd suggest copying the verbatim text and searching it on the issue tracker before you do anything else. Half the launch bugs I've debugged were already documented with the exact string in a thread that took thirty seconds to find. For a wider tour of issues that aren't strictly launch-related, see the full troubleshooting hub at our OrcaSlicer troubleshooting master guide. OrcaSlicer 2.3.2 installer blocked by Windows Defender with Wacatac.H!ml detection Windows: the Defender false positive on v2.3.2 (Wacatac.H!ml) If Defender ate your installer before OrcaSlicer ever got a chance to launch, you're looking at Issue #12923. The detection name is the giveaway: Trojan:Win32/Wacatac.H!ml. That !ml suffix is Microsoft's machine-learning heuristic flag, not a signature match against actual malware. It fires when the ML model sees a packing or behavior pattern it doesn't like, and it has hit GitHub CLI, Docker Desktop, nw.js, and the .NET runtime over the past couple of years, all of which were confirmed false positives in their own trackers. That's not me waving away a real malware report. The OrcaSlicer 2.3.2 installer comes straight from a signed GitHub release, and the cross-project pattern around the Wacatac heuristic is strong enough that I treat it the same way I treat a Defender hit on the GitHub CLI installer: re-download from the canonical source, verify it's the right release page, and add a temporary exclusion if Defender won't let you run it. Here's the path I use. Open Windows Security, click Virus & threat protection, scroll to Virus & threat protection settings, click Manage settings, scroll to Exclusions, and add the OrcaSlicer install folder (and the installer file itself if Defender already quarantined it). Then re-download the installer from the official OrcaSlicer GitHub releases page. There's no maintainer comment yet in #12923 confirming the false positive, so for the full context on the detection and how to report it back to Microsoft, see our breakdown of the Defender false positive. Windows: missing Visual C++ runtime fix When OrcaSlicer just disappears with no error, or you see a generic "VCRUNTIME140.dll not found" style dialog, the cause is usually the Microsoft Visual C++ Redistributable not being installed. The OrcaSlicer README references MicrosoftEdgeWebView2RuntimeInstallerX64 and a vcredist component as runtime dependencies; the README's vcredist note even mentions it's "already available on your computer if you've installed visual studio," which is the giveaway that a clean Windows install often lacks it. Don't chase a specific year (2019 vs 2022). Microsoft ships a unified v14 redistributable that covers everything from Visual Studio 2015 through 2022 in one installer, and that's what the app actually needs. Grab the x64 build directly from Microsoft's official VC++ download page, or use the permalink at https://aka.ms/vc14/vc_redist.x64.exe. Run it, reboot, then try OrcaSlicer again. While you're at it, check that WebView2 Runtime is installed too. Most modern Windows 11 systems have it preinstalled, but a stripped-down Windows 10 image or a "debloated" build sometimes doesn't. If you're rebuilding from scratch anyway, our fresh Windows install guide walks through both runtimes step by step. Windows 11 slow launch and blank home screen (Issue #13239) This one's the freshest pain point in the tracker and it's still unresolved. Issue #13239, opened in April 2026 and OPEN as of my last check, describes OrcaSlicer 2.3.2 taking around 90 seconds to launch on Windows 11 25H2 and then rendering an empty home and project view. The reporter's rig is an Intel i7-13700K, 32GB DDR5, and an RTX 4090 on driver 595.97. The most telling detail in the thread: NVIDIA's GeForce overlay treats OrcaSlicer as if it were a game, which suggests the GPU stack is doing something unusual during the slicer's window initialization. The reporter already tried driver updates, a clean reinstall, deleting the config folder, tweaking Windows graphics settings, disabling the NVIDIA overlay, and even switching the printer connection from LAN to USB. None of that worked. The one workaround mentioned in the thread is temporary, and I want to be clear that it does not actually fix the bug: "changing the language temporarily resolves the problem, though the issue returns upon closing and reopening the application." So you can use it to get into the slicer for one session, but you'll repeat the dance on every relaunch until SoftFever ships a real fix. If you're hit by this, I'd suggest two things. First, subscribe to #13239 on GitHub so you'll get notified when a maintainer responds or a patch lands. Second, try the language-switch workaround (Help, then change the language to anything else, then back) once per session as a stopgap. I haven't seen a permanent fix posted by anyone yet, and I'd rather flag that honestly than recommend the usual driver-roll snake oil. OrcaSlicer Help menu showing Show Configuration Folder option Windows: corrupted profile reset done safely A corrupted user profile is the cause of a surprising number of "OrcaSlicer won't open" reports. Issue #2117 is a clean example: on Windows 10, the app launched into a settings-conflict modal ("Spiral mode only works when wall loops is 1, support is disabled..." and so on), then became unresponsive on either Yes or No, then closed itself. That issue was filed under not-planned without a maintainer fix, but the workaround pattern is well-known: nuke the data folder and let OrcaSlicer rebuild it on first run. The path you're deleting is %APPDATA%\OrcaSlicer, which expands to C:\Users\<username>\AppData\Roaming\OrcaSlicer. Discussion #5781 in the tracker spells it out: "just delete it's data folder %appdata%/orcaslicer/". Close OrcaSlicer first (check Task Manager to be sure), paste the path [truncated] --- ## OrcaSlicer Crashing While Slicing? 8 Fixes for E-Cores, GPU, RAM URL: https://orcaslicer.net/orcaslicer-crashing-slicing/ The first time OrcaSlicer ate my afternoon, I'd clicked Slice Plate on a 600,000-triangle dragon and watched the window vanish without so much as an error dialog. No log popup, no Windows toast, nothing. I opened it again, loaded the same model, and watched it do the same trick. That was Issue #8227, and once I'd traced it to my i9's E-cores I started keeping a list. Different OS, different cause, but the same maddening pattern: you press the big button, OrcaSlicer dies, and the fix depends almost entirely on your platform. This walks through eight specific crash patterns I've debugged across Windows, macOS, and Linux, each tied to a tracked GitHub issue you can verify yourself. Table of contents Why OrcaSlicer crashes when you click Slice Plate First step: find your crash log Fix 1: Windows + Intel E-core crash on complex models (#8227) Fix 2: Linux 75% freeze + OOM (#9094) Fix 3: Windows 11 + 2.3.2 takes 90 seconds to open (#13239) Fix 4: Large-file GPU crash on import (#12142) Fix 5: macOS crashes on slice or launch Fix 6: Organic tree supports that look frozen (#9064) Fix 7: Re-slice slowdown after layer-height change (#9397) Fix 8: Windows Defender blocks the 2.3.2 installer (#12923) Diagnosing RAM without inventing minimums When to file a bug report FAQ Why OrcaSlicer crashes when you click Slice Plate Before you start chasing fixes, you need to know which failure mode you're actually looking at, because the three big categories don't share a root cause and they don't share a workaround. Hard crash vs freeze vs OOM A hard crash is the one I described in the intro: the OrcaSlicer window disappears, sometimes with a generic "OrcaSlicer.exe has stopped working" toast, sometimes without anything at all. On Windows that's overwhelmingly Intel hybrid-CPU E-core scheduling on dense models (Issue #8227). On 2.3.2, certain cylinder and torus geometries reproduce a similar hard crash per Issue #12906. A freeze looks different. The progress bar climbs, hits roughly 75 percent, and then the app locks up while RAM keeps climbing. On Linux that's the memory-leak signature from Issue #9094, eventually fixed via PR #9115. You won't see a crash dialog because the kernel's OOM killer shoots the process before OrcaSlicer can complain. That OOM is really just the tail end of a freeze: mouse stutters, other apps lag, then OrcaSlicer isn't there anymore. The fix for that one rarely lives inside OrcaSlicer's settings. OrcaSlicer 2.3.2 main UI showing the four tabs (Prepare, Preview, Device, Project) The four tabs and why "Performance" isn't one of them OrcaSlicer 2.3.x ships with four top-level tabs: Prepare, Preview, Device, Project. That's it. Tutorials telling you to open a Performance or Optimization tab are either about a different slicer or AI-generated. Settings affecting slice speed and stability live inside Prepare's Process panel. The GPU is for the preview, not the slicer This trips people up, including me at first. OrcaSlicer's GPU runs the preview viewport, not the slicing engine. None of the 2.3.x release notes document any GPU-accelerated slicing feature. When you see nvoglv64.dll in a crash log (like Issue #12142), it's failing during import or preview rendering, not slice computation. Updating GPU drivers helps viewport stability; it won't make slicing faster. First step: find your crash log Before you try any fix, grab your latest log. OrcaSlicer's own bug report template publishes the exact paths, and these are the only paths I'll cite because everything else floating around online is guessed. Windows %APPDATA%\OrcaSlicer\log, which expands to C:\Users\<your username>\AppData\Roaming\OrcaSlicer\log. Paste either into File Explorer's address bar. Most recent dated .log file is what you want. Windows File Explorer showing the OrcaSlicer log folder at AppData Roaming macOS $HOME/Library/Application Support/OrcaSlicer/log. Library is hidden, so in Finder press Cmd+Shift+G and paste the path. Linux $HOME/.config/OrcaSlicer/log. Toggle hidden files, or cd there from a terminal. The in-app shortcut If OrcaSlicer is still running (handy for the freeze pattern where the app is alive but unresponsive), use Help, then Show Configuration Folder. That opens the parent of the log folder, so you'll see profiles and the log directory side by side. I keep this open during any debugging session. For broader triage flow, the OrcaSlicer troubleshooting master guide covers the wider symptom map. Fix 1: Windows + Intel E-core crash on complex models (Issue #8227) This is the one that cost me my afternoon. Issue #8227 opened 28 January 2025 against OrcaSlicer 2.2.0 on Windows 11 Pro 23H2, Intel i9 13800K with 128GB DDR4 and an RTX 4090. Verbatim: "Orca Slicer will consistently trigger errors and crash when slicing complex models (>500000 triangles) with efficiency cores enabled." Exception code c0000005, a plain access violation. Confirm you have an Intel hybrid CPU The fix only applies on Windows with an Intel 12th-gen or newer hybrid CPU (the ones splitting Performance and Efficiency cores). AMD Ryzen, macOS, and Linux aren't affected. Press Win+R, type msinfo32, and check the Processor line. Anything like "i5-12600K," "i7-13700K," "i9-14900K," or "Core Ultra 7 155H" means the hybrid scheduler is in play. Workaround A: Task Manager Set Affinity This is what I use because it doesn't need a reboot. Open Task Manager, switch to the Details tab, scroll to orca-slicer.exe, right-click, and choose Set Affinity. Uncheck the boxes for your E-cores (on an i9-13900K that's CPUs 16 through 31; layouts vary). Leave only P-core threads ticked, click OK, slice again. Windows resets affinity every launch, so you'll redo this each time. There are scripted ways around that, including start /affinity in a batch shortcut or process managers like Process Lasso. Issue #8227 only documents the Task Manager path, but the alternatives work the same way under the hood. Windows Task Manager Set Affinity dialog with E-cores unchecked for orca-slicer.exe Workaround B: disable E-cores in BIOS Verbatim from the issue: "Disabling e-cores in the BIOS or setting a permanent affinity for Orca Slicer to avoid the e-cores mitigates the problem." If you don't care about losing E-core background efficiency, flip the option in your BIOS (on ASUS it's usually under Advanced or AI Tweaker) and the issue goes away permanently. I wouldn't do this on a daily-driver laptop, but it's a clean fix for a dedicated print box. Why this is not coming as a patch The issue is closed as not planned. The reporter's guess, "It seems there is some issue with regards to syncing distributed loads across P vs E cores, maybe it is a timing issue," lines up with how Windows' Thread Director hands work between the two core types. The maintainers aren't shipping a fix, so the workaround is yours to apply. Fix 2: Linux 75% freeze + OOM (Issue #9094) Issue #9094 was opened 26 March 2025 on Arch Linux, OrcaSlicer 2.3.0, AMD Ryzen 5 3600 with 31.3 GiB of RAM and a Radeon RX 6750 XT. Verbatim symptom: "Gets stuck at 75% slicing and leaks memory until kernel kills OrcaSlicer." That's about as clean a description of a memory leak as you'll see in a bug tracker. The signature If your slice climbs steadily, sticks around 70 to 80 percent, and your RAM usage keeps rising in htop or your system monitor, that's the pattern. The kernel's OOM killer eventually fires and OrcaSlicer disappears without any in-app error. Update past 2.3.0 The fix landed in PR #9115 and shipped in 2.3.1, so confirm you're not still on 2.3.0. Help, then About shows your version. If you're behind, grab the latest from the official GitHub releases page. PR #9115 didn't fix every Linux slicing crash; it was specific to this leak. But it does kill the 75 percent signature for the model class that triggered #9094. When updating doesn't help If you're on 2.3.1 or 2.3.2 and still hitting it, the surprise about #9094 is that it reproduced on a 31.3 GiB system, so ad [truncated] --- ## OrcaSlicer Print Stops Mid-Print: Causes and Recovery Guide URL: https://orcaslicer.net/orcaslicer-print-stops-midprint/ An eight-hour print stopped at hour six. I came back to the workshop expecting to peel a finished benchy-cluster off the plate and instead found the toolhead parked over the front of the bed, the hotend cold, and a "Paused" notification glowing on the LCD. No error code. No filament jam. The spool still had three quarters of a kilo of PLA on it. I had no idea what halted the print, and I burned forty minutes scrolling through logs before I figured it out. It was me. Or more accurately, it was a right-click I'd made in OrcaSlicer's vertical layer slider three days earlier on a different project, and the pause marker had quietly survived into the sliced file I sent that morning. That's the thing about mid-print stops: they look like firmware bugs, hardware failures, or cosmic rays, and most of the time they're a setting you (or your slicer) inserted on purpose, then forgot about. This guide walks the four layers where a mid-print halt can originate, in the triage order you should actually use. If you're in the middle of a stopped print right now, jump to the triage section first and come back for the deeper stuff later. What's in this guide Why prints stop mid-job: the four layers that can halt motion Triage in 60 seconds: read the printer screen first OrcaSlicer-side causes and how to verify in the sliced gcode Filament runout: how it actually works under the hood M600 filament change: the deliberate stop Thermal runaway and other firmware aborts Probe and re-home failures mid-print Host and network drops: the silent killers Recovering without restarting: pause-and-resume the right way Skip a failed object instead of aborting the whole plate Bambu Lab printers: specific stop and recovery paths Pre-flight checklist to stop mid-print stops happening at all When to stop and restart (and when not to) FAQ Why prints stop mid-job: the four layers that can halt motion A mid-print stop feels like one event, but it's almost always one of four systems doing exactly what it was told to do. Name the layer that halted motion and you can fix the right thing instead of swapping parts or reinstalling firmware. The slicer (OrcaSlicer) inserts halt instructions into the gcode at slice time. M600, M0, M1, M25, custom pause macros, manual filament change markers, post-processing scripts. They're you, your profile, or a script doing what you configured. The host (your PC running OrcaSlicer, a Pi running Mainsail or Fluidd, a Bambu cloud session, an OctoPrint instance) streams gcode to the printer. If the host crashes or the link drops, the printer stops moving. The firmware (Marlin, Klipper, Bambu's stack, Prusa-Firmware) can decide on its own to halt. Thermal-runaway protection trips. M112 fires. Klipper logs an MCU shutdown. The firmware is protecting itself. The sensors and hardware (filament runout switch, AMS feeder, BLTouch, inductive probe, X1 chamber camera) can pull the cord. The firmware translates a sensor event into a halt, but the trigger is physical. Every section below ladders back to one of those four layers. The first question when diagnosing is not "what broke" but "which layer fired first". Slicer-side halts OrcaSlicer can put halt instructions into your gcode in five places. The right-click menu on the vertical layer slider in the Preview tab inserts pauses, change-filament markers, and custom gcode at a specific Z. The Change Filament G-Code field at Printer Settings > Machine G-Code writes whatever macro you've configured (usually M600 for Marlin, a PAUSE macro for Klipper). The Manual Filament Change toggle at Printer Settings > Multimaterial injects a pause at marked layers. Custom Before/After Layer Change G-Code fields run at every layer transition. And post-processing scripts under Process > Output Options can rewrite the file on its way out. OrcaSlicer's GitHub Discussion #477 spells out how the vertical slider menu inserts pause and M600 markers. Issue #6334 is the still-open enhancement request for adding pause on the horizontal (within-layer) slider, which currently does not support it. If someone told you "just pause anywhere in a layer in OrcaSlicer", that's wrong for 2.3.x stable. Host-side halts If the host is streaming gcode line-by-line over USB or LAN and the host crashes, the printer stops because no more lines are coming. Rare for Bambu users (the printer runs the file off its own storage) and rare for Klipper users with Mainsail or Fluidd. Not rare for OrcaSlicer users sending prints to a Marlin printer via the slicer's send-to-printer feature, especially after 2.3.0. Crash reports under Issues #9081, #9026, #11814, and #7580 describe OrcaSlicer crashing on connection or send. Fix: update to the latest 2.3.x, or better, never stream from PC for long prints. Firmware-side halts The firmware aborts on its own when something exceeds a safety threshold. On Marlin, that's thermal-runaway protection (temperature drifts further than the configured hysteresis for longer than the configured period, with Marlin upstream defaults around 4 degrees C and 40 seconds, though vendor forks change these). On Klipper, it's an MCU shutdown logged in klippy.log, triggered by a missed timer, a thermistor out of range, a stepper driver overheating, or a host-MCU timeout. M112 is the explicit emergency-stop command. Per marlinfw.org/docs/gcode/M112.html, "M112 is NOT a safety-rated 'Emergency Stop' in the formal sense"; it's a control stop that requires EMERGENCY_PARSER for instantaneous response. Sensor and hardware halts Physical inputs. A filament runout switch trips. An AMS feeder reports an empty slot. A BLTouch fails to deploy or retract during a mid-print probe (rare, only if a macro forced a re-home). On Bambu X1, the chamber camera plus AI model flags a high-confidence spaghetti event. The A1 doesn't have this feature because it's an open-frame printer without a chamber camera, so don't blame "AI failure detection" if your A1 stopped on its own. OrcaSlicer's right-click menu on the vertical layer slider showing Add Pause and Add Change Filament options Triage in 60 seconds: read the printer screen first The biggest mistake I see in 3D-printing forums is people opening the slicer first to debug a stopped print. The slicer wasn't running when the print stopped. The printer was. Read the printer's screen, or the web UI talking to the printer, before anything else. Marlin LCD common messages Thermal Runaway / Heating Failed: the hotend or bed temperature didn't track the setpoint within the configured window. The print is halted and heaters are off. Power-cycle to clear, but do not just retry. Marlin Thermal Runaway is almost always a hardware issue (loose thermistor, fan blowing on the heater block, a heater cartridge slipping out), not a false alarm. Probing Failed: a Z probe didn't deploy, retract, or trigger. Far more common at start-of-print than mid-print. If you see it mid-print, look for a custom layer-change macro that called G28 or G29. Print Paused / Filament Change: M600 or M0/M1/M25 ran. Almost always an inserted pause or a runout sensor firing. Not a firmware fault. Vendor LCDs (Creality, Anycubic, Elegoo) translate the same Marlin messages, sometimes badly. The underlying meaning is identical. Klipper-driven UIs (Mainsail and Fluidd) Look at the top banner. A red bar with "MCU 'mcu' shutdown: Timer too close" or similar is a firmware abort. You cannot resume; restart. A yellow banner or "Paused" state means a PAUSE macro ran. You can usually resume from the UI button after fixing the trigger. If neither shows up but streaming stopped, the host (Pi or x86) may have died. The file you want is klippy.log, usually at ~/printer_data/logs/klippy.log. The last hundred lines almost always include the trigger. Mainsail has a Logs section under Configuration; Fluidd has a similar panel. The Klipper community asks newcomers to upload klippy.log to a paste service when reporting a halt because the file contains the answer 90% of the tim [truncated] --- ## OrcaSlicer Top Surface Quality Issues: Pillowing, Gaps, Rough Tops URL: https://orcaslicer.net/orcaslicer-top-surface-quality/ The first time I noticed pillowing on a print I cared about, it was a Bambu X1C running 0.2 mm layers with 15% gyroid infill, default Bambu Studio profile imported into OrcaSlicer 2.3.0. The model was a desk organiser my partner had asked for, in a colour I didn't have a second roll of. I pulled it off the bed, ran my thumb across the top, and felt little raised lumps in a perfect grid pattern, like braille spelling out "you didn't set enough top layers." Three solid layers had not been enough to bridge the gyroid cells underneath. I had to print it twice. That single print is why I have strong opinions about top surfaces, and why I keep coming back to OrcaSlicer's settings menu instead of trusting a vendor profile. The slicer has more top-surface-relevant levers than any free option I've used, and most of them sit unexplored in tabs hobbyists never open. If your tops are pillowing, ghosting, gapping, scarring, or coming out of an ironing pass looking worse than before, the fix is almost always inside Quality, Strength, or the per-filament cooling panel. You don't need to switch slicers. You need to know which six or seven settings do the work. I'm writing this on OrcaSlicer 2.3.0 stable, with 2.3.2 in nightly RC as of this week. A few of the most useful top-surface improvements (Fixed Ironing Angle, per-filament ironing, bridge density above 100%) only landed in 2.3.2-rc2, so I'll flag those clearly. The 2.3.0 advice still works on the stable build. What's in this guide What "bad tops" actually means, five distinct defects How OrcaSlicer handles top surfaces in 2.3.x Fixing pillowing Fixing infill telegraphing Closing gaps between top infill lines Fixing rough, scarred, or zit-covered tops Ironing without ruining your top A 30-minute top-surface tuning workflow Filament-specific notes FAQ and where to download OrcaSlicer 2.3.2 A pillowed top surface on a small PLA print, viewed under raking desk lamp light so the bumps cast visible shadows. What "bad tops" actually means, five distinct defects Most posts about "rough top layer" lump every defect together. They aren't the same problem, and the fix is different for each. Before changing a single setting, spend five seconds identifying which of the five you're actually looking at. I usually tilt the print under a desk lamp so the light grazes across the top, throwing every height variation into shadow. Phone flashlights work too if you hold them low and to one side. Pillowing, the lifted blob between infill cells Pillowing looks like rows of small domes or dimples spread across the top, often matching the underlying sparse infill (gyroid blobs, grid intersections, crosshatch nodes). You sometimes see hairline splits at the centre of each dome where the skin has cracked open. The print otherwise looks finished, but the surface feels lumpy. The physics is straightforward. The top skin has to bridge each open cell of the sparse infill below. Where the strand crosses air, only surface tension and rapid cooling stop it from sagging. If the air on that strand is too warm or there's too little of it, the plastic stays soft, shrinks as it cools, and curls upward. The next pass smears across the curl. The result is a lifted blob. The cause is the same on Bambu, Prusa, and Creality printers because it's thermal rather than slicer-specific. Infill telegraphing, the lattice ghost in your top Telegraphing is the cousin of pillowing that looks subtler at a distance and worse up close. The top sealed surface looks closed and smooth from one angle, but rotate it under light and you see ridges, dimples, or a lattice ghost where the sparse infill below has shown through. Run a fingernail across it and you can feel the height differential, even though there are no visible holes or splits. This is also a bridging problem with a different failure mode. The strand cools faster over the void than over the rib of infill, so it sits slightly lower over the void and slightly proud over the rib. The remaining differential becomes visible under raking light. The fix is more top shell layers, denser infill, or both. Gaps between top infill lines, slim parallel slits This one is easy to spot. It produces visible parallel slits on the top, often running the full length of the surface in the direction of the top infill extrusion. The slits are usually narrower than a hair and you can sometimes see through them to the layer below. They're not the rectangular voids of infill telegraphing; they're the missing space between two extrusions that didn't merge. The root cause is under-flow, not insufficient layers. The top surface line width is set too narrow for the spacing the slicer chose, or surface flow ratio sits below what the printer needs to bridge adjacent lines into a continuous skin. There's also a small family of OrcaSlicer bugs that can leave thin slits even when flow looks right (covered below). Rough, scarred, or zit-covered tops This is the catch-all bucket for tops that look like the printer wandered while working. You see visible toolpath lines as ridges, dots where the nozzle pressed material as it changed direction, alternating diagonal stripes from the ironing pattern flipping every layer, or specks of plastic where a retraction left a scar. These are flow-and-motion problems rather than thermal ones. Pressure advance is off, top surface speed doesn't match outer wall speed, the Z-seam is landing on a visible face, the wall-loop direction is forcing a sharp corner where there shouldn't be one, or small-area flow compensation is over-correcting. Each has its own fix; they're rarely solved by one master setting. Ironing artefacts, banding, melt-back, blobs If you turned ironing on hoping it would smooth a rough top and the result looks worse, you're seeing one of three classic ironing failures. Banding means the surface has visible stripes where each ironing pass left its own ridge. Melt-back means the iron pass dragged extra material into a valley that wasn't there before. Blobs at the start of each ironing line mean the nozzle dribbled while waiting to start the next pass. Ironing is a slow, low-flow pass that smooths the top by re-melting the surface and dragging a small amount of additional plastic across it. The settings are unforgiving. Too much flow and you get blobs. Too little and the pattern stays visible. The wrong angle and you get tiger stripes. Line width subsection, top surface field highlighted at 100-105% range." class="wp-image" loading="lazy" /> OrcaSlicer's Quality tab open with the top surface line width field visible in the line width section. How OrcaSlicer handles top surfaces in 2.3.x OrcaSlicer scatters its top-surface controls across four tabs, which is the first reason new users miss half of them. Once you know where to look, the layout is logical. Where the settings live The Quality tab holds line width, ironing, wall and surfaces (one-wall-on-top, wall loop direction, avoid crossing walls), and small-area flow compensation. This is where most cosmetic decisions are made. The Strength tab carries the structural ones: top shell layers, top shell thickness, top surface density, top surface pattern, top-bottom infill wall overlap, and the sparse infill choices that telegraph into the top. The Speed tab has top surface speed and ironing speed. The Material (filament) tab carries all the cooling settings, including "Keep fan always on", the slow-down-for-layer-cooling toggle, and the per-feature fan overrides. If your defect is thermal (pillowing, splitting, anything that gets worse on tall thin parts), start in Material. If it's structural (telegraphing, gaps, voids), start in Strength. If it's cosmetic (pattern lines, zits, ironing bands), start in Quality. Speed is usually a finishing tweak, though top surface speed is one of the few speeds I'd set consciously rather than letting it inherit from outer wall. What changed in 2.3.2 vs 2.3.0 The 2.3.2 nightly carries the largest set of top-surface [truncated] --- ## OrcaSlicer Support Failures: Painting, Tree & Removal (2026) URL: https://orcaslicer.net/orcaslicer-support-failures/ The print was four hours and twelve minutes in when I heard the click. Not the soft tick of a cooling fan ramping up, the harder, drier click of a printhead clipping something it shouldn't have. I walked over to the printer, looked through the door, and there it was: a tree support branch lying flat on the bed like a knocked-over chess piece, and the nozzle calmly continuing to lay down the next overhang layer over a void that no longer had anything under it. The print was dead. Four hours of PETG, a 0.12 mm layer height, a 70 mm tall figurine, and the only thing I had to show for it was a snapped branch and a slumping chin where the chin used to live. I'm Jake Holloway. I've been printing FDM for more years than I care to date myself with, and I've used OrcaSlicer through the 1.x branch, the awkward jump to 2.0, and now the 2.3.x release line that landed in March 2025. Supports are the part of the slicer I argue with most. So this guide pairs every failure mode I've actually hit with the exact OrcaSlicer setting name that fixes it, points out where 2.3.x is genuinely better than 2.2 was, and stays honest about the bugs still open against 2.3.2-dev. The short version: there are paint-on supports, tree and organic supports, and removability levers like Top Z distance and the interface filament selector. Most failures come from one of those three categories being wrong, not from your printer being broken. What's in this guide Why OrcaSlicer supports fail more than they should The exact setting names that matter (and where they live) Paint-on supports: when auto-detection isn't enough Tree and organic supports in 2.3.x Why your supports won't come off cleanly A failure-mode-to-fix lookup A 10-minute support sanity check before you hit slice Where OrcaSlicer still gets supports wrong Where to download and which version to pick Why OrcaSlicer supports fail more than they should Supports look like one of those features that should be solved by now. Auto-detect the overhangs, drop pillars under them, print the part, snap them off. In practice every step of that sentence is wrong in its own way, and the failures cluster into three patterns that I've seen hundreds of times between my own prints and the GitHub issue tracker. The three failure categories The first is missed overhangs. You slice, the auto-detection misses something, and somewhere on layer 220 there's an unsupported lip the slicer didn't catch. Maybe the angle was exactly at threshold. Maybe the overhang was small enough that "ignore small overhangs" nuked it. Whatever the reason, the print finishes with a sagging tongue of plastic where there should have been a clean edge. The second is fragile trees. You pick organic or one of the tree styles, the slicer draws beautiful branching pillars in preview, and partway through the print one snaps off or gets bumped over by the nozzle on a travel move. From that moment every layer that depended on the branch is printing into thin air. The classic version of this is reported in issue #11456 against 2.3.2-dev, where the Z-hop fires after the nozzle has already travelled into the support body. I've also lived through the issue-#9229 version: a hollow trunk that isn't strong enough to take a sideways knock. The third is welded removal. The print finishes, looks great, you pull it off the bed, you try to snap the supports off, and they take a layer of skin with them. Sometimes you can hear the fibers tearing. Sometimes the support won't come off without pliers, and the down-facing surface ends up looking like a cheese grater. It's the failure mode users complain about loudest, and the one with the most levers to pull. If you're reading because of a specific symptom, jump to the symptom lookup near the end. A failed print with welded support scarring on a down-facing surface Which OrcaSlicer version you're actually on, and why it matters Supports changed materially between 2.2 and 2.3.x, and the version you're running determines what advice applies. OrcaSlicer 2.3.0 stable shipped on 20 March 2025, and 2.3.2 stable followed on 23 March 2025. Both releases sit on top of a foundational support PR that's worth knowing about by number, because it explains why some of the older troubleshooting articles you'll find online are now misleading. That PR is #8212, "Port latest support code from BBS," merged on 26 February 2025 by Noisyfox. It brought in two big things from upstream Bambu Studio: vertical tree support painting, and an "accurate top z distance for tree support" commit that fixes a long-running gap between the value you typed and the gap that actually got printed. The features shipped in the 2.3.0 release line via PR #8212. Be careful how you read the 2.3.0 stable changelog itself; the final stable release notes don't list those features by name, just the post-RC bugfixes. That's a frequent source of confusion when someone reads the changelog and concludes 2.3.0 added nothing for supports. 2.3.2 then added one specific organic-supports change that matters if you've ever broken a tree branch mid-print. The 2.3.2 release candidate notes (1 March 2025) describe the change as enabling base infill patterns including rectilinear, rectilinear grid, and honeycomb for organic tree supports, "which were previously hollow-only." Two explicit limitations: lightning infill is unsupported because it crashes, and double walls remain hollow-only. Both warnings now appear as tooltips. 2.3.2 stable also added a "Hollow base pattern warning" for non-tree supports. So the first question with any support troubleshooting is which version you're on. Help, About in OrcaSlicer gives you the build hash. If you're on 2.2.x, the hollow-tree problem is real. If you're on 2.3.0 stable, you have vertical tree painting and accurate top Z distance but trees are still hollow. If you're on 2.3.2 or any 2.3.x nightly past 1 March 2025, you have the infill patterns. The exact setting names that matter (and where they live) OrcaSlicer's support settings are split across four panels: Support, Advanced, Tree, and Filament. Half of the troubleshooting I do for people in forums boils down to "look two tabs to the left, the setting you want is over there." So before we get into the failure modes, here's the map. Support Type vs Support Style The two top-of-panel dropdowns are Type and Style, and they don't mean the same thing. Type (internal ID support_type) controls whether supports are generated automatically, manually, or not at all. The options are Normal, Tree, Auto, and Manual. Style (support_style) controls the algorithm that decides how the supports are drawn, and the options are Grid, Snug, Organic, Tree Slim, Tree Strong, and Tree Hybrid. The reason this trips people up is that Type and Style cross-couple. If you pick Type = Normal and Style = Tree Slim, the slicer still draws a tree, because Style overrides the geometry choice. If you pick Type = Manual, the slicer only generates supports where you've painted them, regardless of the Style setting. The mental model I use is: Type answers "where do supports go?", Style answers "what shape are they?". Six Style options is a lot. The cleanest community ranking I've seen is from user Marek-Dvorny in OrcaSlicer Discussion #10609, who ranks them by removability and surface quality as Organic first, then Tree Slim, then Tree Strong, then Default (Normal/Grid), and Tree Hybrid last with "poorest contact layer quality." One user's opinion in an official Discussion, not the OrcaSlicer team's position, but it lines up with my experience. OrcaSlicer Support panel showing Type and Style dropdowns The four removability levers If supports are tearing skin off your model on removal, you have four direct levers and one indirect one. They all live in the Advanced tab, except Branch Density which lives in Tree. The first lever is Top Z distance (support_top_z_distance). This is the gap between the top of the support and the bottom of the overhang. Larger [truncated] --- ## OrcaSlicer Overhangs and Bridging Problems: 2026 Fix Guide URL: https://orcaslicer.net/orcaslicer-overhangs-bridging-problems/ I pulled a print off the bed last Tuesday, a small enclosure with a 50-degree sloped lid, and the underside looked like wet cotton candy that someone had hit with a hairdryer. Drooping, hairy, blotchy. The bridges across the cable channel were worse, a curtain of sagging strands that I could lift with a fingernail. I had already done the usual rounds: bumped the fan to 100%, slowed the outer wall to 60mm/s, swapped the nozzle. None of it touched the problem. The reason it didn't touch the problem is that OrcaSlicer doesn't decide an overhang or a bridge based on the things most generic blog posts tell you to tweak. It decides based on a percentage-of-line-width score that almost nobody talks about, and the fix lives in four different tabs, not one. This guide is the one I wish I'd had that night. It's an OrcaSlicer-native walkthrough of the actual 2.3.x setting names, the diagnostic flow from visible defect to the precise setting you need to change, and a printer-class tuning matrix at the end. I'm not going to tell you to "just add supports". If you wanted that answer you wouldn't be five paragraphs deep into a troubleshooting article. Jump to a section What "overhang" and "bridge" mean inside OrcaSlicer Diagnose the defect first, defect-to-setting map The OrcaSlicer overhang stack, every setting that matters The OrcaSlicer bridging stack, every setting that matters Cooling, the half of the fix that lives in the Material tab Running the built-in bridge flow calibration Printer-class tuning matrix Filament-by-filament adjustments Common mistakes to avoid When the slicer can't save you, model-side fixes Recalibration cadence What "overhang" and "bridge" mean inside OrcaSlicer (and why they're not the same) Most articles will tell you an overhang is anything past 45 degrees and a bridge is a horizontal span. That's roughly correct in the real world. It's not how the slicer reasons about it. OrcaSlicer's overhang detector classifies a wall segment by the percentage of its line width that has no material directly below it. The official wiki for Quality > Overhangs is explicit: the "Detect overhang wall" setting will "detect the overhang percentage relative to line width and use different speed to print." Not angle. Line-width percentage. This single fact is the reason "increase cooling" advice fails so often. If the slicer hasn't even classified the segment as an overhang, no overhang-specific setting fires. A bridge, by contrast, is a different beast: an unsupported span that connects two anchor points in the same layer. The slicer treats it as a separate object, with its own flow ratio, density, speed, and fan rules. The settings live in their own block under Quality > Bridging, and they don't share variables with the overhang stack. So when your bridge looks like a wet noodle and your overhang looks like a hairy slope, those are two different bugs with two different cures, even though both end up under "my underside looks bad" in a Google search. How OrcaSlicer's overhang detector actually scores a wall Picture a wall that's sitting on top of nothing for half its width. The line width is 0.42mm. Half of that, 0.21mm, is hanging out over thin air. The slicer scores that segment at 50% unsupported. That falls into the third bracket of the four overhang-speed buckets. We'll get into the buckets in a minute, but the point is: the angle of the slope didn't decide it. The amount of the next line that has no previous line to sit on did. This has a sneaky implication. If you change your line width (say you bumped extrusion width from 0.42 to 0.5 to print faster), the same physical slope now reads as a different percentage of unsupported line width. That's why the same model can print clean on one profile and sag on another with no other changes. I've watched this trip up users in the OrcaSlicer issue tracker more than once. Why a "bridge" can fail to be a bridge (and get printed as an overhang wall) A genuinely fun edge case: sometimes a bridge gets misclassified. Issue #2231 in the SoftFever repo documents "Bridges being sliced as Overhang wall." If the anchor geometry on either end of your span isn't clean (say the wall is too short, or the bridge ends mid-perimeter), the slicer can decide it's actually a steep overhang wall and apply overhang-wall treatment instead. That means bridge flow ratio doesn't apply. Bridge fan speed doesn't apply. You're getting plain wall behavior with overhang slowdown layered on top, which is almost never what you want for an unsupported span. If you see a "bridge" sagging hard and bridge-specific changes do nothing, this misclassification is one of the first things to test for. The fix is usually geometric on the model side rather than a slicer toggle. The Overhangs section under Quality settings in OrcaSlicer 2.3.x Diagnose the defect first, defect-to-setting map Before you change a single number, look at the print. The visible symptom tells you which half of the stack to start with. What you seeMost likely causeFirst setting to touchLives in Smooth drooping curve under a slopeCooling not aggressive enough on the overhangOverhangs and external bridges fan speedMaterial > Cooling Hairy, stringy underside on a slopeOverhang speed bracket too highOverhang speed 50-75% / 75-100%Speed > Overhang Sagging spans of plastic across a gapBridge flow too high, or bridge speed too lowBridge flow ratioQuality > Bridging Pebbly, blobby texture under a bridgeBridge flow too high (over-extrusion)Bridge flow ratio, try 0.9Quality > Bridging Curled-up corner tip on a sharp overhangCurled perimeters detection not triggeringSlowdown for curled perimetersSpeed > Overhang Gappy, starved-looking bridge linesBridge flow too low, or density too lowBridge density, then bridge flowQuality > Bridging Photo-style defect checklist Run this list before you touch anything. It saves an hour of randomly twiddling numbers. Is the defect on a slope or on a span? Slope means overhang stack. Span means bridge stack. They share cooling but nothing else. Is the surface smooth-and-drooping or hairy-and-curled? Smooth droop is a cooling problem. Hair and curl is a speed-bracket problem. Does the defect stop where the supports would be? If yes, the geometry needs supports, not slicer changes. We'll get to tree supports later. Is it on every print or only on this one? If only this one, you have a model-side issue (long unanchored span, sharp corner) before you have a slicer issue. Did anything change since the last good print? New filament spool, new nozzle, ambient temperature drop. Those rule out slicer changes entirely. Has bridge flow calibration been run for this exact filament? If not, run it before tuning anything else. The OrcaSlicer overhang stack, every setting that matters This is the master walkthrough of every overhang-related setting that ships in 2.3.x. I'm using the exact UI labels because the internal variable names occasionally differ and you'll want to be able to find these in the interface without guessing. Detect overhang wall (the master toggle) This is the first switch in the Quality > Overhangs panel. The wiki copy reads: "Detect the overhang percentage relative to line width and use different speed to print." When this is off, the four overhang-speed brackets never trigger. Your wall speed applies everywhere, no matter how steep the slope. There's almost no reason to turn it off, but I've seen profiles imported from older slicer presets where it landed disabled and the user spent a week wondering why their overhang speeds did nothing. Leave it on. Always. If a setting in the overhang stack isn't behaving, this is the first thing to verify. The four overhang-speed brackets (overhang_1_4 through overhang_4_4) The Speed > Overhang Speed page is one of the most under-read pages in the OrcaSlicer wiki. The verbatim quote: "This is the speed for various overhang degrees. Overhang degrees are expressed as a percentage o [truncated] --- ## OrcaSlicer Bed Adhesion Problems: Glue Strategies & Real Fixes (2026) URL: https://orcaslicer.net/orcaslicer-bed-adhesion-problems/ I lifted a fingernail-sized chunk of coating off my first smooth PEI plate four years ago, and I still keep the damaged sheet in a drawer as a reminder. It was a 200-gram PETG bracket, a clean first layer, a perfect print, and then I tried to pry the part off cold. The plate came up with the print. That was the day I learned the most expensive lesson in OrcaSlicer bed adhesion problems: the worst failures are not the ones that come loose, they are the ones that bond too well to the wrong surface. Since then I have run three printers daily (a Bambu P1S, an X1C, and a Voron 2.4) through every material I could justify buying, and I have logged more bed-adhesion failures than I care to admit. This guide is the one I wish I had then. It is not a generic "my print is not sticking" walkthrough, because that question hides at least five different failure modes, and the fix for each is different. If your first layer never lays down cleanly, you have a leveling and Z-offset problem, not an adhesion problem. If your first layer is gorgeous and then a corner peels at layer 40, you have warping. If your whole part pops off the bed as a slab in the middle of the print, that is what most of this article is about. And if your print finished but you cannot get it off without a chisel, you have over-adhesion, which is its own (smaller) section below. I have tried to be specific where it counts and humble where I should be. I am citing Bambu's official wiki, OrcaSlicer's GitHub wiki, MatterHackers' material guides, and Magigoo's own product docs rather than inventing numbers. When I say "do not put bare PETG on smooth PEI," I mean it, and I will explain why. Let's get into it. Table of Contents Diagnose your adhesion problem in 30 seconds Why a print starts well and then detaches Bed surfaces ranked by versatility Material-by-material adhesion strategy Glue strategies: which glue when OrcaSlicer settings that actually fix adhesion Cleaning and maintaining your build plate When over-adhesion is the real problem Per-printer notes (Bambu, Prusa, Creality, Voron, Qidi) Quick reference checklist FAQ Diagnose your adhesion problem in 30 seconds Before you reach for a glue stick or open OrcaSlicer's settings, figure out which problem you actually have. I have wasted whole afternoons cleaning a plate that was fine and dialing settings that were never the cause. The table below is the same triage I now run mentally before I do anything else. Read the symptom column, then go where it points. Symptom Real cause Where to read next First layer never lays down cleanly, lines do not squish, filament curls behind the nozzle Leveling and Z-offset, not adhesion First-layer problems guide First layer is perfect, corners lift around layer 20 to 50, eventually pop free Cooling-induced warping, mostly thermal physics Warping and lifting corners plus the mid-print section below Whole print pops off the bed as a slab, no visible warp on the underside Surface, glue, or maintenance failure This article (keep reading) Print finished but you cannot remove it without a chisel Over-adhesion (common: PETG on smooth PEI) This article, over-adhesion section Layers misalign mid print (visible shift in X or Y) Mechanical layer shift, belts or steppers Z wobble and banding for related mechanical issues Loud clicking from the extruder during the print Extruder skipping (heat creep, jam, or under-extrusion), not adhesion See your printer's extruder troubleshooting docs Your bed sheet has visible peeled spots or texture loss Worn surface, replace or re-coat Maintenance section If the table sent you elsewhere, follow the link. I would rather lose you to the right article than have you adjust settings that will not fix your actual issue. If you are still here because your part pops off the bed mid-print as a clean slab, or because your material does not stick at all, the rest of this guide is for you. If you want the bigger picture across every kind of failure, the OrcaSlicer troubleshooting guide is the parent page that links everything together. The triage I run before touching the slicer. If your symptom is on the wrong branch, you will spend hours fixing the wrong thing. Why a print starts well and then detaches Mid-print detachment is the failure mode that makes you want to throw the printer out the window. The first layer is perfect, the bed is hot, the part is laying down cleanly, and then four hours in you walk past the printer and see a sad lump of plastic being dragged around by the nozzle. There are five common reasons this happens, and they almost never travel alone. 1. Cooling contraction (warping). Every layer above the first cools and shrinks. The amount of shrinkage depends on the polymer: ABS and ASA are roughly 0.8 percent, polycarbonate around 0.7, PETG around 0.4, and PLA around 0.3. Those percentages sound small until you remember they are linear contraction across a 200 mm part, which is several millimeters of force concentrated at the corners. Once the cumulative pull exceeds the bond strength at a corner, that corner lifts, the failure propagates inward, and your print either curls visibly or pops off as a chunk. The fix is not glue, it is thermal management: enclosure, lower part-cooling fan speed, brim, and (for the worst materials) a chamber heater. The full warping playbook lives in the warping and lifting corners guide, which I will not duplicate here. 2. Bed temperature drift during long prints. Some printer mainboards cycle the bed heater in a way that lets the surface dip 5 to 10 degrees between cycles. On marginal adhesion, that is the entire difference between hold and fail. If you can switch your bed control loop to PID instead of bang-bang, do it. If you cannot, pre-heat the bed longer before the print starts, raise your setpoint 5 degrees, or insulate the underside of the bed (cork sheet is the cheap classic). 3. Cooling fans bouncing under the part. A handful of printer designs route the part cooling fan duct so the airflow deflects off the bed and back under one edge of the part. On a 6-hour ABS print, this is enough to chill one corner faster than the others and trigger asymmetric warping. Check the duct alignment, especially on Bambu A1, P1S, and any Voron with a single-fan setup. 4. High-lever-arm part geometry. A 100 mm tall part on a 30 mm square base is a class-one lever pulling on the corners. No surface in the world holds that without help. A brim is mandatory. Mouse ears (small circular tabs at the corners) help further. I have never regretted adding a brim to a tall narrow part, even when the surface was supposedly bulletproof. 5. Worn or stale plate. PEI loses tackiness over roughly 100 to 300 prints, depending on how aggressively you abuse it. The first-layer trace can still look fine to the naked eye while the actual bond strength has dropped by half. If you have replaced glue, cleaning method, settings, and even filament, and you still see detachment, replace the plate. Bambu PEI sheets are around 30 USD and a healthy plate solves a remarkable number of mysterious adhesion problems. There is a sixth cause that does not get talked about enough: re-leveling drift. If you have moved the printer recently, replaced the hotend, re-greased the bearings, or even just bumped the build plate while clearing a print, your stored auto-bed-level mesh might be slightly off. Re-run leveling. On Bambu printers that means triggering the manual full calibration (not just the bed mesh) from the Device tab. On Klipper machines, run BED_MESH_CALIBRATE fresh and save. Bed surfaces ranked by versatility If I could only own one build plate for the rest of my printing life, it would be a textured PEI. It is not the best surface for any single material, but it handles more materials acceptably than anything else on the market. Here is how I rank the common options for general 3D printing use. Textured PEI (Bambu Engineering Plate textured side, Bambu Textured PEI Plate, Wham Bam Sn [truncated] --- ## OrcaSlicer Z-Wobble & Z-Banding: Diagnose and Fix (2026) URL: https://orcaslicer.net/orcaslicer-z-wobble-z-banding/ The first Z-wobble print I ever diagnosed was a 200 mm vase in matte PLA on an Ender 3 V2. I'd just upgraded the hotend, leveled the bed twice, and run a fresh PID tune. The first layer was glass. The seam was clean. And running my thumb up the side of the print, I could feel a soft, regular ripple every eight millimetres. Like corduroy stretched over a glass bottle. I spent the next four hours convinced it was my PETG slicer profile. It wasn't. It was the leadscrew. The cheap, slightly bent leadscrew Creality had shipped with the printer. I'm telling you this because almost every Z-defect article I can find online treats Z-wobble and Z-banding as the same problem. They aren't. They look similar at a glance, they live in the same neighbourhood of your print, and the fixes for one will do absolutely nothing for the other. If you swap your leadscrew because someone on a forum told you to and the artifact was actually a PID tune issue, you've spent thirty quid and a Saturday on the wrong fix. Worse, you'll start to believe nothing fixes this, which is what most "I gave up" forum posts boil down to. This guide is the long version of the conversation I wish I'd had with myself five years ago. I'll separate Z-wobble from Z-banding from VFA from layer shift with a proper disambiguation table, then I'll walk through mechanical causes, thermal causes, the OrcaSlicer settings that actually matter, the math behind the so-called "magic layer heights," and a diagnostic workflow that gets you to the root cause cheaply. There are per-printer notes for Bambu, Voron, Prusa, Creality, Sovol, and Anycubic at the end, and an FAQ for the questions that come up over and over on the orcaslicer.net inbox. If this is your first stop, my OrcaSlicer troubleshooting guide is the broader hub. What's in this guide Z-wobble vs Z-banding vs VFA vs layer shift Root causes of Z-wobble (mechanical) Root causes of Z-banding (thermal and extrusion) OrcaSlicer settings worth investigating The layer height magic numbers Mechanical fixes in order of cost and effort Per-printer notes A diagnostic workflow that doesn't waste filament What input shaper, anti-backlash nuts, and Oldham couplers actually do FAQ Z-wobble vs Z-banding vs VFA vs layer shift Spend the first cup of coffee on this section. If you walk away from this article with one thing, it should be the ability to look at a print and immediately know which of these four defects you have. The fixes diverge from this point. Z-wobble is a smooth, periodic horizontal ripple on a vertical wall, where the vertical period of the ripple matches the leadscrew lead. "Lead" is the axial distance the nut travels in one full rotation of the screw. A TR8x8 leadscrew, the most common in low-cost FDM printers, has an 8 mm lead. A TR8x4 has a 4 mm lead. A TR8x2 has a 2 mm lead. If you can stick a digital caliper against the side of your print and measure that the bands repeat every 2, 4, or 8 mm consistently, you're looking at Z-wobble. The cause is mechanical and lives in the Z drivetrain. Z-banding is a horizontal stripe or set of stripes whose spacing does NOT match the leadscrew lead. It's irregular, or it correlates with feature transitions (a wall count change, the start of an infill region, a tall thin part), or it tracks the bed heater duty cycle. The cause lives in thermal control or extrusion: PID instability, filament diameter swing, fan ramping, or layer-time effects. Most "Z-wobble nothing fixes it" forum threads I've read are actually Z-banding cases where the poster never measured the period. VFA stands for Vertical Fine Artifacts. These are sub-millimetre ridges, often spaced around 2 mm but really tied to belt pitch (one GT2 tooth = 2 mm of belt travel) or motor full-step distance, not the Z leadscrew. VFAs are an XY problem, not a Z one. I've covered VFAs at length in the OrcaSlicer VFA test guide; the short version is that input shaper, pressure advance, and belt tension are the levers, and none of those touch Z-wobble. Layer shift is a catastrophic XY displacement where one layer (or a stack of layers) sits offset sideways from the rest of the print. It is not a Z defect at all. It's caused by lost steps on X or Y, belt slip, or a mechanical strike. There's a full OrcaSlicer layer shifts article for that one because the diagnosis is entirely different. Here's the cross-reference table I keep open on a second monitor when I'm troubleshooting a stranger's print over Discord: DefectWhat it looks likePeriodPrimary cause Z-wobbleSmooth horizontal ripple, regularMatches leadscrew lead (2, 4, or 8 mm)Bent leadscrew or coupler bind Z-bandingHorizontal stripes, irregular or feature-linkedRandom, or tracks thermal cyclePID, filament, fan ramps, extrusion VFAFine vertical lines or sub-mm bumpsSub-mm, ties to belt pitch or motor stepBelt or motor mechanics, XY axis Layer shiftOne-time XY offset between layersOne-time eventLost steps, belt slip, mechanical strike The single best diagnostic move you can make is to measure the band period with a caliper. If you can't measure it (matte filament, very subtle defect), print a thin tall tower in glossy filament and try again under raking light. Without that measurement, every other troubleshooting step is a guess. One more clarification before we move on. Some online guides use "Z-banding" as an umbrella for both defects, then prescribe leadscrew swaps to people whose problem is a 4 °C temperature swing. That's the gap this guide is trying to close. From here on, I'll use "Z-wobble" only for the mechanical, period-matched ripple, and "Z-banding" only for the thermal or extrusion-driven stripe. Root causes of Z-wobble (mechanical only) Z-wobble is a mechanical artifact. Every cause in this section lives in the Z drivetrain or the frame that supports it. None of them can be fixed by changing a setting in OrcaSlicer, although some of them can be masked enough to look fixed if you're not paying close attention. Bent leadscrew Far and away the single most common physical cause, especially on printers under five hundred quid. Cheap TR8 leadscrews ship bent from the factory more often than the manufacturers will admit. The diagnosis is free and definitive: remove the leadscrew, lay it on a flat surface (a kitchen worktop or a piece of float glass), and roll it. If you see a visible gap appear and disappear as it rolls, or if it rocks rather than rolls smoothly, the screw is bent. Straightening it almost never works, and even if it appears to work, the internal stress shows up as wobble again a few months later. Replacement with a quality TR8x8 (or whatever lead your printer uses) is the answer. Rigid coupler transmitting screw runout The coupler between your Z motor shaft and the leadscrew matters more than most people realise. If it's a rigid aluminum sleeve with grub screws, any wobble in the screw and any misalignment between motor mount and screw axis transfers directly to the nut, and from the nut to the bed or gantry. Rigid couplers are the worst-case for Z-wobble. A flexible (helical) coupler is better. An Oldham coupler is best, but it asks more of your installation. Coupler over-tightened or screw bottomed against motor shaft If you cranked the grub screws hard against a screw that's seated against the end of the motor shaft, you've effectively pinned the screw against the motor and lost the small axial play that lets the screw self-center on the nut. The cure is to loosen the grub screws, lift the screw a millimetre or two so it isn't bottomed out, and re-tighten with gentle torque. Free fix. Often missed. Loose Z-motor mount The Z motor sits in a mount at the bottom of the column on most bedslingers, or at the top in some CoreXY designs. If those bolts have loosened (and they do, vibration loosens everything), the motor can shift in place during a Z move. The leadscrew walks. Check during a slow manual Z move and watch the bottom of the screw for any lateral movement. Frame flex Most relevan [truncated] --- ## OrcaSlicer Over-Extrusion, Blobs & Zits: How to Fix Them (2026) URL: https://orcaslicer.net/orcaslicer-over-extrusion-blobs-zits/ The print that broke me was a tiny architectural model. Smooth cylinder, sharp cube on top, nothing exotic. It came off the build plate looking like a teenager going through a bad week. There was a column of round bumps stacked up the side of the cylinder, a few pimply specks scattered across the cube, and the bottom three layers had flared out like the base of a wedding cake. I had been printing this exact model with this exact filament for six months. So I did the wrong thing first. I dropped the temp by 10 degrees, cranked the retraction, and reprinted. Got a different version of the same disaster. That moment is when I learned, with feeling, that "blobs and zits" is not one problem. It is at least three different problems that share a vocabulary, and the fix for one is irrelevant to the others. If you treat them as one bucket, you end up doing what I did: chasing your tail through settings menus, breaking calibration you'd already done, and convincing yourself the printer hates you. This article is the diagnostic walk-through I wish I had that night. We are going to separate the defects first, then attack each one with the right OrcaSlicer setting, in the right order. If your symptoms look like the opposite (gaps in walls, thin top layers, missing infill) you want the sibling piece on OrcaSlicer under-extrusion instead. If you're just dropping in to triage and don't have time for a long read, the broader OrcaSlicer troubleshooting guide has the 30-second flowchart version. Everything below is verified against OrcaSlicer 2.3.x. Table of contents 60-second diagnosis: which defect do you actually have Causes ranked by impact Fix 1: Flow ratio (the real "over-extrusion" lever) Fix 2: Pressure advance (the real "blob" lever) Fix 3: Seam strategy, painting, and scarf joint (honest version) Fix 4: Wipe and retraction Fix 5: First-layer elephant foot Other contributors worth checking Common false fixes FAQ Wrap-up and what to calibrate next 60-second diagnosis: which defect do you actually have Most blob/zit guides online treat every surface artifact as one thing. The Obico blog on blobs and zits, for example, does this. It's a useful resource for setting names, but it conflates two distinct defects, and that conflation is exactly why people end up applying the wrong fix. Before you touch a single slicer value, look at your print and decide which of the four families below you're staring at. DefectWhat it looks likeWhere it appearsPrimary causeWhere to jump Over-extrusion Whole part looks puffy. Top surface is ridged. Holes are undersized, outer dimensions are oversized. Layers bulge between perimeters. Everywhere. Not point defects. Flow ratio too high, or first-layer Z-offset too low. Flow ratio Blobs Round bumps roughly 1 to 3 mm wide. Vertically stacked at one location, often the Z-seam line. Sometimes at the end of a long straight line. Same spot, every layer. Pressure advance miscalibration and seam settings. Pressure advance Zits Tiny pimples, sub-millimeter. Scattered randomly across the surface or along seams. Random spread across walls. Slightly high temp, mild ooze, wet filament, or retraction at start/stop points. Wipe + retract Scars / travel marks Linear smears or diagonal scratches across the surface. Following travel paths. Nozzle dragging during travel (Z-hop or wipe problem), not flow. Other contributors Two things this table is not telling you: Stringing is a separate problem. Thin filament hairs spanning gaps between two separated print islands. That's a retraction and travel-temp problem, and it has its own fix path. If you have hairs more than visible bumps, jump to OrcaSlicer stringing and oozing fixes. Most "blobs" people complain about are really pressure advance problems wearing a costume. If your blobs are stacked vertically on the same outer-wall column on every single layer, that's a pressure release problem at the seam, and changing seam settings will move it around but won't make it go away. The thing that makes it go away is calibrating pressure advance. Hold that thought, it'll matter in the PA section. Look at your print. Pick one of the four. The rest of the article assumes you've done that. Causes ranked by impact Here is the order I attack these defects in. I have spent enough time on this to know that randomly tweaking from the bottom of the list is how you waste a Saturday. Top of the list first, every time. Flow ratio too high. Single biggest cause of overall over-extrusion. Every 0.01 above the true value adds roughly 1 percent material across the whole print. Manifests as elephant foot, ridged top surface, swollen layers, undersized holes. Calibrate with OrcaSlicer's built-in flow ratio test. Do not chase a universal target value, every printer-filament-nozzle combo lands somewhere different. Pressure advance not calibrated. Single biggest cause of corner blobs and seam-start blobs. When PA is too low, pressure built up in the nozzle during acceleration does not get released cleanly on deceleration, so the nozzle dumps excess plastic at corners and at the end of each perimeter loop. That deposit is your seam blob. Seam placement concentrating all artifacts at one spot. Even with perfect flow and PA, plastic has to start and end somewhere. The seam mode you pick decides whether you see a wandering column of small bumps or a clean line in a hidden corner. Temperature too high for the filament. Each 5 degrees C above the ideal point increases ooze and zit count. Run a temp tower. Note: if PA is the real issue, lowering temp is just papering over the symptom. Wipe on retract misconfigured. If retract_before_wipe is 0 percent (the default) and your wipe distance is short, the nozzle is still pressurized during the wipe move and smears plastic, instead of cleaning it. GitHub PR #3094 and Issue #5896 trace this exact failure mode. Filament diameter mismatch. The slicer expects 1.75 mm. The actual filament might be 1.78 or 1.72. Because cross-sectional area scales with diameter squared, a 0.05 mm error is about a 5 percent volume error. Measure with calipers in 5 spots, 10 cm apart, average them, plug the value into the filament profile. Z-hop type wrong for your setup. Normal lifts straight up and can leave a blob at the lift point. Slope and Spiral keep the nozzle moving while lifting. GitHub Issue #10663 documents a bug where Slope and Spiral do not always apply Z-hop to all travel moves, leaving stray blobs. Auto is usually the safe default. Inner wall direction. Inner/Outer is the OrcaSlicer default for most profiles. Inner/Outer/Inner (re-introduced with arachne) gives the best outer surface quality because the outer wall prints between two already-laid inner walls. Don't change it universally, change it when you have a specific surface goal. First-layer flow specifically over-extruding. Separate setting family. A too-low Z-offset can give you elephant foot even when your overall flow ratio is correct. Outer wall slowdown not applied. If you're printing outer walls at the same speed as inner walls, pressure advance has less time to do its job and you get more blobs. Slow outer walls to roughly 50 to 100 mm/s. Notice the top two are calibration tasks, not setting tweaks. That is not an accident. Most articles that say "tweak these eight values to fix blobs" are skipping the only thing that actually works. Fix 1: Flow ratio (the real "over-extrusion" lever) If your symptoms are whole-part over-extrusion (puffy walls, ridged top, holes too small) you almost certainly have a flow ratio that is too high. Lowering nozzle temperature will not fix this. Increasing retraction will not fix this. The only thing that fixes it is calibrating flow ratio so the slicer's expected extrusion volume matches what the printer actually lays down. OrcaSlicer ships with a built-in flow ratio calibration test under Calibration > Flow Rate. There are two variants: YOLO single-pass. Faster. Prints one set of squares at varying flow ratios, [truncated] --- ## OrcaSlicer Under-Extrusion: Diagnosis & Real Fixes (2026) URL: https://orcaslicer.net/orcaslicer-under-extrusion/ The first time under-extrusion really wrecked one of my prints, I spent four hours convinced my Z-axis was loose. I poked at belts, retightened eccentric nuts, ran a bed mesh, swore at my Ender, and only when I held a flashlight at a low angle did I see what was actually wrong: thin pinstripe gaps running between every single perimeter, like the slicer had decided to print stripes on purpose. The belts were fine. The bed was fine. The hotend was starving. That is the thing about under-extrusion in OrcaSlicer. It rarely shouts. It whispers in symptoms that look like ten other problems, and most of us misdiagnose it at least twice before we land on the right fix. I have been printing on OrcaSlicer since the SoftFever fork days, I run a Bambu P1S and a Voron 2.4 next to a stubborn old Ender 5 Pro, and I have probably wasted more filament on bad flow calibrations than I want to admit publicly. So this guide is the ladder I actually climb when prints come out starved, in the order I climb it, with the math, the menus, and the real GitHub issue numbers behind the gotchas. If you skim nothing else, skim Section 3 (the diagnosis ladder). The wet-filament and partial-clog causes account for the majority of the under-extrusion cases I see in Discord and on the OrcaSlicer issue tracker, and they are also the two fixes people skip while they crank flow ratio and make everything worse. What we will cover What under-extrusion actually looks like Under-extrusion vs over-extrusion vs gaps Causes ranked by impact (the diagnosis ladder) OrcaSlicer settings to check first Flow ratio calibration: four methods Max Volumetric Speed (MVS) deep-dive Hardware diagnostic workflow Filament-specific gotchas The 2.3.1 thin-wall regression (Issue #11394) Common false fixes that waste a Saturday FAQ What under-extrusion actually looks like Under-extrusion is the umbrella term for "the printer laid down less plastic than the slicer asked for." That undershoot can be 2% (almost invisible until you side-light a top surface) or 30% (the print falls apart in your hand). What makes it confusing is that the same underlying cause can show up as five different cosmetic symptoms, and a fresh user almost always reaches for the wrong fix. Here is the symptom list I run through in order of how often I actually see them on prints sent to me for diagnosis. Match yours to one of these before you change a single setting. Gaps between perimeters (the headline symptom) Adjacent walls do not fuse. You will see slivers of empty space between the outer wall and the second wall, often as straight pinstripes running parallel to the perimeter. Side-light it with a phone torch held at maybe ten degrees off the surface and the gaps pop visually. This is the canonical under-extrusion look. If you only see it on the first layer, that is not under-extrusion, that is first layer height (see our first-layer guide). Top surface pinholes The top solid layers do not close. You see the infill grid texture bleeding through what should be a clean top, or pinprick holes scattered across the surface. People reflexively crank "top shell layers" from 4 to 8 and the problem stays. The fix is upstream: the lines on each top layer are starved, so adding more starved layers does not help. Parts that split along layer lines Functional prints snap cleanly at a layer line under hand pressure. The plastic was laid down but did not have enough volume to weld to the layer below. If your benchies look fine cosmetically but your mounting bracket cracks at one specific layer, you are almost certainly under-extruding at that height. Inconsistent line width along a single perimeter Look at one continuous wall under a flashlight. If the line visibly thins, fattens, thins again as the toolhead moves, you have either intermittent slippage at the extruder or a partial clog. This one almost always pairs with a clicking extruder. Banding in top surfaces Long parallel grooves in solid fills. The grooves follow the infill direction, not the Z stack. That distinction matters: Z-banding follows the Z axis (lead screw wobble, eccentric Z motor), under-extrusion banding follows the infill angle. If you rotate the part 45 degrees on the build plate and the banding rotates with it, it is extrusion, not mechanical. Missing fine details Small text, fine ridges, gap fill between thin walls just do not print. They show up as dotted lines or simply blank space. Same root cause, different cosmetic outcome. Under-extrusion vs over-extrusion vs gaps This is where I see beginners flip the diagnosis the most. Over-extrusion and under-extrusion are opposite symptoms with opposite fixes, and a wrong call here just deepens the hole. Here is the contrast table I keep pinned in my workshop. Symptom areaUnder-extrusionOver-extrusion Top surfacePinholes, infill showing throughRough, bumpy, "scrubbed" look PerimetersVisible stripe gaps between wallsWalls smear into each other, blobs First layerFilament looks thin, no squishElephant foot, bulging base CornersSharp but starved, dottedRinging, blobs at direction change Z seamOpen gap, visible vertical seam lineZit or blob at the seam DimensionsHoles slightly oversized, walls slightly undersizedHoles undersized, walls oversized If your symptoms point at over-extrusion, hop over to the sibling article on blobs and zits. Everything below assumes you have confirmed under-extrusion. A word on "gaps" specifically. OrcaSlicer 2.3.1 introduced a regression on thin walls roughly 1 to 2 mm wide where adjacent perimeters behave like separate shells rather than fusing. That is tracked as GitHub Issue #11394, and I cover it in its own section below because the fix is "roll back to 2.3.0 or wait for 2.3.2," not "calibrate harder." The YOLO flow calibration plate as it lands on the build surface inside OrcaSlicer. Each block uses a different additive modifier so you can pick the cleanest top. Causes ranked by impact (the diagnosis ladder) This is the spine of the article. I have it ordered by what causes the most under-extrusion cases I see across forums, Discord, and my own bench. Work top to bottom, not bottom to top. Most people land on a fix by step four. 1. Wet filament (the silent killer) Moisture trapped in the filament flashes to steam inside the hotend. You get micro-bubbles in the melt, disrupted flow, and prints that look starved even though everything else is fine. Tell-tale signs include popping, hissing, or crackling sounds at the nozzle, visible steam in a dark room, and surface roughness that comes and goes within the same layer. PETG, TPU, nylon, PC, and PVA are the worst offenders. PLA is more forgiving but still affected after a few humid weeks on the shelf. Fix: dry the spool. Typical temps are 45-55 C for PLA, 65 C for PETG, 80 C for nylon. A purpose-built filament dryer is worth the spend if you live anywhere humid. This is the cause I see missed the most, especially in winter when people put a humidifier in the print room. 2. Partial nozzle clog or hotend gunk Burnt material, dust, or carbonised filament narrows the effective bore. Plastic still extrudes, just less of it. Listen for a "thunk-thunk" as the extruder skips and look at perimeters: if one wall looks consistently thinner than the others under a flashlight, that is your clog signature. Cold pull is the first hardware diagnostic and I cover the procedure in Section 7. 3. Flow ratio miscalibrated OrcaSlicer ships every filament profile at 1.0 flow ratio. Per-spool variance can be plus or minus 0.05 even within the same brand and material, so a generic PLA profile against a thinner-than-spec spool will under-extrude on day one. Calibration is the fix, not cranking flow ratio blindly. Methods in Section 5. 4. Extruder gear slipping or chewed Look under the printer for filament dust. That is the giveaway. The drive gear has grooves clogged with PLA crumbs and cannot bite, or it has chewed a notch into the filament and lost grip. PETG is notorious for this [truncated] --- ## OrcaSlicer Layer Shifts: Causes, Diagnostics, Fixes (2026) URL: https://orcaslicer.net/orcaslicer-layer-shifts/ I lost a 14-hour print to a layer shift last winter and spent the next two days convinced my mainboard was dying. It wasn't. It was a single grub screw on the Y-axis pulley that had backed off about a millimeter, and the motor was happily spinning the pulley loose against the shaft every time the bed slung itself across the gantry at 8,000 mm/s². I had stared at that little brass screw a hundred times during normal maintenance and never thought to check whether it was actually doing its job. After I tightened it (against the flat of the shaft, the right way, which I'll get to below), the printer ran for the next 400 hours without a single shift. That's how layer shifts go. They feel mechanical and electrical and software all at once, and the actual fix is almost always smaller than your panic. I've written this guide as the troubleshooting tree I wish I'd had that week. If you're reading this with a half-failed print sitting on your bed and the top half offset two centimeters from the bottom, I want to get you to "fixed" in the shortest path possible, with no false fixes and no fabricated cause-of-the-week. We'll cover what a layer shift actually is (and what it isn't), the four categories of root cause, the OrcaSlicer settings that matter, the Klipper and Marlin and Bambu-specific differences, a cheap-first diagnostic workflow, and the common bad advice you'll find on Reddit so you can stop chasing it. If you'd rather start with general slicer troubleshooting before diving into this one specifically, our OrcaSlicer troubleshooting guide is the umbrella piece. This article is the deep dive for one specific failure mode. Table of contents What a layer shift actually is What's not a layer shift (VFAs, Z-banding, salmon skin, ghosting) Root cause categories Mechanical causes (start here) Electrical causes Slicer and firmware causes The print itself OrcaSlicer settings that matter Klipper-specific notes Marlin-specific notes Bambu Lab-specific notes The diagnostic workflow Common false fixes FAQ What a layer shift actually is A layer shift is a sudden, catastrophic offset of the upper portion of a print relative to the lower portion. You'll see a clean step at one Z height where everything above is translated a few millimeters (sometimes a few centimeters) in X, or Y, or occasionally both. The shift is binary. There's the layer where it happened, and every layer above continues from the new wrong position, because the firmware doesn't know it lost track. Most consumer FDM printers don't have closed-loop position feedback, so when a stepper skips, the controller keeps issuing G-code as if the carriage were still where it commanded it to be. The print continues, just two centimeters off. That's the key thing I want you to take away before we get into causes. A layer shift is a loss of synchronization between commanded position and actual position. Everything we're going to discuss is either a reason the motor couldn't deliver the commanded motion, or a reason something physically prevented the carriage from following. The other thing worth understanding: by the time you can see a shift, the failure already happened. You can't "tune" your way out of one in real time. You can only prevent the next one. And because the cause space is wide (belts, pulleys, drivers, current, acceleration, collisions, warping), the worst thing you can do is start by lowering acceleration and hoping. That's the equivalent of taking a painkiller for a broken wrist. It might make the pain go away long enough to ignore the actual problem, until the problem gets worse. What's not a layer shift I see this confusion constantly in forum threads. Users come in saying "I have a layer shift" and post a photo of subtle banding or a scale-pattern wall texture, and the entire thread spins for forty replies before someone gently points out they're chasing the wrong problem. So before we go further, here's the disambiguation table. What you seeWhat it actually isWhere it belongs Clean horizontal step, entire top of model translated in X or YLayer shift (this article)Keep reading Subtle repeating vertical lines on flat walls, roughly 2 mm periodVFAs (vertical fine artifacts), caused by belt-tooth meshing or stepper resonanceSee our VFA test guide Horizontal banding throughout the print at regular Z intervalsZ-banding, usually leadscrew or Z-axis mechanical issueDifferent topic, not a layer shift Repeating wavy "fish-scale" pattern on perimetersSalmon skin, stepper-driver microstepping artifact (classic DRV8825 issue)Different topic, swap to TMC drivers Echoes or ripples next to sharp corners that fade awayGhosting / ringing, vibration after direction changesSee our input shaper guide First few layers are scrunched, dragged, or missing materialFirst-layer issue, not a layer shiftSee our first layer guide The fastest way to know you have an actual layer shift: look at the side of the print and see if you can identify a single Z height where the entire upper section is offset. If there's a clean step, that's a shift. If the texture is repeating or fading or distributed, it's something else. VFAs in particular get mislabeled as layer shifts on a daily basis. They're typically 10 to 40 microns wide, appear at intervals of about 2 mm along the Z axis, and persist across the entire print. They look like a problem you should be able to solve, which is why people get worked up about them, but they're a surface-finish issue rooted in resonance and tooth meshing, not a position-tracking failure. Z-banding sits in a similar bucket, just at the Z axis instead of XY. And salmon skin is a microstepping artifact that historically plagued DRV8825 boards and largely went away when builders moved to TMC2208/2209 drivers. If any of those describe your problem better than what I've described above, close this tab. None of the fixes here will help you and most of them will make your print times worse for no reason. Root cause categories I find it useful to group the causes into four buckets, because the diagnostic path is different for each. Most prints that shift fall into the first or fourth category. The middle two are real but rarer in well-maintained machines. CategoryExamplesFrequency (anecdotal)First check MechanicalBelt tension wrong, pulley grub screw loose, loose idler, frame flexMost common single bucketGrub screws on the shifting axis ElectricalStepper driver thermal throttling, run_current too low or too high, sense_resistor wrongCommon on overclocked or fan-failed setupsTouch the driver after a long print Slicer / firmwareAcceleration too high for hardware, jerk or Junction Deviation too aggressive, travel speed exceeds mechanical capabilityCommon after a user crank-up of "speed"Reduce acceleration and reprint Print-inducedWarped corner the nozzle hits, detached part on the bed, tall narrow geometry that vibratesVery common, often overlookedLook for a previously failed corner on the bed Important framing: I see this online all the time and it drives me up a wall. "Layer shifts are belt tension." No they aren't, not always. Belt tension is a frequent cause but it's one of many, and if you spend a week obsessing over belt tension when your actual issue is a heat-throttled driver, you'll be exactly where you started except now you've over-tensioned a belt and you're going to wear out your idler bearings. Treat the four categories as a search space, not a recipe. Diagnose first, then fix. Mechanical causes This is the heaviest section because it covers the majority of cases. I'll walk through each one in roughly the order I'd check them on a printer that just shifted on me. Belt tension too low This is the classic first guess and it's right often enough that it earned the reputation. When a belt is loose, the teeth don't seat fully into the pulley grooves under load. The motor turns, the pulley turns, but the belt slips a tooth or two under high acceleration. The result is a position offset that the motor [truncated] --- ## OrcaSlicer Warping & Lifting Corners: Fixes That Actually Work (2026) URL: https://orcaslicer.net/orcaslicer-warping-lifting-corners/ I lost a 14-hour ABS print last winter because of a single lifted corner. The part was a tool holder for my workbench, 180 mm long, printed on a P1S that I'd convinced myself was "basically enclosed enough." Around layer 90, one corner popped off the textured PEI plate, the head dragged across the loose section, and by the time I noticed, the whole thing looked like a piece of toast that had argued with the toaster. I had skipped the brim because I "didn't want to deal with peeling it off." That print cost me about $9 in filament and the better part of an afternoon, and it taught me something every ABS user eventually learns: warping is not a slicer bug. It is physics expressing itself through a sticky surface that finally gives up. If you're reading this because corners on your prints are lifting, curling, or popping off the bed mid-job, I'll save you the suspense up front. The fix is rarely a single OrcaSlicer toggle. Warping is primarily a thermal and hardware problem, and slicer settings help at the margins. That said, the margins matter, especially when you're running a passive enclosure like the P1S or X1C, or trying to stretch ASA on an open-frame bedslinger. I've spent the last year breaking, rebuilding, and re-tuning prints on three different printers (P1S, X1C, and an enclosed Voron 2.4 with a PTC chamber heater), and what follows is the playbook that actually works. I'll be clear about what's a slicer fix, what's a hardware fix, and what's a myth that gets repeated on Reddit. One more thing before we get into it. If you're printing PLA and you think you're seeing warping, you almost certainly aren't. I'll explain why later, but the short version is that PLA shrinks so little that "PLA warping" is nearly always a bed adhesion failure in disguise. Save yourself an afternoon and skip to the first-layer problems guide instead. Table of Contents Why warping happens (the physics in plain English) Material warping ranking (which filaments curl, which don't) OrcaSlicer settings that actually move the needle Hardware: the real fight Enclosure and chamber options Brim techniques in OrcaSlicer (the deeper playbook) Common false fixes Per-printer notes Worked example: dialing in ABS on a P1S FAQ Why warping happens (the physics in plain English) I want you to picture a single bead of molten ABS leaving the nozzle at around 260°C. It hits the cooling line below, fuses, and immediately starts shedding heat. As it cools, it contracts. The amount of contraction depends on the polymer, but for ABS, ASA, and PC it's somewhere around 1.5 to 2.0 percent linear shrinkage. On a 100 mm part, that's up to 2 mm of contraction trying to happen. Two millimeters doesn't sound like much, but stretched across a corner with a few square centimeters of bed contact, it's plenty of force to peel a lifted edge. The chain of causation goes like this. Layer N goes down hot and bonds to layer N minus one. Layer N then cools and contracts. But layer N minus one has already cooled and is now rigid. So layer N, in shrinking, pulls layer N minus one inward. Repeat that fifty times and you have a stack of layers all trying to pull the corners inward. The stress concentrates at the perimeter, and especially at corners, because corners receive contraction force from two axes simultaneously. Eventually the cumulative force at one corner exceeds the bed adhesion at that corner, and that corner lifts. From the lifted corner, the failure usually propagates inward, sometimes catastrophically. Two material properties decide whether this matters or not. The first is the glass transition temperature, or Tg. This is the temperature at which the polymer transitions from a rigid glassy state to a softer rubbery state. Above Tg, the polymer is ductile enough to dissipate stress by deforming slightly without losing dimensional integrity. Below Tg, stress is locked into the part as residual internal stress. ABS has a Tg around 105°C. PC is much higher, around 145°C. PLA is around 60°C, and PETG is around 80°C. Materials with high Tg trap more stress before they "freeze" into their final shape, which is why ABS, ASA, and PC are the materials that fight you hardest. The second property is the coefficient of thermal expansion, or more practically, the shrinkage rate per degree of cooling. ABS, ASA, PC, and nylon shrink a lot. PLA and PETG shrink less. Carbon-fiber-filled variants like PLA-CF, PETG-CF, and PA-CF shrink less than their base resin because the fiber doesn't shrink and mechanically restrains the matrix. That's why CF-filled engineering filaments are often easier to print than their unfilled cousins, even though the matrix material is the same. This is also why a heated chamber matters so much for the worst offenders. If you can keep the bulk of the print near Tg while subsequent layers are deposited, the lower layers stay ductile and dissipate the contraction stress instead of transmitting it to the corner adhesion zone. Drop the chamber temperature, and the lower layers go rigid, and now every new layer's contraction force gets concentrated at the bed bond. That's the moment corners start lifting. It's why an enclosed printer with a chamber heater can print 200 mm ABS plates with a 3 mm brim and finish clean, while an open-frame Ender 3 with the same plate and a 10 mm brim will still curl. The slicer is identical in both cases. The chamber is doing the heavy lifting. Material warping ranking (highest to lowest severity) Before you fight your warping problem, it helps to know what you're up against. Here's the ranking I've settled on after a year of testing, cross-referenced against the Bambu Lab material table and a stack of community threads. Tier Material Warping severity Why 1 Polycarbonate (PC) Severe Highest Tg of common filaments (~145°C), high shrinkage, brittle when cold. Practically requires a heated chamber for parts over 80 mm. 1 Nylon (PA, PA6, PA12) Severe High shrinkage and absorbs moisture, which compounds the problem. Enclosure mandatory, dry filament mandatory. 2 ABS / ASA Severe without enclosure Tg around 105°C. The classic warping case. Open frame ABS is a coin flip on anything bigger than a coin. 3 PA-CF / PAHT-CF Severe but less than unfilled nylon Still needs enclosure. Fiber helps but doesn't eliminate. Less moisture-sensitive in the spool, but still requires drying. 4 PETG Minor on tall or large prints Low shrinkage, low Tg, semi-amorphous. Usually only warps when a print is over 150 mm long or has a very thin first layer. 4 PLA-CF, PETG-CF Less than base resin Fiber matrix mechanically resists shrinkage. Often warps less than its base. 5 PLA Minimal Lowest practical shrinkage of common filaments. Most users will never see PLA warp on prints under 150 mm. Very large flat prints can still lift, but it's rare. 5 TPU Minimal Flexible, low Tg. It absorbs stress by stretching instead of warping. If you're on ABS, ASA, or PC, the rest of this article is for you. If you're on PLA and you're seeing corner lift, you have a bed adhesion problem, not a warping problem. Common causes: dirty plate, first layer too high, too-cool bed (PLA prefers 60°C, not 50°C), or a part with a tiny footprint sitting on textured PEI without a brim. Read the first-layer troubleshooting guide instead of this one. For PETG users seeing warping, the usual culprit is either too much fan on a tall print or a too-cool bed. PETG actually adheres extremely well to textured PEI, sometimes too well. If yours is lifting, look at the fan curve and the bed temperature before adding a 10 mm brim. OrcaSlicer settings that actually move the needle I'm going to walk through every setting in OrcaSlicer that meaningfully affects warping, in rough order of impact. None of these will save a 250 mm ABS plate on an open-frame Ender 3. But on a P1S or X1C, getting these right turns "occasional failures" into "reliable prints," which is the realistic goal for passive enclosures. Brim: the single biggest slicer-side [truncated] --- ## OrcaSlicer Stringing & Oozing Fixes: Ranked by Impact (2026) URL: https://orcaslicer.net/orcaslicer-stringing-oozing-fixes/ The first time I ran a "stringing test" benchy farm with PETG, I came back to the printer twenty minutes later and laughed out loud. The pillars looked like a doll-sized harp. Strings everywhere. I did what every panicked user does: I cranked retraction from 0.8 mm to 4 mm, ran it again, and watched a fresh nest of strings appear, slightly thicker this time. I spent the next three nights chasing settings before I noticed the spool had been sitting on a shelf next to a humidifier since Christmas. The fix was not in the slicer. It was in a $40 dryer. That experience is the spine of this guide. If you came here looking for "the magic retraction value for OrcaSlicer," I'm going to disappoint you on purpose. There isn't one. What there is, instead, is a stack of causes ranked by how often they're actually the problem. Work the stack in order and you'll stop chasing your tail. I've ordered the fixes by impact, not by how interesting they are to fiddle with, because I want you to spend an hour solving stringing rather than three weekends. This article covers what stringing actually is (and is not), the impact-ranked fix list, OrcaSlicer's stringing-related settings and where to find them in 2.3.x, material-specific drying and retraction tables, direct drive vs Bowden numbers, how to read the built-in retraction tower, and the false fixes that waste everyone's time. Pour a coffee. We're going deep. Table of contents What stringing is vs oozing vs blobs Causes of stringing, ranked by impact OrcaSlicer's stringing-related settings walkthrough Material-specific stringing patterns Direct drive vs Bowden retraction Drying guide table Stringing test prints (built-in) Common false fixes FAQ Final cleanup and CTA What stringing actually is (and what it isn't) Half the forum threads I read mash three different artifacts into one word. Before you spend money on a dryer or rip apart your hotend, get the vocabulary right, because each of these has a different root cause and a different fix. Stringing is the thin, hair-like filament strand dragged between two separate features of a print during a travel move. The nozzle moves fast across open air, the filament is still molten, and a strand peels out as the hotend leaves the part. You see this most clearly between the pillars of a stringing test or between the towers and ears on a multi-feature model. It's a travel-move artifact. If your strings appear only during travel moves, you have a stringing problem. Oozing is the broader phenomenon: a slow, continuous leak of plastic from the nozzle anytime the printer is not actively extruding. Residual pressure in the melt zone, thermal expansion of plastic inside the heatbreak, and moisture flashing to steam can all force molten filament out the orifice on their own. Stringing is one visible result of oozing during travels. Oozing also produces start-of-line zits, blobby seams where the perimeter restarts, and a noticeable puddle on the bed during long pauses or tool changes. Blobs (also called zits, pimples, or scars depending on which forum you're on) are localized aggregations of plastic at very specific points. Most of the time those points are the seam, the start and stop of a perimeter, layer changes, or the spot where a print paused for a filament swap. Blobs are not travel artifacts. They come from pressure-advance miscalibration, an imbalanced retract-and-restart, or seam-handling settings. If your "stringing problem" is actually a string of tiny zits running up the seam line, you don't need this guide. You need the pressure advance and seam articles. I want to flag this hard: stringing is not the same problem as Z-seams, vertical scars, or surface pimples. I see readers conflate them constantly. If your travel paths are clean but your seam line has bumps, close this tab and go read the seam material first. If your travel paths look like a spider's leftovers and the seam itself is fine, you're in the right place. One more boundary worth setting. A small amount of stringing on a torture-test pillar print is normal. PETG will often show faint hairs that wipe off with a heat gun even after textbook calibration. You're aiming for "good enough to clean up in ten seconds," not "absolutely zero strands under a microscope." Chasing perfection on a torture model leads to over-tuned settings that hurt actual prints. Causes of stringing, ranked by impact (the 80/20 list) I cannot overstate how much time I've watched people waste by starting at the bottom of this list. Retraction is the popular knob because it's the obvious one, but it's not where most stringing problems live. Work from the top, not the middle. Wet filament. This is the #1 cause and the one that can't be slicer-tuned around. Water trapped inside the filament flashes to steam at the melt zone, blows micro-bubbles out the nozzle, and oozes continuously regardless of how aggressive your retraction is. PETG, Nylon (PA), PC, PVA, and TPU are aggressively hygroscopic. PLA and ABS are less hungry for moisture but still affected. Fix: dry the spool before you touch the slicer. Print temperature too high. Hotter plastic is runnier and oozes more. Try dropping nozzle temperature 5 to 10 °C and rerunning the test. OrcaSlicer's Temperature Tower calibration is the right tool to find the lowest clean band. Pressure advance not calibrated. Uncompensated pressure in the melt zone keeps pushing plastic after the extruder motor has technically stopped. This produces both end-of-line ooze and stringing on travels. Calibrate pressure advance before you touch retraction. If you skip this step you're tuning retraction around a moving target. Retraction distance wrong for your hardware. Too low and the pressure isn't relieved before the travel starts. Too high (rare on modern direct-drive hotends) and you cause heat creep, grinding, or blobs at restart. Run the Retraction Test tower and pick the lowest clean value. Retraction speed wrong for the firmware or extruder. Too slow and pressure can't drop in time for the travel. Too fast and the extruder skips steps or grinds the filament. Bowden setups and weaker direct-drive steppers are especially sensitive here. Travel speed too low. This one is counterintuitive. Slow travels mean the nozzle hovers over open air longer, which gives ooze more time to manifest as a visible strand. Bumping travel speed to 150 to 250 mm/s often kills marginal stringing instantly. Z-hop adding ooze opportunity. Z-hop adds two short vertical moves to every travel. If your retraction is undertuned, Z-hop will visibly worsen stringing because the nozzle stays full of pressurized melt while it lifts and re-lowers. Tune retraction first, then enable Z-hop. OrcaSlicer's Slope and Spiral lift types are gentler on stringing than the Normal Lift. Nozzle gunk, partial clog, or worn nozzle. A burr around the orifice catches filament during travels and physically drags strands. A cold-pull cleans most partial clogs; brass nozzles wear out faster on abrasives like carbon fiber blends, glow-in-the-dark, or wood-filled PLA. If you're three months into a CF spool with a stock brass nozzle, the nozzle is the problem. Read that list again. If you skip steps 1 and 2 and jump straight to retraction, you will not fix a wet-filament print. I've tested this. I once dialed retraction from 0.8 mm to 6 mm on a soaking-wet roll of PETG just to prove the point to a friend. The strings got slightly shorter and slightly bumpier. They never went away. Twelve hours in the dryer fixed what no slicer setting could touch. OrcaSlicer's stringing-related settings walkthrough (2.3.x) OrcaSlicer puts stringing-related controls in three places and sometimes hides one behind another. Here's where to look, with the caveat that field labels occasionally shift between minor releases. Verify against the OrcaSlicer wiki at slice time if something looks renamed. Filament Settings (per-filament, usually the right place to edit) Most retraction parameters [truncated] --- ## OrcaSlicer First Layer Problems: A Symptom-Based Diagnosis Guide (2026) URL: https://orcaslicer.net/orcaslicer-first-layer-problems/ I have a 3D printer that lived in my garage for two years before I realized almost every failed print started failing in the same place: the first layer. Not halfway up. Not at some tricky overhang. The first ten seconds of motion against the build plate. Once I started treating the first layer as the whole job, my scrap pile shrank by maybe 80%, and the prints that did fail tended to fail for genuinely interesting reasons rather than because I rushed the start. This guide is the diagnostic flow I wish I had when I switched to OrcaSlicer. I am writing it as a symptom-to-cause manual, the way an automotive shop binder reads, because that is how people actually arrive here. You see something wrong on the plate, and you want to know which knob to turn. Every recommendation below maps to a real OrcaSlicer setting with its menu path, and where the slicer can't fix it, I will say so and point at the hardware fix instead. Quick note before we start. I'm a long-time OrcaSlicer user who runs a Bambu A1, a P1S, and an Ender 3 V2 with a Klipper conversion. The advice below works on all three, but the menu paths assume OrcaSlicer 2.x. If you're on an older 1.9 release, the panels are similar but the elephant foot compensation panel lived under a slightly different parent. The full OrcaSlicer troubleshooting hub covers cross-version differences if you need them. Table of contents The fast diagnostic table Gaps between first-layer lines First layer not sticking at all First layer too thin or smashed First layer too thick or ridged Elephant foot Corners lifting on layer 1 Lines start strong, then taper One side good, the other side bad OrcaSlicer first-layer settings reference Bed leveling and mesh Z-offset workflow Bed surface adhesion table Per-material bed temperature table First-layer flow ratio calibration Common mistakes FAQ The fast diagnostic table If you only read one section, read this one. Find your symptom in the left column, then work top-down through the likely causes. The "first thing to try" column is the smallest possible change you can make to confirm or rule out the cause, so you don't end up changing four variables at once and not knowing which one mattered. What you seeMost likely causes (ranked)First thing to try Gaps between first-layer lines, bed showing through1) Nozzle too far from bed. 2) First-layer flow ratio too low. 3) Initial layer line width too narrow.Baby-step Z down by 0.02 mm during a real print and watch the gaps close. Filament won't stick anywhere on plate1) Dirty build plate (skin oils). 2) Wrong plate type selected in slicer. 3) Bed temp too low. 4) Z too high. 5) First-layer speed too fast.Wash the plate with warm water plus dish soap, handle by edges only, retry the same file. First layer smashed flat, nozzle drags audibly1) Z-offset too negative. 2) Initial layer height set too low for the nozzle.Baby-step Z up by 0.02 mm and re-evaluate. Lines stand up like ropes, visible spacing between them1) Z-offset too positive. 2) Initial layer print height too high (above 65% of nozzle diameter).Baby-step Z down by 0.02 mm, then verify layer height matches the wiki table for your nozzle size. Bottom of the model bulges outward (elephant foot)1) Bed temp too high for material. 2) Z over-squish on layer 1. 3) Part cooling fan off too long.Enable elefant_foot_compensation at 0.1 mm and spread it across 2 layers. Corners curl up off the plateAdhesion or warping issue. Bed temp low, draft on the print, no brim on a tall narrow part.See the dedicated warping and lifting corners guide. Lines start nicely then thin out across the movePressure advance miscalibrated for layer-1 speed.Re-run pressure advance calibration and accept that layer 1 may need its own tune. Left side perfect, right side gapped (or vice versa)Bed not level, mesh stale, or gantry tilted.Re-run bed leveling and rebuild the mesh. Whole first layer offset in XYNot a first-layer problem. Belt tension, axis crash, or layer-shift cause.See layer shifts in OrcaSlicer. Stringy blobs on layer 1 onlyRetraction or ooze before the print starts, plus first-layer speed too slow over the gaps.See stringing and oozing fixes. The rest of this article expands each row into a proper diagnosis. I'll keep the prose tight and lean on tables for anything you'd otherwise be scanning to find. Gaps between first-layer lines This is the most common symptom I see in Discord pictures, and people misdiagnose it constantly. The fix order matters. Most users immediately reach for flow ratio, when nine times out of ten the actual problem is Z. Here's the confirmation test. Slice a 20 by 20 by 0.25 mm square. One wall, zero top layers, zero infill, and let it print. Pull it off when it finishes (it'll be one pass and done). Now look at the underside. If the lines are round, separated, and the square peels off cleanly with the bed texture barely visible on the underside, your Z is too high. Baby-step down 0.02 mm at a time on the next attempt. If the lines are flat, fused side by side, and the underside has the bed texture clearly imprinted (you can see the PEI peaks), your Z is correct. The gap is a flow or line-width issue. If you've confirmed Z is right and you still have gaps, then you can look at flow. There's a sharp caveat here that I see ignored everywhere on Reddit: the setting bottom_solid_infill_flow_ratio in OrcaSlicer only affects first-layer solid infill, not first-layer walls. That's per OrcaSlicer issue #11838, which clarified the scope of that knob. If your gaps are between perimeters, raising bottom_solid_infill_flow_ratio will do nothing. You'd need to nudge the global filament flow ratio (carefully, since it affects every layer) or wait for the dedicated first-layer wall flow override that's been requested in issue #11000. Line width matters less but is worth a glance. In Strength > Walls, initial_layer_line_width defaults to roughly 0.42 to 0.45 mm for a 0.4 mm nozzle. If somebody set it down to 0.4 flat, the line spacing will be tight and any imperfection in flow will show as bed-visible gaps. Bump it to 110% of nozzle diameter for a layer that's noticeably more forgiving. First layer not sticking at all This is the panic symptom. Filament balls up around the nozzle, you scrape, you try again, same result. I have a checklist I run before I touch any setting, because the slicer is almost never the cause. Wash the plate. Warm water, dish soap, fingers, rinse, towel dry. For textured PEI specifically, Bambu's own cleaning guide and forum consensus is dish soap, not IPA. Isopropyl alcohol on textured PEI tends to spread oils into the cavities of the texture instead of dissolving them out. On smooth PEI, IPA is fine. Check the plate selection in OrcaSlicer matches the plate on the machine. I cannot count how many times I've sliced for Cool Plate and printed on Textured PEI. The bed temp and the first-layer profile default to the wrong values and adhesion craters. Check bed temp. See the material table below. PLA at 35 C lifts off in seconds. ABS at 60 C is hopeless. Slow down layer 1. If you cranked your print speed up for production parts, drop initial_layer_speed to 50 mm/s or below for the diagnostic attempt. Confirm Z. Only after the above four steps. Z-offset is the most-blamed and least-likely cause of total adhesion failure on a freshly cleaned plate. The Bambu Lab forum thread on this topic (the one titled simply "First Layer Does Not Stick") is full of users who burned through hours of Z-offset tuning before finally washing their plate and discovering it was finger oils all along. I include this story because I did exactly the same thing on my A1 for a week. First layer too thin or smashed Symptom signature: the filament looks ironed flat. You can see ridges at the line edges where the plastic was pushed sideways. The square is hard to peel off even when the bed is cool. On some printers you can hear the nozzle dragging across the previously-extruded line as it lays the [truncated] --- ## OrcaSlicer Can't Connect to OctoPrint: Every Fix That Works URL: https://orcaslicer.net/orcaslicer-octoprint-troubleshooting/ I once spent thirty minutes staring at a "401 Unauthorized" toast in OrcaSlicer before I realized the API key was fine. The actual culprit was an nginx block I'd added six months earlier to put OctoPrint behind a real domain, and it was quietly stripping the X-Api-Key header before the request hit OctoPrint. The browser worked because cookies survive proxies. The slicer didn't, because it doesn't speak cookies. If you're here, you've already done the setup. The Test button is throwing red, the upload stalls, or the file lands but never prints. This is the flowchart, ordered by how often each cause shows up in the OrcaSlicer and OctoPrint trackers. If you haven't wired things up yet, start at the OrcaSlicer OctoPrint setup guide. On Klipper with Mainsail or Fluidd, hop to the Klipper setup guide. Bambu users belong in the Bambu cloud and LAN guide. What's in this guide The fast triage: which error are you seeing? A. "Connection refused" or wrong IP/port B. "403 Forbidden": wrong or missing API key (yes, 403 not 401) C. "401 Unauthorized": almost always reverse proxy or nginx D. Reverse proxy and nginx misconfig E. Cloudflare Tunnel or Tailscale quirks F. SSL self-signed certificate failure G. OctoPrint plugin conflict (safe mode) H. 413 Payload Too Large I. Pi running out of disk J. Network drops or DHCP changed your IP When to use Mainsail, Fluidd, or PrusaLink instead What success actually looks like Frequently asked questions The fast triage: which error are you seeing? OrcaSlicer surfaces what libcurl tells it plus whatever HTTP status OctoPrint hands back. Match your symptom, jump to the section. Most readers land in A, B, or C. What you seeWhat it meansJump to Connection refused, Error 7Nothing listening on that IP/portA 403 ForbiddenAPI key wrong, missing, or user lacks upload permissionB and C 401 UnauthorizedReverse proxy ate the auth headerD 404 Not FoundPath prefix mismatch behind a proxyD 409 ConflictPrinter not in operational stateA bottom 413 Payload Too LargeUpload bigger than the proxy or OctoPrint will acceptH 500 Internal Server ErrorA plugin crashed during the requestG 502/503/504Proxy can't reach OctoPrint, or upstream timed outD Operation timed out, Error 28Wrong IP, firewall, or wifi packet lossJ Empty reply, Error 52 / Send fail, Error 55Connection died mid-transferJ SEC_E_UNTRUSTED_ROOTlibcurl rejected your self-signed certF One thing to flag before you start clicking. OctoPrint returns 403 Forbidden when the API key is wrong, not 401. This is documented in OctoPrint issue #2706; the docs specify 401, but the server has long behaved this way in practice. Treat 403 as "the key didn't work" first, "the user can't upload" second, or you'll burn an hour in the wrong place. This is where almost every connection failure starts. Three fields, three ways to ruin your afternoon: the host, the port, and the key. Host Type stays on Octo/Klipper for both OctoPrint and Klipper-via-OctoPrint setups. A. "Connection refused" or wrong IP and port This is the biggest category. Curl Error 7 means the TCP handshake never completed, because nothing accepted on that address and port. Almost never an OctoPrint problem; it's a "Orca is talking to the wrong place" problem. Diagnostic: From the slicer machine, ping the Pi by IP. If that fails, network issue, not OctoPrint issue. Try octopi.local. If ping resolves to 127.0.0.1, you've hit OctoPrint #3852, where Bonjour misbehaves on Windows after 1.5.0. Switch to the literal IP. SSH in, run sudo systemctl status octoprint. If it's not running, that's your answer. Run sudo ss -tlnp | grep 5000. On stock OctoPi, HAProxy fronts on port 80 with OctoPrint on 5000. On a manual install, you might be on 5000 only. Fix is almost always the literal IP in Orca's host field plus the right port. 192.168.1.145:5000 for a manual install, 192.168.1.145 (port 80) for stock OctoPi. Don't use octopi.local in production unless you've confirmed mDNS works on the slicer machine; Windows in particular drops Bonjour resolution after restarts. One subtle case: 409 Conflict. Network is fine, OctoPrint is alive, key is right, but the printer isn't operational. OctoPrint refuses uploads when the printer is mid-print, SD-only, or not connected. Click Connect in the OctoPrint UI first, then retest. B. "403 Forbidden": wrong or missing API key (yes, 403 not 401) If ping works, OctoPrint loads in your browser, and Orca still bounces you, the key is the next suspect. As I said, OctoPrint hands you 403 for a wrong key. That tripped me up the first time, because every other API I'd touched used 401 for auth failures. Generate a fresh key the modern way: OctoPrint Settings, then click the Application Keys entry in OctoPrint's Settings sidebar. Type "OrcaSlicer", click Generate. Click Allow on the confirmation popup. Use the copy icon next to the key. Don't double-click and Ctrl+C; that frequently grabs a trailing space, and a trailing space fails auth with no helpful error. Paste into Orca, hit Test. I lost two hours to a trailing space once. The strings looked identical when I diffed them. Always use the copy icon. The legacy global API key under Settings > API is deprecated in favour of Application Keys (see Application Keys documentation). It still works in 1.x but it's living on borrowed time. C. "401 Unauthorized": almost always reverse proxy or nginx OctoPrint itself rarely returns 401 to Orca, because Orca uses API keys, not session cookies. If you're seeing 401, something between Orca and OctoPrint is intercepting. Nine times out of ten that's a reverse proxy with basic auth in front, or a header policy stripping X-Api-Key. This is the trap I fell into. I'd added basic auth to nginx to keep the OctoPrint UI off the public internet. The browser stored a cookie, so I never saw a prompt. Orca got a flat 401 because it doesn't speak basic auth, and the original X-Api-Key never reached OctoPrint. If your OctoPrint is behind nginx, HAProxy, or Caddy and you see 401: Check the proxy's auth_basic directive. If it exists, remove it for the API path or add the basic-auth credentials to Orca's Username and Password fields. Confirm the proxy doesn't strip Authorization or X-Api-Key headers. Bypass the proxy as a test by pointing Orca at the Pi's LAN IP and port 5000. If LAN-direct works and proxied doesn't, you've confirmed the proxy. Detail: if your API key was issued for a user without FILES_UPLOAD permission, you'll see 403 on POST /api/files/local while GETs to /api/version pass. That permission split came in with OctoPrint 1.4.0 and broke a lot of slicer integrations (see the 1.4.0 ACL break thread). Fix: put the user in Operators, or grant FILES_UPLOAD and PRINT on your custom group. If you skipped Access Control during the first-run wizard (this screen, the moment you click Finish), every Orca request hits an unauthenticated OctoPrint and gets a 403. Re-run the wizard from System if you bypassed it. D. Reverse proxy and nginx misconfig Reverse proxies are the most common production-grade trip-up. The OctoPrint UI works in your browser through your custom domain, the API does not, and you start questioning your sanity. The fix is the canonical proxy block published by foosel in the OctoPrint reverse proxy thread. The headers matter and the body size matters. Bare-minimum nginx config that won't fight you: location / { proxy_pass http://127.0.0.1:5000; proxy_http_version 1.1; proxy_set_header Host $host; proxy_set_header X-Real-IP $remote_addr; proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for; proxy_set_header X-Scheme $scheme; proxy_set_header Upgrade $http_upgrade; proxy_set_header Connection $connection_upgrade; client_max_body_size 1024M; proxy_read_timeout 600s; } Note: the $connection_upgrade variable must be defined via a map $http_upgrade $connection_upgrade { default upgrade; '' close; } directive at the http scope of your nginx config, or nginx will refuse to start. Then in ~/.octoprint/config.yaml: server: reverseP [truncated] --- ## OrcaSlicer Troubleshooting: Master Index of Every Common Fix URL: https://orcaslicer.net/orcaslicer-troubleshooting-master/ Last March I had a print fail seven times in a row. Bambu A1, fresh PLA, simple model. The first three failures looked like classic stringing, the fourth like under-extrusion, the fifth like the print head was crashing into earlier objects. By the seventh I was ready to throw the printer out of a window. The actual problem turned out to be a Local Network privacy bug in macOS 14.5 that was making OrcaSlicer randomly drop the connection to the printer mid-print, sending it into emergency stop. Twenty seconds in System Settings fixed it. That's the pattern with OrcaSlicer troubleshooting. The symptoms are obvious. The actual causes are often two layers deeper than where you'd think to look. This guide is the index I wish I'd had: every common OrcaSlicer failure I've seen on Reddit, in GitHub issues, on forums, organised by what you'd actually search for at 11pm when nothing's working. Plus the fix sequence, the real root cause, and the GitHub issue numbers if you want to verify the bug isn't just you. It's organised in three categories: software issues (slicer crashes, won't open, settings won't save), print quality issues the slicer can fix (stringing, warping, elephant foot, layer shift), and connectivity issues (Klipper, OctoPrint, Bambu Cloud, macOS specifically). At the end there's a decision tree that maps "what you see" to "where to start". Skim the categories, find your symptom, click through. Written against OrcaSlicer v2.3.2, current stable. The decision tree (where to start) Match your situation to the symptom and skip to the matching section. OrcaSlicer won't launch at all → A2 Won't Open OrcaSlicer crashes mid-slice on a complex model → A1 E-core crash OrcaSlicer slices fine, but settings I save vanish next session → A3 Settings not saving Slice button stays disabled with "object outside print volume" → A7 "Spiral mode wall loops" toast → A4 Print time estimate is wildly wrong → A6 Slicing a 100MB+ STL takes 10+ minutes → A5 Print fails with strings between every feature → B1 Stringing Corners lift off the bed → B2 Warping Bottom layers wider than the rest → B3 Elephant foot Print suddenly offsets at a specific height → B4 Layer shift Echo / wave pattern after corners → B5 Ringing First layer doesn't stick or nozzle is scraping → B6 Walls have gaps and pinholes → B7 Under-extrusion Walls look bumpy with corner blobs → B8 Over-extrusion Top surfaces have small holes → B10 Top gaps Can't connect to Klipper / Mainsail / Fluidd → C1 Can't connect to OctoPrint → C2 Can't connect to Bambu Cloud → C3 Mac specifically: printer doesn't appear in the Device tab → C4 Printer just won't show up at all → C5 Category A: Software issues A1. OrcaSlicer crashes while slicing complex models (the E-core bug) Symptom: the app vanishes mid-slice on Intel 12th, 13th, or newer-gen CPUs. Windows Event Viewer logs ExceptionCode c0000005 with FaultingOffset in OrcaSlicer.dll. Triggers reliably on models above 500K triangles. Some users see kernel-mode BSOD. Root cause: thread synchronisation issues across heterogeneous P-cores and E-cores. OrcaSlicer's slicing scheduler doesn't gracefully handle the asymmetric core architecture introduced by Intel's hybrid CPUs. Memory pressure is a secondary cause. This is GitHub issue #8227, closed-not-planned. Fix sequence: Right-click OrcaSlicer.exe → Properties → Compatibility, then set CPU affinity to P-cores only. Process Lasso ($40 for permanent affinity) is the cleanest path. Free option: Task Manager → Details → right-click OrcaSlicer.exe → Set affinity, uncheck E-cores. Reduce the model's triangle count in Blender or MeshLab. Decimate modifier targeting -50% almost always saves the slice. Increase Windows page file to 32GB minimum. Close browser tabs. Try Windows 8 compatibility mode on the OrcaSlicer shortcut. Slower slicing but stable. Last resort: disable E-cores in BIOS. Costs general-purpose performance but eliminates the bug. Workaround at the slicing-settings level: increase perimeters to 4 or layer height to 0.2mm, both reduce per-layer compute load. A2. OrcaSlicer won't open or won't launch Symptom on Windows: splash flashes, then nothing. Or "OpenGL version too low" toast. Symptom on macOS: "OrcaSlicer quit unexpectedly" dialog after a macOS 14.5 or 15 update. Symptom on Linux: terminal output version GLIBC_2.38 not found (required by OrcaSlicer.AppImage). Root causes (in priority order): GPU drivers missing OpenGL 4.5 support (pre-2014 Intel iGPUs and older AMD APUs all fail this) Corrupted user config in AppData / Application Support Linux distro libc older than 2.38 (issue #10990, breaks Ubuntu 22.04, Debian 12, Mint 21) macOS sandbox or Local Network permission denied Antivirus quarantine (Defender SmartScreen, McAfee) Fix sequence: Update GPU drivers from the vendor (NVIDIA Game Ready / AMD Adrenalin / Intel Arc Control). Verify with GPU-Z (Windows) or glxinfo | grep "OpenGL version" (Linux). Delete user config: %AppData%\OrcaSlicer\ on Windows, ~/Library/Application Support/OrcaSlicer/ on macOS, ~/.config/OrcaSlicer/ on Linux. Run as Administrator on Windows. Whitelist OrcaSlicer in Defender / your AV. On Linux, download the nightly with the "Ubuntu" suffix (built against Ubuntu 20.04, libc 2.31 compatible) or downgrade to v2.3.0. Last resort on Windows: copy opengl32.dll from a PrusaSlicer install folder into the OrcaSlicer install folder. Forces software fallback rendering. For the GLIBC issue specifically, the Flatpak install path is the safest. Flatpak ships its own libc and works on every modern distro. A3. Settings not saving (the profile reset bug) Symptom: a custom filament or process profile saves during the session but vanishes after closing OrcaSlicer. Or all custom presets revert to factory defaults after an update. Root causes: Stale data from a prior major version conflicting with the new schema Write permission issue on the AppData folder Running stable and nightly side-by-side, both writing to the same config Fix sequence: Close OrcaSlicer fully. Navigate to %AppData%\OrcaSlicer\ (or equivalent) and back up the user subfolder. Delete any *.bak files and OrcaSlicer.conf. Relaunch. If still failing, clean uninstall, delete the entire OrcaSlicer AppData folder, reinstall the latest stable. Save your custom profile under a unique name. Avoid overwriting "Default" presets. After saving, verify the JSON file actually exists at %AppData%\OrcaSlicer\user\<num>\filament\<name>.json. Evidence: GitHub issues #1182, #1435, #3152, #5759, #6545, #980, Discussion #1395, #13482. A4. "Spiral mode wall loops" error Symptom: toast that says "Spiral vase mode requires only one wall loop" (wording varies). Slice button disabled. Root cause: Spiral mode requires a specific configuration. Wall loops = 1, Top shells = 0, Sparse infill = 0%, only one object on the plate. Fix sequence: Process → Strength → Wall loops = 1 Process → Strength → Top shell layers = 0 Process → Strength → Sparse infill density = 0% Make sure only one object is on the plate If error persists despite correct settings, the config is corrupted. Edit %AppData%\Roaming\OrcaSlicer\user\<id>\process\<name>.json, change "spiral_mode": 1 to 0, save, then re-enable spiral mode through the UI. A5. Slow slicing on large models Symptom: slicing a 100MB+ STL takes 10 or more minutes. UI freezes during preview generation. Root causes: Arachne perimeter generator on high-poly meshes (much heavier compute than Classic) Adaptive layer height enabled Tree supports enabled Insufficient RAM (under 16GB) Fix sequence: Process → Quality → Wall generator: change Arachne to Classic Disable adaptive layer height Switch tree supports to Normal supports Decimate the mesh in Blender, target under 500K triangles Close browser, Discord, anything not essential. 16GB RAM minimum. A6. Print time estimate wildly wrong Symptom: slicer says 7h 30min, actual print is 2h 10min. Or vice v [truncated] --- # Section: Settings & Workflow ## OrcaSlicer Settings Hierarchy: Printer, Filament, Process URL: https://orcaslicer.net/orcaslicer-settings-hierarchy/ I bumped my PLA nozzle temp to 215 one Tuesday because a fresh roll of matte was under-extruding, hit Save, and went to bed feeling competent. A week later I noticed every PLA print on every machine in the workshop was running at 215, including the one rated for 200, and I had no idea why. That's the trap nobody warns you about in OrcaSlicer, and once you understand how the slicer actually composes Printer, Filament, and Process presets together at slice time, the "wait, where did my setting go" feeling stops happening. I've been digging through the source headers, the wiki, and a pile of GitHub issues for a while now, and the short version is this: OrcaSlicer doesn't have one big settings object. It has three independent profile families that get merged in a specific order, with a few overrides stacked on top, and the order matters more than most tutorials let on. Let's walk through it the way it actually works. Table of contents The three tiers at a glance Tier 1: Printer settings Tier 2: Filament settings Tier 3: Process settings How the tiers compose: the merge order System presets vs user presets Inheritance via the inherits key Compatibility conditions and hidden dropdowns Per-object overrides via modifier meshes Import, export, and bundle extensions Gotchas pulled from the issue tracker FAQ The three tiers at a glance Open OrcaSlicer and look at the right-hand panel. You'll see three dropdowns stacked on top of each other: Printer, Filament, and Process. Those aren't three views of one config. They're three completely separate preset families. Inside the source, in src/libslic3r/Preset.hpp, they're three distinct entries in the Preset::Type enum: enum Type { TYPE_INVALID, TYPE_PRINT, TYPE_SLA_PRINT, TYPE_FILAMENT, TYPE_SLA_MATERIAL, TYPE_PRINTER, TYPE_COUNT, TYPE_PHYSICAL_PRINTER, TYPE_PLATE, TYPE_MODEL, }; For the FFF (filament) workflow that 99% of us run, the relevant types are TYPE_PRINT (which the UI calls "Process"), TYPE_FILAMENT, and TYPE_PRINTER. The "Print" name is a Slic3r-era leftover and doesn't mean a print job; it means slicing strategy. The other types are either SLA-only (legacy in OrcaSlicer) or for separate concepts like network printer instances. You can see the enum yourself in the Preset.hpp header on GitHub. The three settings tabs at the top of OrcaSlicer's right panel Here's the quick mental model. Printer is the machine. Filament is what's on the spool. Process is the slicing strategy. None of them inherit from another. They get composed together when you hit Slice, and that composition follows a defined merge order that I'll get to in a minute. One thing worth saying upfront: OrcaSlicer didn't invent this three-tier model. It's inherited from PrusaSlicer, which inherited it from Slic3r. If you've ever used PrusaSlicer or SuperSlicer, the concepts map across almost cleanly, which is why the OrcaSlicer dev team explicitly references PrusaSlicer behaviour in Discussion #1090 when answering questions about how presets work. Tier 1: Printer settings, the machine itself Printer is the deepest tier. It describes the physical machine: build volume, number of extruders, nozzle diameter on each, kinematics (CoreXY, bedslinger, delta), max velocities and accelerations, retraction defaults at the firmware level, and the Start G-code / End G-code blocks that wrap every print. The DeepWiki summary describes it as "machine capabilities, extruder config," which is a fair shorthand. Printer settings page showing kinematics and build volume The thing people miss about Printer settings is that they cap what the other tiers can ask for. If your Printer preset says max X/Y velocity is 300 mm/s and your Process preset asks for 500 mm/s outer wall speed, you don't get 500. You get 300, silently, because the firmware would clamp it anyway and the slicer is being honest about it. Same goes for max acceleration. A lot of the "why isn't my print actually going as fast as my Process tab says" questions come down to a conservative Printer preset. Printer settings are also where Printer Variants live. A variant is, per Obico's profile guide, a "particular configuration of a printer model, often tailored to a specific nozzle diameter." That's why your X1C might show up three times in the dropdown: once for 0.2 mm, once for 0.4 mm, once for 0.6 mm. They're separate Printer presets that point at the same physical machine but with different nozzle metadata. Tier 2: Filament settings, what's on the spool Filament covers material properties. Nozzle temperature, bed temperature, retraction (length and speed, which can override printer defaults), flow ratio, cooling fan curve, max volumetric speed, pressure advance values if you tune per material, and filament-specific Start/End G-code. On multi-extruder machines, Filament presets are per-extruder, which is why a tool-changer with five colours has five Filament dropdowns. Filament settings showing temperature and retraction fields This is the tier where I burned myself with the 215 PLA temp. When I bumped that value and clicked Save, I wasn't editing the print job. I was editing my Filament preset (or, more accurately, OrcaSlicer prompted me to save a new user preset based on the current one, and I'd done the same thing the week before without realising). Every subsequent slice that loaded that filament inherited the higher temperature. That's working as designed; the Filament preset is meant to outlive a single print. Tier 3: Process settings, how you slice this print Process is "everything about how to convert this STL into G-code that isn't a function of the machine or the material." Layer height, perimeter count, infill density and pattern, top/bottom solid layers, support generation, speed targets per feature (outer wall, inner wall, infill, travel), seam placement, ironing, fuzzy skin. The wiki sidebar groups Process settings into categories like Quality, Strength, Speed, and Support, and most tutorials follow that grouping. Process settings showing layer height, walls, and infill Process is also the tier most people think of as "the slicer settings," which is why it's the tier they accidentally edit when they mean to tweak a one-off print. If you're saving Process changes from inside a project just to test something, you're populating a user preset that will then appear in your dropdown forever. Worth knowing. How the three tiers compose: the merge order This is the section I wish I'd read three years ago. The DeepWiki configuration system page documents the merge order at slice time, lowest to highest priority: Factory defaults (lowest) Printer preset Print/process preset Project settings Filament presets (per extruder) Per-object/region overrides (highest priority) Read that order again because it's load-bearing. Filament sits ABOVE Process in priority. That's the opposite of what most "Process is your main slicer config" tutorials imply. If your Filament preset specifies a layer height range or a cooling override or a speed cap, that wins over what Process asks for. If your Process preset says 220 nozzle and your Filament preset says 210, you get 210, because Filament merges in after Process. Per-object overrides sit above everything. They're the top of the stack. We'll get to those in the modifier mesh section. The practical takeaway: when a value isn't doing what you expect, ask which tier owns it. Temperature questions are almost always Filament tier. Layer height questions are usually Process tier (unless an object-level override is in play). Machine kinematics questions are Printer tier. If you remember nothing else from this article, remember the order and which tier owns which knob. System presets vs user presets Every preset in your dropdown is either a System preset or a User preset. The distinction is enforced in the source via an is_system flag on the preset object, and DeepWiki describes the two classes like this: System presets: "Read-only, vendor-provided, Always visible if comp [truncated] --- ## OrcaSlicer Keyboard Shortcuts Cheat Sheet (Every Verified Key) URL: https://orcaslicer.net/orcaslicer-keyboard-shortcuts/ I lost an embarrassing amount of time to right-click menus before I learned that pressing M jumps straight to the move gizmo. The shortcut that actually broke my mouse-only habit was Ctrl + R for Slice plate, because I'd spent months hunting for the slice button at the top right every time I tweaked a setting. If you're clicking everything, you're paying a tax measured in seconds per action, and it compounds across a busy plate. I'll walk through every documented OrcaSlicer keyboard shortcut, group by group, the way the official wiki organizes them, and flag the three or four shortcuts almost everyone gets wrong (including a couple I had wrong myself). What's in this guide Global shortcuts (file, edit, the universal commands) Prepare tab: moving, selecting, and the camera The gizmo single-key roster Color painting and the 1-9 filament keys Preview tab: scrubbing layers and reading G-code Objects List shortcuts Windows vs macOS vs Linux Cheat sheet FAQ Global shortcuts: file, edit, and the universal commands These are the keys that work regardless of which tab you're on. They're the ones I'd hammer into muscle memory first because they cover the actions you repeat all day. Every entry below is pulled verbatim from the official OrcaSlicer keyboard shortcuts wiki. Ctrl + N: New Project Ctrl + O: Open Project Ctrl + S: Save Project Ctrl + Shift + S: Save Project as Ctrl + I: Import geometry data from STL/STEP/3MF/OBJ/AMF files Ctrl + G: Export plate sliced file Ctrl + R: Slice plate Ctrl + Shift + G: Print plate Ctrl + X: Cut Ctrl + C: Copy to clipboard Ctrl + V: Paste from clipboard Ctrl + P: Preferences Ctrl + M: Show/Hide 3Dconnexion devices settings Ctrl + Tab: Switch table page Del: Delete selected ?: Show keyboard shortcuts list The trap here is the slice-vs-export confusion. Ctrl + R slices, Ctrl + G exports the sliced file, and Ctrl + Shift + G sends the print. I used to assume Ctrl + G would slice because "G" feels like "go," but it doesn't. Memorize R-for-Slice and you'll stop sending half-sliced jobs. The other one worth highlighting is ?, which opens an in-app cheat sheet listing every binding for your current build. As of OrcaSlicer 2.3.1, the popup also reflects swapped mouse buttons if you've remapped LMB and RMB in preferences. Prepare tab: moving, selecting, and the camera The Prepare tab is where the keyboard pays for itself fastest. The four tabs that make up OrcaSlicer's UI all have their own bindings, but Prepare is where you spend most of your time arranging, scaling, and orienting parts before you slice. The camera lives on the mouse: left mouse rotates the view, right mouse pans, and the wheel zooms. Once you've got that, the keyboard handles selection and movement. A: Arrange all objects Shift + A: Arrange objects on selected plates Q: Auto orients selected objects Shift + Q: Auto orients all objects on active plate Shift + Tab: Collapse/Expand sidebar Ctrl + Any arrow: Movement in camera space Alt + Left mouse button: Select a part Ctrl + Left mouse button: Select multiple objects Shift + Left mouse button: Select objects by rectangle Arrow Up: Move selection 10 mm positive Y direction Arrow Down: Move selection 10 mm negative Y direction Arrow Left: Move selection 10 mm negative X direction Arrow Right: Move selection 10 mm positive X direction Shift + Any arrow: Movement step set to 1 mm Esc: Deselect all Ctrl + A: Select all objects Ctrl + D: Delete all Ctrl + Z: Undo Ctrl + Y: Redo I: Zoom in O: Zoom out Tab: Switch between Prepare/Preview Arrow keys default to 10 mm steps. Hold Shift and they drop to 1 mm. That's the one I wish I'd known on day one. I'd been clicking into the position field and typing numbers to nudge parts into corners, and the Shift+arrow combo is far faster. The camera presets are a separate cluster you should at least try once: Ctrl + 0: Camera view: Default Ctrl + 1: Camera view: Top Ctrl + 2: Camera view: Bottom Ctrl + 3: Camera view: Front Ctrl + 4: Camera view: Behind Ctrl + 5: Camera Angle: Left side Ctrl + 6: Camera Angle: Right side If you're checking whether a feature sits flush on the bed, Ctrl + 1 for top-down and Ctrl + 3 for front view will save you a lot of right-mouse orbiting. The gizmo single-key roster This section delivers the biggest practical speedup. The left-side gizmo bar holds move, scale, rotate, cut, and paint tools, and each has a single-key binding. No need to aim at a tiny icon when you can tap a letter. M: Gizmo move R: Gizmo rotate S: Gizmo scale F: Gizmo place face on bed C: Gizmo cut B: Gizmo mesh boolean H: Gizmo FDM paint-on fuzzy skin L: Gizmo SLA support points P: Gizmo FDM paint-on seam T: Gizmo text emboss/engrave U: Gizmo measure Y: Gizmo assemble E: Gizmo brim ears Three corrections trip up users coming from older guides. First, L is SLA support points, not FDM support painting. If you tried L expecting the FDM support brush, you hit the wrong gizmo. The wiki lists no single-key shortcut for FDM support paint; reach it through the gizmo bar icon. Second, N does nothing. I've seen it claimed as the Color Painting hotkey in old posts. It isn't documented in the wiki and pressing it does nothing. Color Painting is a gizmo-bar icon for now. Third, C behaves differently by context. In Prepare, it activates the cut gizmo. In Preview, it toggles the G-code window. Same key, two behaviors. While we're here, the cut and boolean gizmos are the entry points for modifier meshes. Color painting and the 1-9 filament keys Here's the shortcut most people get wrong, and I include myself. The keys 1-9 are not brush sizes. The wiki is unambiguous: in Prepare, 1-9 sets the filament for the selected object or part. In the Objects List, the same range sets the extruder number. Inside the Color Painting gizmo, those keys pick which filament index you're painting with. Brush size lives on the mouse wheel. Specifically: Ctrl + Mouse wheel: Support/Color Painting: adjust pen radius Alt + Mouse wheel: Support/Color Painting: adjust section position Esc: Deselect All (works in the toolbar context too) Shift + drag, Move: press to snap by 1 mm If you've been wrestling with brush size by clicking the +/- icons, stop. Hold Ctrl, spin the wheel, done. The full color painting workflow changes once you internalize those two modifiers. One subtle bit: inside the gizmo, 1-9 switches the active brush filament. Outside the gizmo with a part selected, the same keys assign that part to a filament. Same range, different effect depending on focus. Preview tab: scrubbing layers and reading G-code Once a plate has been sliced, the Preview tab is where you sanity-check layer transitions, look for retraction artifacts, and scrub through tricky overhangs to make sure the toolpath is going where you think it should. Preview tab vertical slider with G-code window toggle The Preview shortcuts revolve around the two sliders (vertical for layer, horizontal for move-within-layer) and the G-code window. Arrow Up: Vertical slider: Move active thumb Up Arrow Down: Vertical slider: Move active thumb Down Arrow Left: Horizontal slider: Move active thumb Left Arrow Right: Horizontal slider: Move active thumb Right L: On/Off one layer mode of vertical slider C: On/Off G-code window Tab: Switch between Prepare/Preview Shift + Any arrow: Move slider 5x faster Shift + Mouse wheel: Move slider 5x faster Ctrl + Any arrow: Move slider 5x faster Ctrl + Mouse wheel: Move slider 5x faster Home: Horizontal slider: Move to start position End: Horizontal slider: Move to last position L for one-layer mode is the one I use constantly. It isolates the current layer so you see exactly which moves happen on that pass. Pair it with C to open the G-code panel and you've got a fast diagnostic loop without touching the mouse. Objects List shortcuts The Objects List panel on the right side has its own keyboard set, mostly overlapping with the global shortcuts but with one extra binding worth knowing. 1-9: Set extruder number for objects and [truncated] --- ## How to Update OrcaSlicer Without Losing Your Profiles (2026) URL: https://orcaslicer.net/orcaslicer-update-without-losing-profiles/ I lost a year of dialed-in process presets the night I updated from 1.9.1 to a 2.0 beta because I was too excited about scarf seams to back anything up first. I'm not the only one: issue #4459 tells almost the same story, issue #8982 shows it again when a user uninstalled 2.3.0-rc on prompt during the 2.3.0 install, and as recently as Feb 24 2026 issue #12452 logged a 2.3.2 beta2 user opening the slicer to find every filament profile gone except one stranded "unsupported" entry. So no, you're not paranoid. Profile loss is rare, but it's real and documented. In almost every reported case the profiles were still sitting on disk. Here's the procedure I run every time now, on Windows, macOS, and Linux. Table of contents Why profile loss happens during an update Where OrcaSlicer actually stores your profiles The in-app backup: File > Export > Preset Configs The manual backup: copy the user folder Updating on Windows, macOS, and Linux Restoring after an update went wrong Rolling back to the previous version Vendor preset re-sync after a major update FAQ Why profile loss happens during an OrcaSlicer update OrcaSlicer's installer isn't supposed to wipe your user data dir, and in most updates it doesn't. But the tracker tells an honest story. Issue #1182 caught a 1.6.3 overwrite in May 2023. Issue #4679 in March 2024 is the one that worries me most: the user's profiles were still in %appdata% after the update, the app just refused to load them. Issue #7679 describes a 2.2.0 portable build clobbering the installed 2.1.1 folder structure. All closed as "not planned". The pattern is consistent. Profiles are usually still on disk, but one of three things broke: the installer deleted a subfolder, an inheritance chain snapped (a parent profile got overwritten by a vendor refresh, so any child preset that inherits from it errors out, as the PeterVRC writeup explains), or, the sneakiest, the new build started writing to a different numbered user-id folder than the old one was reading from. Discussion #5033 nailed that last cause: "the new installation is using a different number than the old." Where OrcaSlicer actually stores your profiles Every backup and restore path starts here. Discussion #1594 is the canonical source for where the user data dir lives on each OS. These are the paths verbatim: Windows: %APPDATA%\OrcaSlicer\ (resolves to C:\Users\your_name\AppData\Roaming\OrcaSlicer) macOS: ~/Library/Application Support/OrcaSlicer/ Linux (native): ~/.config/OrcaSlicer Linux (Flatpak): ~/.var/app/io.github.softfever.OrcaSlicer/config/OrcaSlicer If you don't want to type any of that, open Help > Show Configuration Folder inside OrcaSlicer. That pops open File Explorer (or Finder, or your file manager) right at the correct path for whichever build you're using. Discussion #530 calls this out as the no-typing way to find your data dir on any OS, and it's the move I use every time. Windows File Explorer showing C:\Users\\AppData\Roaming\OrcaSlicer with the user, system, and numbered subfolder visible Open that folder and you'll see two subdirectories that matter. user/ is where your custom presets live, and inside it there's a numbered subfolder per account. Discussion #5033 confirms it: "Within C:\Users\<username>\AppData\Roaming\OrcaSlicer\user there is a numbered folder that contains your user presets." Those presets are stored as .json files. The sibling system/ folder holds vendor and system presets, and the bundled vendor source ships separately at C:\Program Files\OrcaSlicer\resources\profiles\, which is part of the installer, not your user data. Quick honesty check: people online sometimes list presets/ or cache/ as top-level folders here. I haven't found a primary source that names either, so I'm not going to claim they're there. user/, system/, and the numbered subfolder are what's confirmed. The in-app backup: File > Export > Preset Configs This is the first thing you do before any update. It's not a full backup, but it's the only fully GUI-driven export OrcaSlicer ships. From inside OrcaSlicer, click File > Export > Preset Configs. The official wiki page documents the five outputs: OrcaSlicer File menu open showing File > Export > Preset Configs highlighted Printer bundle (.orca_printer): exports a printer plus all the filament and process presets that belong to that printer. This is what you want if you're sharing a fully tuned machine with someone, or if you only print on one machine. Filament bundle (.orca_filament): exports your filament presets as a single file. Note: the wiki page actually renders this extension as the plural form, but on disk and in practice the extension is singular. Use .orca_filament when you're naming things. Printer presets zip: a flat zip of your user printer presets. Filament presets zip: a flat zip of your user filament presets. Process presets zip: a flat zip of your user process presets. Export Preset Bundle dialog with the four format options visible (printer bundle, filament bundle, printer presets zip, filament presets zip, process presets zip) Here's the catch nobody tells you. There is no "export everything in one click" button. Issue #3012 was opened to request exactly that, and it was closed as not planned. So if you've got five printers, you're either running Export Preset Configs five times or falling back to the manual folder copy below. I do both. The import path back is symmetrical: File > Import > Preset Configs, point at the bundle, done. The wiki phrases it as "the application will extract the presets contained in the bundle and make them available for use in projects." For a deeper walkthrough see my import and export profiles guide. The manual backup: copy the user folder This is the one I actually trust. Close OrcaSlicer fully (check the system tray on Windows, quit it from the dock on macOS), then go to the config folder via Help > Show Configuration Folder while OrcaSlicer's still open, or just navigate there yourself after you've closed it. Help menu in OrcaSlicer showing Show Configuration Folder option Copy the entire user/ subfolder somewhere safe. I name mine with a date and version, like user_backup_2026-05-15_pre_2.3.2.zip, and drop it on my NAS plus a USB stick. Because there's no built-in "export all" per issue #3012, this is the only real "back up everything" option. It captures every numbered subfolder, every .json, every inheritance parent your custom profiles depend on. Glance at the numbered subfolder name while you're there: take a screenshot if you don't trust yourself to remember. If you like belt and suspenders, there's a third option from PR #6780 (Sep 2024): portable mode. If you create a folder called data_dir next to the OrcaSlicer executable, the app uses that as its data location instead of %APPDATA%. I keep a portable copy on a non-system partition for nightly testing so an update can never touch my main install's profiles. Updating on Windows, macOS, and Linux without losing presets The update procedures themselves come straight from the SoftFever README. None of them are supposed to wipe your data dir. None of them are guaranteed not to. Always run the backup first. Windows Grab the Windows installer exe from the releases page and run it. The installer overwrites C:\Program Files\OrcaSlicer\ but is not supposed to delete %APPDATA%\OrcaSlicer\. If you've got winget, winget install --id=SoftFever.OrcaSlicer -e gets the same result. Microsoft Edge WebView2 and Visual C++ 2019 redistributables are on the releases page too if you need them. See the Windows install guide for a walkthrough, and the OrcaSlicer download page for current links. The lesson from issue #8982 is worth internalising: if the installer prompts you to uninstall a previous release candidate or beta, your custom user data may not survive that uninstall step. Don't accept blindly. Make sure your backup is on a different drive first. macOS Download the DMG (arm64 for Apple Silico [truncated] --- ## OrcaSlicer Cooling and Fan Curves Explained (2.3.x Guide) URL: https://orcaslicer.net/orcaslicer-cooling-fan-curves/ My first serious PETG print sagged like wet cardboard because I'd left OrcaSlicer's cooling tab on the default PLA-style fan curve, and I'd assumed (wrong) that fan settings lived under Process. They don't. I spent the next hour digging through the wiki, the JSON profiles in the OrcaSlicer repo, and a couple of late-night GitHub issues before I figured out that part cooling is a Filament setting, not a Process setting, and that every filament family wants a different shape of curve. This guide is the brain dump I wish I'd had that night, written for the 2.3.x release line and pulled almost entirely from the live wiki and the base JSON profiles in the OrcaSlicer source tree. I'm going to walk through every cooling setting OrcaSlicer exposes, show the verified per-filament defaults straight from the repo, explain how overhang and bridge fan logic actually works (it's not what most people think), and finish with the failure modes I've seen on my own bench. There's no "perfect" fan curve waiting at the end of this article. There is, however, a clear mental model that will let you build one for your printer, your filament, and your specific part. Table of contents Why cooling makes or breaks a print Where the cooling settings actually live in OrcaSlicer 2.3.x Every cooling setting explained, in order Per-filament fan defaults pulled from the source Overhang and bridge fan tuning Layer-time slowdown logic Aux fan, chamber fan, exhaust fan Per-printer notes (Bambu, Klipper, Marlin) Troubleshooting cooling-related failures Calibrating without a built-in Cooling Tower Known cooling bugs and feature requests in 2.3.x FAQ Why cooling makes or breaks a print Cooling is the single setting category that decides whether your overhangs droop, whether your bridges sag halfway across a span, whether PETG pulls long whiskers between travel moves, and whether ABS lifts off the bed like a sad taco. It's also the category most people get wrong on their first few prints because OrcaSlicer puts the cooling tab in a non-obvious place, and because the default values are tuned for PLA and only PLA. If you switch from PLA to PETG without touching the Cooling tab, you're using a curve that's roughly twice as aggressive as the base JSON profile recommends for that material. Here's the rough rule of thumb I work from on my own bench. PLA wants the fan slammed on as early as possible because the polymer recrystallizes fast and benefits from quick solidification. ABS, ASA, and PC want almost no part cooling, and they actively want a warm enclosure to slow contraction. PETG sits in the middle and is genuinely case-by-case: bridges and steep overhangs need fan, but flat infill layers usually don't. TPU is its own thing entirely. Aggressive airflow on a flexible filament weakens layer bonds and can drag the print sideways since the rubbery filament doesn't tolerate the same mechanical stress as rigid plastics. OrcaSlicer's cooling model is genuinely richer than vanilla Bambu Studio or PrusaSlicer. You get separate fan speeds for overhangs and external bridges versus internal bridges, a configurable "no cooling for the first X layers" plus a linear ramp-up to "full fan speed at layer N," a part-cooling auxiliary fan channel on the M106 P2 G-code address (used on Bambu hardware and any Klipper rig that's configured for it), and a separate exhaust or air-filtration fan on M106 P3. The cooling logic also drives OrcaSlicer's "slow down for layer cooling" behavior, which enforces a minimum layer time by automatically reducing print speed. That's crucial for tall thin parts that otherwise stack hot layers on top of each other with no time to cool. One thing I want to set straight before we dig in: OrcaSlicer didn't invent the overhang-fan concept. PrusaSlicer had overhang_fan_threshold and overhang_fan_speed first, and OrcaSlicer inherited and extended the idea (notably with separate handling of internal bridges). The 2.3.0 release notes don't actually mention any cooling or overhang-fan changes either, so if you've read a blog post crediting 2.3.0 with the overhang split, that's wrong. The split is documented on the wiki, it's just not tied to that specific release in the changelog. Where the cooling settings actually live in OrcaSlicer 2.3.x Open OrcaSlicer, look at the right-hand panel, and click the Filament icon (it's the spool, not the gear). Inside the Filament settings dialog, you'll find a tab called Cooling. That's where 90% of the settings in this article live. The wiki page for this group is Material Cooling, and the URL slug itself (material_cooling) confirms it's a material setting, not a process setting. This trips people up because Cura, for example, exposes fan settings under print profile, and PrusaSlicer puts the slowdown and fan logic under Filament Settings but in a different visual location. In OrcaSlicer, Cooling is filament-scoped because the slicer assumes (correctly, in my experience) that the right fan curve depends mostly on the polymer you're extruding, not on the part you're printing. OrcaSlicer Filament settings panel with the Cooling tab highlighted There's a second, smaller cooling page tucked under Printer Settings → Basic Information → Cooling Fan. Don't confuse it with the main Cooling tab. The machine-level page only covers timing parameters: fan speed-up time ("Start the fan this number of seconds earlier than its target start time") and fan kick-start time ("Emit a max fan speed command for this amount of seconds before reducing to target speed to kick-start the cooling fan"). Those two settings exist because real fans have spin-up inertia. They're a printer characteristic, not a filament characteristic. If your fan is slow to come up to speed, raise the kick-start time. Otherwise leave it alone. There's a full reference to that page on the OrcaSlicer wiki Cooling Fan entry. If you're new to the slicer's overall layout, the broader walkthrough at our OrcaSlicer settings master guide covers the Filament vs Process vs Printer tab split in more detail. For this article, just remember: part cooling sliders live under Filament, fan-timing seconds live under Printer. Every cooling setting explained, in order Here's every setting on the Material Cooling page, in the order the wiki lists them. I'm using the verbatim setting labels and the JSON keys, because the labels sometimes change between releases but the keys don't. Material Cooling for Specific Layer No cooling for the first (key close_fan_the_first_x_layers). The wiki defines it as "Number of initial layers during which part-cooling fans are disabled." This exists because bed adhesion depends on the first layers staying warm and pliable. Cool them off and they shrink, lift at the edges, and the rest of your print is doomed. The common-base default is 3 layers. PLA overrides this down to 1 (PLA's first-layer warp risk is low). PETG inherits the 3. ABS, ASA, PC, and TPU also inherit the 3, but if you're printing without an enclosure on a high-temp material, you may want to raise it further. Full fan speed at layer (key full_fan_speed_layer). Verbatim: "Fan speed is increased linearly from 0% starting at the layer specified by close_fan_the_first_x_layers up to the maximum part-cooling speed at this specified layer." This is the linear ramp. If close_fan_the_first_x_layers is 3 and full_fan_speed_layer is 6, the fan will go from 0% at the start of layer 4 to 100% at layer 6, scaling linearly across those three layers. Most users leave this at 0 (which disables the ramp and just snaps to full speed at the layer after the close-fan-first layers). I prefer a short ramp on PLA prints with delicate first features. Material Part Cooling Fan Min fan speed threshold and Max fan speed threshold. These are the heart of OrcaSlicer's auto-cooling logic. Verbatim from the wiki: "When a layer's estimated print time falls below the minimum time, the slicer begins enabling the part-cooling fan" and "Layers with [truncated] --- ## OrcaSlicer M600 Manual Filament Change: Single-Extruder Multi-Color (2026) URL: https://orcaslicer.net/orcaslicer-m600-manual-filament-change-2/ I was halfway through a sign that needed white letters on a red background, single-extruder Ender 3, no AMS, no MMU, no fancy four-color setup. The plan was simple in my head: print the base in red, pause at layer 22, swap to white, finish the top, peel a clean two-color part off the bed. The execution was less simple. My first attempt didn't pause at all because I had the wrong G-code in the Machine settings. My second attempt paused, but the nozzle just sat where it was and oozed a fat blob onto the top layer. The third attempt finally worked, and that's the run that taught me what M600 actually is, what OrcaSlicer is actually emitting when you right-click that little slider in Preview, and why the same workflow behaves three different ways on Marlin, Klipper, and Bambu firmware. This is the guide I wish I'd had open on a second monitor that night. I'm going to walk through the whole pause-for-color flow in OrcaSlicer 2.2.x and 2.3.x, but I'm also going to drag the curtain back on the firmware side so you understand why the same slicer button does very different things depending on what board is running your printer. If you've ever sliced a print, clicked "Add color change" in Preview, hit go, and watched the printer cheerfully ignore your pause, this article is going to make a lot more sense of that than the official wiki ever did. Table of contents What M600 actually is (and why it isn't universal) Where the "Add color change" button lives in OrcaSlicer The full workflow: prepare to print, layer by layer Marlin specifics: ADVANCED_PAUSE_FEATURE and the lengths that matter Klipper specifics: the M600 macro you have to write yourself Bambu specifics: why it doesn't run M600 at all Common mistakes I see (and have made) Tips for clean swaps FAQ What M600 actually is (and why it isn't universal) M600 has a clean Wikipedia-style definition that's basically useless once you sit in front of an actual printer. The clean version is: "M600 is the Filament Change G-code command. It parks the head, retracts the old filament, waits for the user to swap, primes the new filament, then resumes." The messy version, the one you need to know, is that M600 is firmware-defined, not standard. The slicer can emit the three characters M, 6, 0, 0 into your file all day long, and what happens at the printer end depends entirely on whether the firmware on the board has any idea what to do with them. Here's how the three big firmware families actually handle it. Marlin: built-in, but locked behind a compile flag Marlin has M600 as a real, first-class G-code, but the catch is that the entire pause-for-filament-change subsystem lives behind a compile-time flag called ADVANCED_PAUSE_FEATURE in Configuration_adv.h. If the person who built your firmware didn't enable that flag before flashing, M600 is unknown to the board. Your slicer can ship a beautifully formatted file with M600 at layer 22 and the printer will treat it as a comment, blow right past it, and finish the print in whatever color started the job. The flag itself has a small history that's worth knowing because it changes what version-checking advice you take seriously. It used to be called FILAMENT_CHANGE_FEATURE. It was renamed to ADVANCED_PAUSE_FEATURE in Marlin pull request #6407, which was merged into the bugfix-1.1.x branch on May 26, 2017. From the 1.1.x line forward and throughout the entire 2.x series, ADVANCED_PAUSE_FEATURE is the gate. Anyone telling you "it works on Marlin X.Y.Z" without checking that flag is guessing. The version doesn't gate it; the build does. When M600 fires on a properly compiled Marlin board, the sequence is: park the head at NOZZLE_PARK_POINT, retract by PAUSE_PARK_RETRACT_LENGTH (2 mm default), unload by FILAMENT_CHANGE_UNLOAD_LENGTH (100 mm default, but you'll want about 600 mm on a Bowden Ender 3), beep, wait for the user to confirm at the LCD, fast-load the new filament by FILAMENT_CHANGE_FAST_LOAD_LENGTH, slow-prime by FILAMENT_CHANGE_SLOW_LOAD_LENGTH, optionally continue-purge with ADVANCED_PAUSE_PURGE_LENGTH (50 mm default), then resume. The full Marlin reference for M600's inline parameters is at marlinfw.org/docs/gcode/M600.html, and the companion commands are M701 (load filament), M702 (unload filament), and M603 (set the defaults persistently, savable with M500 to EEPROM). Don't try to remember every parameter. The ones you'll actually touch are E (initial retract), L (load length), U (unload length), X, Y, Z (park position), and R (resume temperature). Klipper: not built-in, you write the macro This is the one that catches more people out, because they read "Klipper has every G-code you could ever want" online and then send a file with M600 in it and watch their printer do absolutely nothing. Klipper does not have M600 built in. M600 doesn't exist until you define it as a [gcode_macro M600] in printer.cfg (or in an included file). The canonical sample, the one that lives in Klipper's own repository at klipper/config/sample-macros.cfg, looks like this: [gcode_macro M600] gcode: {% set X = params.X|default(50)|float %} {% set Y = params.Y|default(0)|float %} {% set Z = params.Z|default(10)|float %} SAVE_GCODE_STATE NAME=M600_state PAUSE G91 G1 E-.8 F2700 G1 Z{Z} G90 G1 X{X} Y{Y} F3000 G91 G1 E-50 F1000 RESTORE_GCODE_STATE NAME=M600_state For that to work you also need [pause_resume] declared in your config, otherwise the PAUSE and RESUME commands themselves don't exist. A common pattern from Ellis' Print Tuning Guide is to make M600 a one-line alias that delegates to a heavyweight custom PAUSE macro: [gcode_macro M600] gcode: PAUSE Ellis' PAUSE saves state, parks the nozzle near the front-left corner, lifts Z, can either cool the hotend or keep it at target, extends the idle timeout to 12 hours so the heater doesn't shut off while you fight a stuck spool, and disables the runout sensor for the duration. His RESUME re-heats with M109 S{etemp|int}, restores position, and continues. If you want to copy a known-good pattern instead of reinventing one, his version is the cleanest I've found and the source is at ellis3dp.com. One thing that bites Klipper users specifically: Klipper does NOT automatically purge on resume. After the swap, you have to either extrude manually from the Mainsail or Fluidd extruder panel (the little plus button with a length field, usually pre-set to 5 or 10 mm) or call a LOAD_FILAMENT macro if you've defined one. If you just click RESUME, Klipper will move straight back to your print path and start drawing with whatever's currently in the nozzle, bleed-through and all. Bambu Lab: doesn't run M600 at all This is the one that surprises Bambu owners who came over from a Creality or Voron background. Bambu Lab firmware on the X1C, X1E, P1S, P1P, A1, and A1 Mini does not execute M600. The Bambu firmware has its own pause-and-change-filament command set internally, and the M600 line in a sliced file simply does nothing. When you slice a model in OrcaSlicer with a Bambu printer profile selected and you right-click the layer slider to add a color change, OrcaSlicer does not actually emit M600. It emits the Bambu pause sequence that's configured in that profile's "Change filament G-code" field (usually an M400 U1-style command that the Bambu firmware recognizes as "stop and prompt the user"). The printer then halts at the layer and a "filament has run out" or "change filament" dialog appears on the touchscreen. The community-verified critical step here is: when that prompt appears, you should NOT tap "Resume Printing" first. Tap X or Back to dismiss the prompt, navigate to the Filament menu on the touchscreen, run Unload, swap the spool manually, run Load, purge until the color runs clean (on the X1C you can purge directly into the poop chute), and only then return to the paused screen and tap Resume. The whole flow is documented in the long-running Bambu forum thread at forum.bambulab.com/t/filament-change-without-ams-mid-print/2589, which is worth scroll [truncated] --- ## OrcaSlicer Wipe Tower & Prime Tower Settings: Multi-Color Guide (2026) URL: https://orcaslicer.net/orcaslicer-wipe-prime-tower-2/ The first multi-color print I ever sliced in OrcaSlicer came out looking like someone smudged ketchup across the side of a white toy. The white surface had pink streaks running through the first few millimeters of every layer change, and I had no idea why. The tower next to my model looked fine. The colors in the slicer preview looked perfect. But the actual print? A mess. It turned out I had the same problem almost every multi-color beginner runs into. I'd been quietly nudging the flush multiplier down to save filament, and the purge volume had dropped below the point where the hotend could actually clear the previous color before laying down the next one. That smudgy result is the exact thing the wipe tower (or prime tower, depending on which panel you're looking at) is built to prevent. And if you've ever wondered why OrcaSlicer uses two different names for what looks like the same thing, or why your tower is sometimes twice as tall as your model, or why the AMS doesn't magically eliminate that ugly purge structure next to your print, this guide is going to walk you through all of it. I'll cover every setting in the panel, the matrix that controls purge volumes, the three wall types, and how to actually tune the thing instead of just leaving the defaults and hoping. I'll also call out the GitHub issues that are still open in 2.3.x so you know what's a real limitation versus what's just a setting you haven't found yet. Table of Contents Prime Tower vs Wipe Tower: Why OrcaSlicer Uses Both Names Why the Tower Is Needed at All Where the Settings Live in OrcaSlicer Settings Walkthrough: Simple Mode Settings Walkthrough: Advanced Mode The Three Wall Types: Rectangle, Cone, Rib Purge Volume Tuning and the Flush Matrix AMS vs Non-AMS Behavior When You Can Skip or Minimize the Tower Common Problems and Fixes How to Calibrate Purge Volume With a Test Print FAQ Wrap-Up Prime Tower vs Wipe Tower: Why OrcaSlicer Uses Both Names Let's clear this up before anything else, because almost every confused forum post about this feature starts here. OrcaSlicer uses the term "Prime Tower" in one place and "Wipe Tower" in another, and a lot of new users assume those are two different features they need to configure separately. They're not. It's the same physical structure with two names, and the naming inconsistency is inherited from the slicer's two ancestors. Here's the breakdown. In Print Settings > Others, you'll see a section called Prime Tower. That panel holds the per-print toggles: enable on or off, width, brim, wall type, prime volume, and everything to do with the shape and behavior of the structure on a print-by-print basis. This is the language Bambu Studio uses, and Orca inherited it. In Printer Settings > Multimaterial, you'll see a section called Wipe Tower. That panel holds the printer-level configuration: X position, Y position, rotation angle, which extruder owns the perimeter, and a few flush-related toggles. This language comes from PrusaSlicer, which has called it a wipe tower since the original MMU days. And if you dig into the config keys themselves (the ones you'd see if you exported a profile and opened it in a text editor), the naming is split too. You'll find enable_prime_tower, prime_tower_width, prime_volume, prime_tower_brim_width on one hand, and wipe_tower_x, wipe_tower_y, wipe_tower_rotation_angle, wipe_tower_extruder, wipe_tower_wall_type, wipe_tower_cone_angle on the other. Same feature, two prefixes, no clean rule for which is which. For the rest of this article I'll use "the tower" when I'm talking about it generally, and I'll use the exact UI label when I'm pointing at a specific setting. If you remember nothing else from this section, remember this: the structure that gets printed next to your model to absorb purge is the tower, and you configure parts of it in two different panels because of OrcaSlicer's mixed lineage. The Prime Tower panel under Print Settings. The matching Wipe Tower panel lives under Printer Settings and holds position, rotation, and extruder fields. Why the Tower Is Needed at All To understand why you can't just delete this structure and save filament, you need to think about what's happening inside the hotend during a filament swap. When the printer decides to switch from filament A to filament B, it retracts A out of the melt zone, feeds B in, and starts pushing B toward the nozzle. But the hotend isn't empty. There's still a slug of melted A in the melt zone, in the throat, and in the bore of the nozzle itself. As B pushes through, the first material that comes out isn't pure B at all. It's mostly leftover A, then a transitional gradient of A and B mixed together, and only after a certain volume of filament has passed through do you finally get pure B coming out the nozzle. If you skipped the tower and let that gradient land on your model, you'd see exactly what I described in the intro. Streaks of the previous color smeared across the new surface for the first few millimeters of every perimeter after a swap. On a dark-to-light transition like black to white, you'd see grey stripes. On a red-to-yellow, you'd see orange smudges. It's not subtle and it's not fixable in post-processing. So the tower's primary job is to absorb that contaminated transition material somewhere off the model. The slicer routes the nozzle to the tower after every toolchange, extrudes a known volume of filament into the tower (purging the old color and priming the new one), and only then sends the toolhead back to the actual part where it lays down the now-clean new color. There are a couple of secondary reasons too. After a toolchange the filament pressure in the hotend is unstable, and priming a known volume into the tower lets that pressure equalize before the printer tries to start a fresh perimeter. The tower also guarantees the nozzle isn't dry or under-extruding when it starts on the part. And for IDEX and toolchanger setups (like the Prusa XL or Snapmaker U1), the tower gives idle tools a known clean starting point too, which matters for tool reliability over a long print. The wiki phrasing from the Bambu Lab docs (which Orca largely inherits) puts it like this: the tower "cleans up the residue on the nozzle and stabilizes the chamber pressure inside the nozzle, in order to avoid appearance defects when printing objects." That's the formal version. The informal version is: without it, your multi-color print looks like crap. Where the Settings Live in OrcaSlicer Before we go through every parameter, here's the lay of the land so you know where to click. Print Settings > Others > Prime Tower is where you find the per-print toggles. This panel changes when you switch process profiles (so a "0.20mm Standard" profile and a "0.16mm Optimal" profile can have different tower widths if you've tuned them). The settings here are: enable, width, prime volume, and (when you switch the UI mode to Advanced) brim width, rotation angle, max bridging distance, purge line spacing, extra flow, max print speed, wall type, cone angle, rib parameters, fillet wall, and no-sparse-layers. Printer Settings > Multimaterial > Wipe Tower is where you find the physical placement and extruder assignment. This panel changes when you switch printer profiles. The settings here are: X position, Y position, rotation angle (yes, also appears here on some printers), and extruder. Filaments > Flushing Volumes (or the equivalent button in the filament dialog) is where you find the matrix of mm-cubed values that controls how much filament gets purged for each pair of colors. There's also a global multiplier that scales the entire matrix at once. The flush-into destination toggles (flush into infill, into objects, into support) live in a slightly different place depending on the OrcaSlicer version: in 2.3.x they're under Print Settings > Multimaterial alongside the prime tower section, but check the search box at the top of the settings panel [truncated] --- ## OrcaSlicer Color Painting: How to Paint Multi-Color Models (2026) URL: https://orcaslicer.net/orcaslicer-color-painting/ My first multi-color print in OrcaSlicer was a tiny pixel-art keycap, and I spent maybe ninety minutes painting individual triangles before I realised the Fill tool exists. I'd been clicking facet by facet like a person carving runes into bark. When the keycap finally came off the plate with three clean colours, I felt smug for about ten seconds before noticing the purge waste tower next to it was bigger than the keycap itself. That's color painting in OrcaSlicer in one paragraph. It's powerful, it's not difficult once you understand what the five brush tools actually do, and it'll teach you very fast that purge volume is the real cost of multi-colour 3D printing. I've now painted enough Venn diagrams, Pokeballs, name plates, and Hextraction tiles to have a clear sense of which workflows survive contact with reality and which ones look great on YouTube but collapse the moment your printer doesn't have an AMS bolted to its side. This guide covers what color painting actually is inside OrcaSlicer 2.3.x, the five brush tools you'll use to mark up models, how painted regions map to filament slots on hardware that ranges from a Bambu X1C with four chained AMS units down to a stock Creality Ender 3 with a Klipper macro. I'll also cover the stuff the official wiki glosses over: paint depth (it doesn't exist yet), MMU3 (use PrusaSlicer instead), and how to keep purge volumes from eating an entire spool on a four-colour Benchy. What's in this guide What color painting actually is (and what it isn't) Finding the Color Paint mode in OrcaSlicer 2.3.x The five brush tools, one by one Height range, color filter, and the missing "paint depth" Importing colors from images: HueForge, Fuzzyficator, and the myth of native projection Painting workflow on a Bambu printer with AMS Painting workflow on a single-extruder printer with M600 or PAUSE IDEX dual-extruder workflow, plus the MMU3 sidebar Flushing volumes and the purge matrix Where the prime tower fits in Common mistakes (and how I've fixed them) Per-printer notes: X1C, A1, J1, SV04, Voron, Ender, Anycubic FAQ What color painting actually is (and what it isn't) Color Painting is the in-slicer mode that lets you mark different regions of a single mesh with different filament slots, so the slicer generates filament-change events at the right points during the print. The official OrcaSlicer wiki puts it like this: "Color Painting allows you to paint specific areas of your 3D model with different colors in the Prepare tab. This feature is useful for visualizing different sections of your model, planning multi-material prints, or simply adding a creative touch to your design." That's accurate, but it understates how distinct the workflow is from two other things people confuse it with. The first is multi-object color assignment. If your scene has three separate STL files on the build plate, you can right-click each object and assign it to a different filament slot. That's not painting; that's by-object coloring. It's the right answer when each colour belongs to a completely separate part. The painted-region workflow is for when you've got one mesh and you want different surfaces of that mesh to be different colours, like the differently-coloured rings of a 3-color Venn diagram printed as one model. The second confusion is what people call "MMU split" or "AMS color split." That's the automatic slot management that happens when a 3MF you downloaded from MakerWorld or Printables already encodes per-region colours. OrcaSlicer reads the file, shows you the painted regions, and lets you remap which physical filament slot maps to each painted color. You aren't painting anything; you're consuming someone else's paint job. The reassignment happens from the Filament panel before slicing. Color Painting is the only one of those three that takes a single contiguous mesh and puts multiple colors on different surfaces without splitting the STL. If you find yourself trying to import three separate STLs and paint them in the scene, stop. You want the by-object filament assignment instead, and the right-click "Change Filament" submenu is your friend there. One more concept that has to land before the brushes make sense: color painting is binary per facet. Each triangle of the mesh belongs to exactly one filament slot at slicing time. There's no blending, no gradient, no anti-aliased edge. If you want a soft transition between two colours, you have to fake it by alternating thin horizontal bands of each color within a height range, and the print result still looks like banding because that's literally what it is. Finding the Color Paint mode in OrcaSlicer 2.3.x Open OrcaSlicer, load an STL or 3MF, and stay on the Prepare tab. On the left-hand vertical toolbar you'll see icons for Move, Rotate, Scale, Place on Face, and a cluster of painting icons further down: Support Painting, Seam Painting, and Color Painting. The Color Painting icon is the one that opens a side panel with the five brushes and the active-filament slot selector. You can also right-click any object in the scene and choose Color Paint from the context menu. Same destination, different route. Clicking the toolbar icon a second time exits the painting mode and bakes your strokes into the mesh's painted-region metadata. The painted regions are saved into the 3MF when you save the project, so if you do all your painting work, save, then come back the next day, your strokes are still there. The Prepare tab is where Color Paint lives. The left-hand toolbar has the painting icons, and the right-click menu lets you add primitives or jump straight into Color Paint mode. Version compatibility is worth a quick note. The five brush set, the Edge Detection toggle on Fill, and the Gap Fill option are all present in OrcaSlicer 2.3.x. Earlier 2.0 builds had the same UI shape but suffered from issue #5152, where painted vertical walls didn't trigger filament changes on the vertical shells when filament shrinkage wasn't set to 100%. That bug was closed by PR #6507, so if you're on a 2.3.x build (and you should be), painted vertical surfaces produce filament changes on vertical walls exactly as you'd expect. The 2.3.2 release also added a wipe-tower-type selector (including a sturdier Type 2 variant for MMU and tool-changer setups), which matters less to color painting itself but does affect how the purge column on your bed looks at print time. You can grab the current release from the OrcaSlicer GitHub releases page; the official orcaslicer.com download links route there anyway. I'd avoid pre-release alpha builds for any actual print job you care about, because painting bugs in alpha builds aren't always backwards-compatible with project files saved in stable builds. The five brush tools, one by one OrcaSlicer's color painting interface exposes exactly five brushes. PrusaSlicer users will recognise the family but the names are different, and you'll see writers online refer to a "Smart Fill" tool that doesn't exist in OrcaSlicer's UI by that name. What OrcaSlicer actually labels them is what I'll use here. Circle The wiki calls Circle "a circle-shape pen tool used to draw any curves on the model's surface." Practically, it's the brush you'll use most. Hold left-click and drag, and Circle paints every visible facet under the cursor with the active filament colour. The brush radius is set with the slider on the side panel, not by mouse wheel by default, which is the first thing I trip over every time I come back to OrcaSlicer after a break. Circle is "visible facets only," meaning it doesn't paint hidden interior facets even if they'd geometrically fall inside the brush radius from your viewing angle. That's a good default for most cosmetic painting work. Sphere Sphere is Circle's evil twin. Same shape, same drag behaviour, but it paints every facet inside the sphere volume around the cursor, including hidden interior facets and the back side of thin walls. This is occasionally what you wan [truncated] --- ## OrcaSlicer Project Files (.3mf): How to Save and Share (2026 Guide) URL: https://orcaslicer.net/orcaslicer-project-files-3mf/ I shared my first OrcaSlicer file with a buddy last spring. He's a Cura guy, runs an old Ender 3 he refuses to retire. I sent him a .3mf because someone on a forum told me .3mf was "the universal one." He opened it, and half my settings were just gone. Tree supports? Vanished. Color paint? Plain orange. Custom seam position? Reset. He printed it anyway, and the print looked nothing like the slice I'd shown him on Discord the night before. That was my first real lesson in what a .3mf file actually is, and what it isn't. The format is great, but the way OrcaSlicer uses it is specific, and the way other slicers read it is even more specific. If you're saving and sharing your work, you need to know which version of .3mf you're handing over, who's on the receiving end, and where the cracks are. This guide walks through saving an OrcaSlicer project, sharing it without trashing the recipient's settings, uploading to MakerWorld or Printables without getting rejected, and recovering when things go sideways. I've kept it accessible because most of the questions I see in the OrcaSlicer Discord are from people on their second or third week with the slicer. If you're past that point, skim the section headers and skip ahead. What this guide covers What actually lives inside a .3mf file OrcaSlicer's .3mf flavor (and the cross-slicer reality) The File menu, from Save Project to Export .3mf vs .gcode.3mf, and which one to send Sharing a .3mf the right way Common problems and how to fix them A backup workflow that doesn't suck MakerWorld and Printables specifics FAQ What actually lives inside a .3mf file Let's start with the basics because the format itself is misunderstood. The 3MF (3D Manufacturing Format) is a ZIP container. Rename a .3mf to .zip, double-click it, and you'll see an XML manifest, a model definition, thumbnails, and (in OrcaSlicer's case) a bunch of preset XML blobs sitting in subfolders. It is not magic. It's a sensibly organized archive. The 3MF Consortium built the format back in 2015. The members list reads like a who's-who of additive manufacturing: Microsoft, Autodesk, HP, Materialise, Shapeways, Ultimaker, and (later) Bambu Lab. The core spec stamps units explicitly inside the model file, which kills STL's chronic millimeter-vs-inch guessing game. It supports extensions for materials, colors, slice data, lattice structures, secure content, and even signed-distance-function geometry. STL, by comparison, is a triangle soup with no units, no colors, no metadata, and no defined origin. If you've ever opened an STL and watched it appear at 1/25th the expected size because someone exported in inches, you know exactly what I'm talking about. OrcaSlicer treats .3mf as its native project format. When you hit Ctrl+S in Orca, the file that lands on your disk contains: Mesh geometry for every model on every plate, with units explicitly stamped. No scaling ambiguity. Plate layout, which means multi-plate arrangement, object positions, rotations, and scales. Multi-plate is a Bambu/Orca extension, so a third-party slicer may only show one plate. Print, filament, and printer profile snapshots. The full preset values at slice time, baked in, not just references. This is the reason opening someone else's .3mf can clobber your active profile (we'll come back to that). Per-object overrides like wall count, infill density, layer height, support flag, brim, and seam position when those are set per-object instead of per-plate. Manual support paint and color paint data, stored as vertex-level paint plus MMU-style triangle splits inside the mesh XML. Modifier parts, which include cutters, infill modifiers, support blockers, and any negative/positive volumes you've added. Custom G-code blocks: start G-code, end G-code, layer change, tool change, pause-at-height, and any post-processing scripts you've wired in. Calibration metadata like flow ratio, pressure advance, and Z-offset for the assigned printer profile. Thumbnail PNGs shown by the printer LCD or by MakerWorld's model card. The sliced G-code itself, but only if you exported a .gcode.3mf rather than a plain .3mf. Project notes, which can hold images, spreadsheets, PDFs, and free text. Orca's Project tab on the top bar is where you find these. The Prepare tab in OrcaSlicer. The asterisk before the project name in the title bar means there are unsaved changes. Compared to the formats sitting next to it on your hard drive: STL: geometry only. No color, no units, no settings, no support paint. The receiver has to start from a blank profile. OBJ: geometry plus a material reference, but slicers ignore the material side. Same practical limits as STL. G-code (.gcode): toolpath only. Locked to one printer, one filament, one slice. The receiver cannot re-slice, cannot scale, cannot edit. .3mf (project): geometry plus settings plus paint plus modifiers. Fully re-sliceable. .gcode.3mf: everything a .3mf has, plus the baked toolpath. Ready to print without re-slicing. So when someone on Reddit says "just send the .3mf," what they really mean is "send the project file." And when someone says "send the .gcode.3mf," they mean "send the file that's already sliced and ready to run on the printer." Two different files, two different jobs, and the difference matters more than people give it credit for. OrcaSlicer's .3mf flavor (and the cross-slicer reality) The 3MF spec is intentionally extensible. That's its strength and its weakness. Bambu Studio created proprietary XML extensions to carry its presets, multi-plate state, color paint, and AMS metadata. OrcaSlicer, which is a fork of Bambu Studio (which itself is a downstream of PrusaSlicer), inherited those extensions and added more of its own. The honest practical consequences look like this. Bambu Studio and OrcaSlicer read each other's .3mf files in most cases. Each may warn about unknown keys from the other, and color paint occasionally needs reapplication. There's even a third-party converter tool that exists specifically to translate paint between them, which tells you everything about how clean the round trip really is. PrusaSlicer opens Orca .3mf geometry fine, including multi-plate. The print settings rarely import cleanly because the preset key names diverge. Treat PrusaSlicer-to-Orca and Orca-to-PrusaSlicer as "geometry transfer only" and you'll save yourself a lot of swearing. Cura doesn't recognize Bambu or Orca custom extensions. Cura users report .3mf files from Orca behave inconsistently: sometimes you get geometry without color, sometimes only one plate loads, sometimes the import fails outright. If you're sending a file to a Cura user, export the geometry as STL or as a Generic 3MF and stop pretending it'll all just work. Version mismatch inside OrcaSlicer itself is also a thing. A .3mf saved by a newer Orca version warns "this 3MF was generated by a newer version" and may load with geometry-only fallback. The fix is to update Orca, not to downgrade the file. Issue #11436 on the OrcaSlicer GitHub is one of the better threads documenting this. So the honest framing, the one I wish someone had given me on day one: .3mf is a great geometry interchange across all major slicers. Orca's full project fidelity only round-trips cleanly back into Orca, and mostly into Bambu Studio. Everything else is a partial import. The File menu, from Save Project to Export OrcaSlicer's File menu is the one place every option lives, and once you've memorized the layout it stops being scary. Verified against the OrcaSlicer GitHub wiki and OrcaSlicer 2.3.x, here's what each entry does. A loaded model in OrcaSlicer's Prepare tab. The title bar shows the current project name, and the green dot at top-right means the file is ready to slice. Save Project (Ctrl+S) This saves your current project as a .3mf, overwriting the existing file if the project was previously saved or opened from disk. If the project has never been saved before, OrcaSlicer will (usually) prompt you for a name and locati [truncated] --- ## OrcaSlicer Brim, Raft & Skirt: When and Why (2026 Settings Guide) URL: https://orcaslicer.net/orcaslicer-brim-raft-skirt/ I printed a 200mm tall vase in PETG without a brim and watched the corner curl off the bed at hour two. Three reslices later I had it stuck down with a 5mm brim and a glue stick, same printer, same filament, same model, just one setting changed. Brim, raft, and skirt are three different tools for three different problems, and most people use the wrong one. Here's the decision tree I wish someone had handed me before I wasted that spool. What follows is the practical, opinionated guide I'd give a friend who's tired of failed first layers. I'll cover what each option actually does inside OrcaSlicer 2.3.2, when to reach for it, and the exact numbers that work in the real world. I'll also flag the new brim_use_efc_outline toggle that landed in v2.3.2 and the underused brim ears feature that's quietly the best fix for corner warping on otherwise-flat parts. If you want the broader picture of how these settings sit alongside everything else, the OrcaSlicer settings master guide covers the full panel. Table of contents Quick decision tree, which one do you actually need Skirt, the free 30 seconds you should always spend Brim, the workhorse for warping and small footprints Brim types, Outer / Inner / Both / Auto / Painted Brim width tuning, the 3 / 5 / 8 / 10mm rules Brim object gap and flow ratio, the two settings that decide removal pain Brim ears (mouse ears), when to paint instead of widen New in v2.3.2, brim_use_efc_outline Raft, the last resort that you probably don't need Raft anatomy, layers, contact distance, expansion, density Bed surface matters more than your slicer settings Material-specific recommendations Common problems and fixes Honest verdict, what 95% of users should do FAQ Quick decision tree, which one do you actually need? Before we touch a single setting, here's the flowchart that decides everything else. I run this in my head every time I drop a model on the plate. If you only remember one thing from this article, it's this. SituationRecommended setup First layer is reliable on this bed and filament today, flat-bottomed modelSkirt only, 1 to 3 loops. Done. Tiny footprint, tall, narrow, spindlyBrim 5 to 8mm. Or paint mouse ears at the corners. Mild corner warping (PETG, ABS, ASA on a tuned bed)Brim 5mm, brim object gap 0.1mm Severe warping, open frame printerBrim 8 to 10mm plus draft shield, or move into an enclosure Bed is severely uneven and you can't fix it nowRaft 3 layers, contact distance 0.15mm Material is Nylon, PC, PC-CF, or polypropyleneGarolite plate or PEI with glue plus brim 8mm minimum. Raft only if brim still fails. Brim already failed and you're out of ideasRaft 3 to 4 layers, expansion 3 to 5mm The default recommendation, for a flat-bottomed slab on a tuned printer with PEI Smooth or PEI Textured, is skirt only. No brim, no raft. Most prints don't need anything more than this and adding a brim "just in case" creates work after the print and a small scar on the part that you didn't have to live with. Skirt, the free 30 seconds you should always spend A skirt is a perimeter loop printed around the model on the first layer that doesn't touch the part. It does two jobs. It primes the nozzle so the first millimeter of actual model has consistent flow, and it gives you a visual sanity check on first layer height, squish, and bed leveling before the model starts. A skirt does nothing for adhesion. It's basically free, costs maybe 30 seconds of print time, and there's almost no scenario where running a 1 to 3 loop skirt is a bad idea. You'll find the controls in Process panel → Others tab → Skirt section. Switch to Advanced or Expert mode via the gear icon at the top of the Process panel, otherwise some of these are hidden in Simple mode. OrcaSlicer skirt settings, line by line SettingVariableDefaultWhat it does Loopsskirt_loops1Number of perimeter loops. Set to 0 to disable. Two to three is what I run. Distanceskirt_distance2mmDistance from skirt to brim or object. Heightskirt_height1 layerLayers tall. Raise this to turn it into a draft shield. Min extrusion lengthmin_skirt_length0mmForces a minimum extrusion length so the nozzle is fully primed even on tiny parts. Typeskirt_typeCombinedCombined wraps all objects. Per object is recommended with object sequence printing. Draft shielddraft_shieldDisabledDisabled, Enabled (use skirt height), or Limited (follow tallest object). Single loop draft shieldsingle_loop_draft_shieldDisabledAfter layer 1, use only one loop. Saves filament, weaker shield. Skirt speedskirt_speed0 (use first-layer speed)Slower is more reliable. Skirt start angleskirt_start_angle-1 (auto)Cosmetic only. I run 2 loops at 2mm distance, 1 layer high, on essentially every print regardless of material. It's the cheapest insurance in 3D printing. If I'm dropping a fresh spool I haven't used before, I'll bump it to 3 loops just to watch the extrusion settle in. When to turn the skirt into a draft shield Set draft_shield to Enabled and crank skirt_height up to roughly the model's height (or use Limited mode and let the slicer match the tallest object) and you've got a wall around the print that blocks cold air. This matters if you're printing ABS or ASA on an open-frame machine. Heat retention next to the part keeps each new layer hotter, which keeps the layers below from contracting and pulling the corners up. It's not a substitute for an enclosure, but it helps. Bear in mind that a tall draft shield uses real filament, and on a long ABS print it can add 10 to 30g. One annoying caveat: per Issue 12050, skirt and brim don't always cooperate in per-object mode. If you're running object sequence printing and the skirt vanishes, switch skirt_type to Per object and re-slice. Brim, the workhorse for warping and small footprints A brim is extra material attached to the model's first layer, fused to the outer (or inner) perimeter and extending outward by a chosen width. It glues the edges of the model to the bed by giving the first layer way more contact area. You peel it off by hand or with a fingernail or a deburring tool after the print. Brims fix roughly 90% of warping and edge-lift problems on materials like ABS, ASA, PETG, and PLA on textured surfaces. The trade-off is removal effort and a small scar at the edge of the part where the brim met the perimeter. Honest take: if you're not sure whether you need a brim, you probably don't. Brims are for known-warpy materials, tall skinny parts, parts with tiny first-layer footprints, and recurring bed adhesion failures. Slap one on a flat-bottomed PLA cube and you've just made yourself extra cleanup work for no benefit. The settings live in Process panel → Others tab → Brim section, right below the skirt block. Below is a sliced view of a tower with a generous brim. You can see the green Brim line type lighting up around the base, which is how I sanity-check that the brim actually attached itself to the perimeter before I send the file to the printer. Sliced preview of a tall part with a brim, the green Brim entries in the legend confirm the brim is attached to the outer perimeter. OrcaSlicer brim settings, line by line SettingVariableDefaultWhat it does Typebrim_typeAutoNone, Outer brim only, Inner brim only, Outer and inner brim, Auto, or Painted (Mouse Ears). Widthbrim_width5mm (manual)If Type is Auto, this is computed for you. Brim object gapbrim_object_gap0.1mmRange 0 to 2mm. Larger means easier removal but less adhesion. Brim flow ratiobrim_flow_ratio1.0Above 1.0 sticks better but is harder to remove. Brim use EFC outlinebrim_use_efc_outlineOffNew in v2.3.2. Aligns the brim with the post-Elephant-Foot-Compensation perimeter so the brim actually touches the model. Combine brimscombine_brimsOnMerges touching brims from neighboring objects into continuous loops for a cleaner first layer. Brim types, Outer / Inner / Both / Auto / Painted The Type dropdown is where most people pick wrong. Here's how I think about each option: TypeBest forTrade-off None [truncated] --- ## OrcaSlicer Fuzzy Skin: Add Texture to Any Print (2026 Settings Guide) URL: https://orcaslicer.net/orcaslicer-fuzzy-skin/ I printed a phone case in glossy black PETG and within a week it looked like fingerprint art. Reprinted with fuzzy skin enabled at 0.3mm thickness, and six months later it still looks like a brand-new case. Fuzzy skin is the cosmetic setting nobody talks about until they need it, and OrcaSlicer's v2.3.2 added two new noise modes worth knowing about (Ripple and a new Hole-aware Contour mode) that change how I think about textured prints. This guide covers what fuzzy skin actually does, what changed in v2.3.2, the two settings that matter most, and the honest tradeoffs nobody mentions until your snap-fit refuses to snap. For the broader settings context, the full OrcaSlicer settings master guide is the parent reference. Table of contents What fuzzy skin actually does (and what it doesn't) Where to find fuzzy skin in OrcaSlicer The five fuzzy skin modes (with the Contour rename explained) Thickness and point distance: the two settings that matter Noise types: Classic, Perlin, Voronoi, Ripple, and friends Generator modes: Displacement, Extrusion, Combined Per-region application: paint tool and modifiers Reading the sliced preview When to use fuzzy skin When to skip fuzzy skin Material reality check Performance and gcode size Common mistakes and fixes FAQ What fuzzy skin actually does (and what it doesn't) Fuzzy skin takes the outer wall extrusion path, resamples it at a regular step you control, and offsets every new sample point inward or outward by a random amount you also control. The printhead still moves one layer at a time. The Z axis isn't touched. What you see when you look at a finished print is per-layer randomness stacked on top of itself, and your eye reads that as a vertical pebble or sandpaper texture. It's a clever optical illusion, not a real 3D fuzz. I want to put one myth to bed up front. Fuzzy skin is purely cosmetic and tactile. It doesn't make your part stronger in any meaningful way. CNC Kitchen has noted that very thin walls can pick up a tiny bit of stiffness the same way corrugated cardboard is stiffer than flat cardboard, but that's a niche effect on single-line walls. For 99% of prints, fuzzy skin is decoration and grip, nothing else. What fuzzy skin reliably hides: Z-seams, layer lines on vertical walls, light scratches and scuffs, fingerprints on dark glossy filaments, and mild under-extrusion blemishes. What it cannot hide no matter how hard you crank thickness: stringing, blobs, ringing, layer shifts, or bad first-layer adhesion artifacts. If you're hunting Z-seam visibility specifically, fuzzy skin is one tool, but it's not the only one. The OrcaSlicer seam placement guide covers the geometric approach, which is usually the right first move before reaching for fuzz. Where to find fuzzy skin in OrcaSlicer Open the Process panel on the right side of the OrcaSlicer window. In v2.3.2 the fuzzy skin section lives under the Others tab toward the bottom of the Quality category, depending on which sub-tab layout your profile uses. If you can't find it, type "fuzzy" into the search box at the top of the Process panel and it'll filter straight to the right block. The full Quality tab in OrcaSlicer v2.3.2. Fuzzy skin sits in the Others section near the bottom, with mode, thickness, point distance, and the first-layer toggle all in one block. Two important navigation notes for anyone coming from older guides or from PrusaSlicer: "Outside walls" got renamed to "Contour" in v2.3.2. If a YouTube tutorial from 2023 tells you to pick "Outside walls," that's now Contour in your dropdown. Same behaviour, new label. The Hole-aware option is new. v2.3.2 added "Contour and Hole" as a separate dropdown choice, so you can fuzz the outer perimeter and the inside of any cutout holes without fuzzing every internal wall. Older builds forced you to pick All Walls if you wanted holes textured, which was overkill. You can also access fuzzy skin per-object by right-clicking a model on the build plate, then Add Modifier with Fuzzy Skin overrides. I'll cover that workflow in the per-region section further down. The five fuzzy skin modes OrcaSlicer v2.3.2 ships with five mode choices in the Fuzzy Skin Type dropdown. Each one targets a different set of perimeters. ModeWhat it texturesBest for NoneNothing. Fuzzy skin off.Default. Use when you want clean walls. ContourOuter perimeter only (the "outside" of the model)90% of cosmetic use cases. Fast, clean, no internal mess. Contour and HoleOuter perimeter plus inside-of-hole perimetersVases, planters, lampshades where the inside of cutouts is visible. All WallsEvery perimeter, internal and externalDecorative shells where the perimeter count is low and you want texture everywhere. Adds the most print time. Painted OnlyOnly regions you mark with the paint toolMixed prints where one region needs grip and the rest stays smooth. I default to Contour for almost everything. It's the cheapest in print time and the safest for dimensional accuracy. Contour and Hole gets used on planters and any model where I can see inside through a cutout. All Walls is rare for me because internal walls don't show on a finished print, so why pay the time cost. Thickness and point distance: the two settings that matter If you only learn two parameters in this article, learn these. Everything else is fine-tuning. Fuzzy skin thickness (default 0.3mm) Thickness is how far each sample point can wander inward or outward from the original wall path. A higher number means deeper bumps and a coarser texture. The shipped default is 0.3mm and that's a sane starting point for a 0.4mm nozzle running 0.2mm layers in PLA. Here's the rule that catches most beginners: thickness needs to stay roughly between 70% and 125% of your nozzle diameter. For a standard 0.4mm nozzle that's a safe envelope of about 0.28mm to 0.5mm. Push it higher and the extrusion path starts skipping segments, dropping blobs, or punching holes through thin walls. Push it lower than about 0.1mm and the texture becomes invisible to the eye. If you've got a 0.6mm nozzle, scale thickness up to roughly 0.4–0.5mm. On a 0.2mm nozzle drop it to 0.1–0.15mm. The relationship to nozzle diameter is what keeps the extruder happy, not some arbitrary number. Fuzzy skin point distance (default 0.8mm) Point distance is how often the slicer takes a new random sample along the perimeter. Smaller numbers mean more samples per millimetre, which means a denser, finer-grained texture. Larger numbers mean fewer samples, which gives chunkier pebbled bumps. Default is 0.8mm and that produces what most people picture when they hear "fuzzy skin." Drop to 0.3–0.4mm for a fine sandblast feel. Push to 1.2–1.5mm for a chunky pebbled look that almost reads as basket weave on tall vases. One warning: don't go below about 0.2mm. The gcode size balloons (your perimeter goes from a handful of long line segments to thousands of short ones), slicing slows down, and your physical nozzle can't actually resolve detail finer than that anyway. Recipe table for the impatient LookThicknessPoint distanceMode Subtle (hide layer lines on a display piece)0.10–0.15mm0.5–0.8mmContour Standard fuzzy aesthetic (default)0.30mm0.8mmContour Aggressive grip (tool handles, controller shells)0.50–0.60mm0.4–0.6mmContour or All Walls Decorative wood-grain (with woodfill)0.40mm0.6mmContour, Perlin noise, feature size 1–2mm Vase / planter with visible interior0.30mm0.8mmContour and Hole The first-layer toggle (off by default for a reason) There's a third boolean called "Fuzzy skin first layer" and it ships off. Leave it off unless you know what you're doing. The first layer is what holds your print to the bed, and replacing a clean perimeter with a wobbly one cuts the contact area at the worst possible spot. Issue 6414 on the OrcaSlicer tracker shows exactly this confusion: users in Contour mode noticed the bottom layer wasn't textured, assumed bug, the maintainers explained it was the first-layer toggle behaving as intended. If you actually [truncated] --- ## How to Import & Export OrcaSlicer Profiles: 2026 Backup Guide URL: https://orcaslicer.net/orcaslicer-import-export-profiles/ I migrated OrcaSlicer to a new MacBook last month and almost lost six months of filament profile tuning. The profiles weren't where I expected them, the bundle import threw an error on first try, and the migration guide I found online still referenced the pre-2.0 .orca_preset extension that current builds don't even use anymore. Here's what actually works in OrcaSlicer 2.3.2, written by someone who's done this both the easy way and the hard way. This guide covers three workflows: the official preset bundle export and import (the cleanest path for most people), single preset sharing for when a friend just wants your tuned PETG profile, and the filesystem copy fallback for when the UI bundle workflow misbehaves. I'll also cover the inheritance trap that makes "imported" profiles silently vanish, the cross-slicer migration story (spoiler: PrusaSlicer .ini import is not natively supported), and a backup strategy that has actually saved my bacon during a Windows reinstall. If you're brand new and haven't gotten through a first print yet, I'd start with my OrcaSlicer beginners first print guide before worrying about backups, but if you've already invested time tuning settings, this is the article to read before the next slicer update. Table of contents What a "profile" actually is in OrcaSlicer Where OrcaSlicer stores profiles on your computer Workflow A: Export your profiles as a preset bundle Workflow B: Export a single profile to share with one person Workflow C: Manual filesystem backup (the fallback) Importing on a new computer Importing a community profile (and the inheritance trap) Migrating from Bambu Studio, PrusaSlicer, and Cura Why profiles disappear after updates, and how to recover them A backup strategy that actually works FAQ What a "profile" actually is in OrcaSlicer Before you can back something up, it helps to know what you're backing up. I lost two hours on my first migration because I assumed "my profile" was one file. It isn't. OrcaSlicer splits settings into three preset types, and your real "configuration" is the combination of all three. Preset typeWhat it storesHow often you touch it Printer (machine)Bed shape, build volume, hot end, max acceleration, retraction defaults, start and end G-code, change filament G-codeSet once per printer, tweak rarely FilamentNozzle temp, bed temp, fan speed, cooling overrides, pressure advance, max volumetric speed, flow ratioTuned per spool or per brand ProcessLayer height, line widths, wall count, infill density and pattern, support style, speed group, seam settingsSwitched per object There's a fourth conceptual layer the wiki calls "printer model" (for example "Bambu Lab X1 Carbon" before you pick the 0.4 nozzle variant), but end users rarely see this directly because the setup wizard picks a variant for you and creates the linked machine preset automatically. If you want a deeper dive into how the printer side works, the add a custom printer guide covers that piece in detail. The thing that trips people up: every user filament preset usually inherits from a system parent like "Generic PLA" or "Bambu PLA Basic." Only the diffs are saved in your user JSON. If the parent disappears (because of a slicer update or a missing printer on the destination machine), the child appears to vanish from the dropdown even though the JSON file is still sitting on disk. I'll come back to this in the inheritance trap section, because it's the single biggest reason "I imported my profiles but they don't show up" threads exist on GitHub. Where OrcaSlicer stores profiles on your computer I'll save you the AppData scavenger hunt I went on. Here are the paths verified against 2.3.2 builds on each platform. OS / install typeUser preset path Windows%APPDATA%\OrcaSlicer\user\default\ macOS~/Library/Application Support/OrcaSlicer/user/default/ Linux (native or AppImage)~/.config/OrcaSlicer/user/default/ Linux (Flatpak, current ID)~/.var/app/com.orcaslicer.OrcaSlicer/config/OrcaSlicer/user/default/ A few quick notes from doing this on every OS at least once. On Windows, AppData is hidden by default, so paste the path into File Explorer's address bar instead of trying to browse to it. %APPDATA% resolves to C:\Users\YourName\AppData\Roaming\ if you want to know where you actually are. On macOS, the Library folder is also hidden, so hold Option, click the Go menu in Finder, and Library appears in the list. On Linux, nothing's hidden but the dot prefix on .config means you need to enable "Show hidden files" in your file manager. The Flatpak path is the one that has shifted recently. Older Flatpak builds used io.github.softfever.OrcaSlicer, and the application ID changed to com.orcaslicer.OrcaSlicer in 2.3.2. If your folder isn't where I said it should be, check both IDs. Inside the user/default/ folder you'll find three subfolders that matter and a handful of loose files: user/default/ machine/ <-- your printer JSONs filament/ <-- your filament JSONs process/ <-- your process JSONs OrcaSlicer.conf (general preferences, not part of the bundle) OrcaSlicer.log cache/ Note that the folder is called machine even though the UI calls it "Printer." That's a holdover from the Bambu Studio fork and it has confused me more than once when I was poking around at 1am. The three preset subfolders are the ones that actually matter for a backup. The OrcaSlicer.conf file holds preferences like window layout and your default startup behaviour, and it is not included in any bundle export. I'll come back to that. There's also a portable mode worth knowing about. If a folder literally named data_dir sits next to orca-slicer.exe, OrcaSlicer uses that as the config root instead of AppData. You can also force a custom location with --datadir "C:\path\to\folder" on a shortcut. I keep my workshop PC and laptop syncing a portable folder through OneDrive so they always share the same profiles. More on that in the strategy section. Workflow A: Export your profiles as a preset bundle This is the workflow you'll use 80% of the time and the one I recommend for almost everyone. Bundles preserve the inheritance links between presets, the printer associations, and the JSON structure that OrcaSlicer expects on import. The filesystem copy works too, but the bundle is what the slicer is designed around. To start, open OrcaSlicer and go to File → Export → Export Preset Bundle... (note the ellipsis, that's the exact menu wording in 2.3.2). A dialog opens with five radio button options. Here's what each one actually does in plain English, because the dialog itself is terse: Note: the screenshot shows the German UI ("Preset-Bundle exportieren"). The English equivalent is "Export Preset Bundle." The five radio options translate as Printer config bundle (.orca_printer), Filament bundle (.orca_filament), Printer presets (.zip), Filament presets (.zip), and Process presets (.zip). Printer config bundle (.orca_printer): the entire setup for one or more printers, including any custom filaments and process presets linked to that machine. This is the "give me everything for this printer" option. If you tuned an A1 with three filaments and two process presets, exporting an .orca_printer packages all of it. Filament bundle (.orca_filament): custom filaments only, but with their printer connections preserved. Pick this when you've tuned one spool across multiple printers and want to share that filament without dragging the whole machine config along. Printer presets (.zip): raw JSON files for custom printer configs, zipped. No bundle wrapper, just the files. Useful if you want to inspect or edit the JSON before someone else imports it. Filament presets (.zip): raw JSON filament configs, zipped. Same idea as above. Process presets (.zip): raw JSON process configs, zipped. For 90% of backup use cases, you want option 1 (.orca_printer) per printer you've tuned. It packages everything related to that machine in one file. The plain .zip options are for when you spec [truncated] --- ## How to Add a Custom Printer to OrcaSlicer (2026 Step-by-Step) URL: https://orcaslicer.net/orcaslicer-add-custom-printer/ I built a custom Voron Trident with a 350x350 bed, fired up OrcaSlicer expecting to see it in the dropdown, and instead got "Generic Voron 2.4 (300mm)" as the closest match. The Custom Printer flow exists for a reason, and it works fine, but the documentation glosses over the gotchas that crash your first print into the bed. Here's the actual step-by-step. I've done this dance maybe twelve times across DIY CoreXY builds, off-brand Marlin cartesians, a Klipper-converted Ender 5, and a couple of Sovols that the wizard refused to recognize. So this isn't a regurgitated wiki page. I'm going to walk you through the three real paths to adding a printer that isn't in OrcaSlicer's stock list, tell you which one to actually pick (hint: it's almost never Path A), and give you the exact Start G-code templates for both Marlin and Klipper so you don't end up copy-pasting from a 2022 forum post that uses placeholder syntax that no longer parses. By the end you'll have a working printer profile, you'll know where OrcaSlicer stashes the JSON file so you can back it up before the next update, and you'll have a checklist for the five things that always break on the first print after a manual add. Table of contents The three paths at a glance Path C: Import a community profile (try this first) Path B: Manually add a printer in Printer Settings Clone, don't build from scratch Basic Information tab walkthrough Bed shape definitions Extruder tab settings Machine G-code: Marlin and Klipper templates Network connection (Device tab) Path A: The wizard's Create Printer option Save, find, and share your custom printer Common pitfalls (and how I've fixed them) When to stop fighting the slicer FAQ The three paths at a glance OrcaSlicer gives you three routes to get an unsupported printer running. I've ranked them by what actually works fastest in 2026, not by what the wiki suggests first. PathWhat it isPick this if...Time to first print Path C: Import community profile Download a .orca_printer bundle, File > Import > Import Configs Someone else has already done the work (vendor GitHub, Printables, MakerWorld) 5 minutes Path B: Manual via Printer Settings Clone the closest existing printer, edit bed/origin/G-code in Printer Settings Your printer is uncommon, DIY, or a variant of a listed model 20 to 40 minutes Path A: Wizard Create Printer File > Configuration Wizard, then the "Create Printer" button You want a quick stub and you'll edit it in Printer Settings anyway 10 minutes (but you'll redo it in Path B) My honest advice: check Path C first because if it exists you're done in five minutes. Fall back to Path B with a clone of a similar listed printer (the "Save As" trick I'll cover below). Treat Path A as a wrapper around Path B that saves you maybe two clicks. If you skipped the initial wizard entirely and want to revisit it, see my OrcaSlicer setup wizard walkthrough. If you're not even sure where the Printer Settings dialog lives in the interface, the OrcaSlicer UI tour shows you every panel. Path C first: Import a community profile (the fastest path if it exists) I'll start with the path that gets most people unstuck in five minutes, because if a community profile exists for your printer, everything else in this article becomes optional reading. Where to find profiles, ranked by trust Not all profiles are equal. I've burned hours debugging "why is my retraction tower a blob" only to discover the GitHub fork I grabbed was tuned for a different hot end. Use this trust order: Vendor official GitHub release. Sovol-OrcaSlicer, ELEGOO Wiki downloads, the Bambu repo, Anycubic and Creality vendor pages. These are tested against real production printers. Community profiles with prints shown. Repos like mjonuschat/OrcaSlicer-Profiles and DarthPope0327/orca-slicer-profiles (both excellent for Voron and Klipper builds). I trust these when the maintainer posts actual benchy and calibration cube photos. Printables.com OrcaSlicer profile tag. Search "OrcaSlicer profile [your printer]." Quality varies. Read comments before downloading. MakerWorld profile downloads. Skewed toward Bambu but increasingly hosts Voron and Prusa community profiles. Random GitHub forks. Last resort. Always verify Start G-code before printing. The import steps Download the .orca_printer file. If it arrives as a .zip, extract it first. The importer wants the .orca_printer directly, or a folder of .json files. In OrcaSlicer: File > Import > Import Configs. Browse to the file, select it, click Open. OrcaSlicer will prompt if it would overwrite an existing user preset. Confirm or rename. Confirm the printer now appears in the Printer dropdown on the right sidebar. Select it. Open Printer Settings (the pencil icon next to the dropdown) and verify Bed shape, Origin, and Start G-code before hitting Slice on a real print. That last step is the one nobody does and the one that has saved me from at least three failed first layers. Imported profiles can have origin offsets baked in that don't match your physical printer. Verify before you trust. For a deeper dive into bundle formats, .orca_filament vs .orca_printer, and round-tripping profiles between machines, see my OrcaSlicer profile import and export guide. There's also a curated list of vetted community profiles at the OrcaSlicer printer profiles hub. The first-print verification checklist Whatever profile you imported, run this five-minute check before slicing anything real: Slice a 20x20mm calibration cube. Look at the gcode preview. Does the toolhead start in the corner you expect? If it crashes into Y-min on a center-origin printer, the origin is wrong. Check the Start G-code in Printer Settings > Machine G-code. Does it call G28 (Marlin) or PRINT_START (Klipper)? If it's a Klipper profile dressed up as Marlin, fix it. Check Max nozzle temperature. Some community profiles cap at 250C even when your hot end is all-metal hardened. Bump it up. Run the cube. Watch the first layer. Don't walk away. Path B: Manually add a printer in Printer Settings (the canonical method) This is where you'll spend most of your time if no community profile exists. It's also where you have the most control, which is good news and bad news. The bad news is that you can absolutely set a value that crashes the toolhead into a Z-endstop. The good news is that the dialog is well organized and most fields have sane defaults. Clone, don't build from scratch I cannot stress this enough. The Printer Settings dialog has a "Save As" button. Use it. Here's the workflow that takes you from zero to a working profile in 20 minutes instead of 2 hours: Open the Printer dropdown in the right sidebar. Pick the closest existing printer. Match on bed size, kinematics (Cartesian vs CoreXY vs delta), and firmware first. Brand doesn't matter much. Click the small pencil/notepad icon to the right of the dropdown. Printer Settings opens. Click Save As at the bottom of the dialog. Name it something descriptive, like "MyVoron0.1 0.4 nozzle" or "DIY CoreXY 300x300 Klipper." Now edit only the differences: bed size, origin, Start G-code, nozzle diameter, max temp. That's usually 5 fields instead of 50. Here are my go-to clone bases for common DIY situations: Your buildClone fromEdit afterward Custom Voron 0.1 buildGeneric Voron 0.2Bed size, max Z, possibly origin DIY CoreXY 300x300 KlipperGeneric Voron 2.4 300Origin if not center, Start G-code Off-brand Marlin cartesian 235x235Generic Ender 3 V2Bed size, max temp, Start G-code Klipper conversion of an Ender 5Generic Klipper printerBed size, PRINT_START params Off-brand deltaGeneric FLSUN V400 or Anycubic KosselDiameter, max Z IDEX DIY buildSnapmaker J1Extruder offsets, bed size Cloning is faster than from-scratch by an order of magnitude because every Marlin / Klipper / RepRap default value is already correct. Retraction settings, jerk, acceleration, max feedrate, all baked in. Don't be a hero. Hit Save As. Basic Information tab walkthrough The Prin [truncated] --- ## OrcaSlicer UI Tour: Prepare, Preview & Device Tabs Explained (2026) URL: https://orcaslicer.net/orcaslicer-ui-tour/ I finished the OrcaSlicer setup wizard, landed on the Prepare tab, and stared at it for a solid minute. There are four tabs across the top, a toolbar with ten buttons, paint tools hidden behind a left-side icon strip, and a right-side panel with three dropdowns that each have their own sub-tabs. Here's what every section actually does, in the order I wish someone had explained it to me. This tour is built against OrcaSlicer v2.3.2 on Windows 11. The layout is mostly identical on macOS and Linux, with the usual Cmd-key swaps. If you haven't finished the first-launch wizard yet, do that first in our OrcaSlicer setup wizard walkthrough and then come back here. If you want the end-to-end "press print and don't melt your bed" version, that's the beginner's first print pillar. This article is the in-between piece: I'm naming every panel, button, and dropdown so the rest of the cluster stops feeling like a maze. Heads-up before we start: the UI is dense. Some buttons are unintuitive. A lot of features live behind right-click menus that aren't obvious. The Device tab looks broken until you actually connect a printer. I'll flag those potholes as we hit them. Table of contents The four-tab anatomy (Prepare, Preview, Device, Project) The top toolbar and the Slice button Inside the Prepare tab The right-side settings panel (the dense one) Paint tools: Color, Support, Seam The Preview tab The Device tab The Project tab (the one nobody mentions) Keyboard shortcuts cheat sheet Where features hide (right-click and other easter eggs) FAQ The four-tab anatomy (Prepare, Preview, Device, Project) The first thing to clear up: OrcaSlicer has four top-of-window tabs, not three. A lot of older blog posts (and an earlier draft of my own notes) said three. They're wrong about v2.3.2. The actual order, left to right, is Prepare, Preview, Device, Project. TabWhat it's forWhen you use it PrepareImport, arrange, scale, rotate, paint, configure print settingsEvery single print, this is the default landing tab PreviewScrub through the sliced toolpath layer by layer, inject color changes or pausesOnly after you press Slice plate (Ctrl+R) DeviceLive printer control: status, camera, AMS, queue, temperaturesOnly useful with a connected networked printer ProjectProject notes, attached docs, bill of materialsHonestly, almost never. More on that later. Here's the first real pain point: there's no keyboard shortcut to switch between these tabs. None. You have to click. Issue #527 requested a tab hotkey and it was closed as "not planned." I check that issue every few months hoping it reopens. It hasn't. If you're coming from Cura or PrusaSlicer where the workflow is more linear, you'll click between Prepare and Preview about a hundred times per project and you'll notice the missing hotkey by hour three. The other thing I want to be honest about: Prepare and Preview do most of the daily work. Device only matters if you have a Bambu, a Klipper rig, or an OctoPrint setup connected over the network. Project is essentially metadata storage. I'll spend most of this guide on the first two. The top toolbar and the Slice button Across the top of the 3D viewport (still on the Prepare tab) is a horizontal toolbar. These are the buttons you'll touch the most. I'll go left to right. ButtonWhat it doesNotes AddOpen file picker for STL, STEP, 3MF, OBJ, AMFAlso accepts drag-and-drop directly onto the viewport Add PlateCreates a new build plate inside the same projectBuggy. See note below. Auto-ArrangePacks all models on the current plateHotkey: A Auto-OrientDrops each model onto its most stable faceSometimes flips logos upside-down, eyeball it after SplitSplits a multi-body STL into separate objectsUseful when an STL packs multiple parts in one file Split PartsSplits at part boundaries inside one objectDifferent from Split, this one stays inside a single mesh Variable Layer HeightOpens the per-layer height editorAdaptive layer heights live here Plate selectorThumbnails of every plate in the projectClick a thumbnail to switch plates Undo / RedoStep backwards or forwards through editsCtrl+Z, Ctrl+Y Slice plateSlices the current plate onlyCtrl+R. The dropdown arrow next to it gives Slice all. Export / SendExport sliced file or send to printerAppears after a successful slice. Split-button: main click exports, arrow opens "Print" or "Send and Print." The Add Plate button has a real bug I want to flag because it tricked me for a week. After you click Add Plate, the button looks unresponsive. It won't accept a second click. Issue #5042 and #5899 confirm it: move your mouse off the button and back onto it, and it'll work again. It's not a real lockout, just a UX bug. Multi-plate workflows feel rough because of this. If you regularly print batches across multiple plates, you'll learn to expect the weird mouse dance. The Slice plate button is the one you came for. Ctrl+R does the same thing. After it finishes, the Preview tab unlocks and the export button appears in the same toolbar slot where Slice was. If you click the small arrow next to Slice plate, you get "Slice all" which processes every plate in the project. Useful when you have batched five plates for an overnight run. Inside the Prepare tab Prepare is where you'll spend 80% of your time. The tab is divided into four zones: the top toolbar (covered above), the left-side gizmo strip, the central 3D viewport, and the right-side settings panel (which I'll handle in its own section). The left-side gizmo strip This vertical strip of icons only lights up after you select a model. Click your imported STL once on the bed, and the icons become active. Here's the full lineup with hotkeys. ToolHotkeyWhat it does MoveMDrag the model on the bed, or type exact X/Y/Z coordinates ScaleSResize uniform or per-axis, in percent or millimeters RotateRRotate around any axis. Hold Shift while dragging for 45-degree snaps Lay on FaceFClick a face, the model rotates so that face sits flat on the bed CutBPlane cut with optional connectors. Some older docs list C, the current wiki says B. Confirm in your install if it matters. Mesh Boolean(none)Union, difference, or intersect on two or more selected meshes Color PaintingNPaint faces with assigned filament colors Support PaintingLEnforce or block supports per face Seam PaintingPForce or block seam placement per face TextTAdd embossed or debossed text on a face SVG(none)Add embossed or debossed SVG art on a face Lay on Face is the one I undersold to myself for a month. If a model is sitting awkwardly, you don't have to math out rotation angles. Press F, click the face you want flat on the bed, and OrcaSlicer rotates the model into place. It's the single biggest time-saver in the gizmo strip. The paint tools (N, L, P) get their own section below because they're deep enough to deserve it. The thing to know up front: those keys only fire when a model is selected and the viewport has focus. If you press N and nothing happens, click your model first. The 3D viewport and camera The viewport is the central area showing your build plate and any loaded models. The build plate grid has an axis indicator in the lower-left corner: red is X, green is Y, blue is Z. Standard CAD coloring. Camera controls aren't explained anywhere obvious, so here's the list. Middle-mouse drag orbits around the focal point. Shift plus middle-drag pans. Scroll wheel zooms. The number keys 0 through 6 jump to preset views: 0 is default isometric, 1 through 6 are top, bottom, front, back, left, right (the exact mapping is on the keyboard-shortcuts wiki page if you want to memorize it). In the bottom-left corner of the viewport there's a small info HUD showing the selected model's dimensions in millimeters and its triangle count. If you ever wonder why a slice is taking forever, glance at that triangle count. Anything over a couple million triangles is going to chug. Right-click is where features hide This is the single biggest thing I missed when I started. Right-click [truncated] --- ## OrcaSlicer Ironing Settings: Glass-Smooth Top Surfaces (2026) URL: https://orcaslicer.net/orcaslicer-ironing-settings/ The first time I tried OrcaSlicer's ironing setting on a 100mm flat sign I printed for my desk, it added 45 minutes to a 90-minute print and the surface came out worse than no ironing at all. Brown discoloration, ridges, the whole disaster. Three months and a stack of test cubes later I've got ironing dialed in for PLA and PETG, and the trick is that almost every default in OrcaSlicer is wrong for it, or at least wrong for the kind of plaque-and-sign work most of us actually want this feature for. I'm going to walk through every ironing parameter OrcaSlicer exposes, what each one does in plain language, what numbers to start with for the common materials, and the failure modes I kept running into so you don't have to repeat them. This is the long version because ironing has more knobs than people think, and getting one of them wrong wastes a print. If you're new to OrcaSlicer's process settings in general, the OrcaSlicer settings master guide is the broader pillar that this article sits under. Table of contents What ironing actually does to your top layer When to use ironing, and when to skip it Every ironing setting in OrcaSlicer, explained The base settings that decide whether ironing even works How to tune ironing on your printer (step by step) Recommended starting points by material When ironing goes wrong: common fixes Ironing on multi-color and multi-material prints Frequently asked questions Wrap up What ironing actually does to your top layer Ironing is an extra pass the nozzle makes across your freshly printed top solid layer at the same Z height. The nozzle moves slow, extrudes a tiny trickle of fresh filament (something like a tenth of normal flow), and the hot tip glides over the still warm surface. The heat re-melts the existing plastic and surface tension drags it across the ridges between the previous extrusion lines. The trickle of new filament fills the micro gaps. It's literally an iron, in the same way you'd iron a shirt, except the shirt is made of warm PLA and the iron is your hotend. If you look at a normal top solid layer under a desk lamp tilted at a low angle, what you see as "layer texture" on a flat top isn't actually the layer height. It's the ridges where adjacent extrusion lines on the top layer almost touch but don't quite, and the valleys catch the light. Ironing flattens those into a single smoother sheet because the molten plastic redistributes when you sweep across it. The technique came out of PrusaSlicer 2.3 back in 2021. OrcaSlicer inherited it through the Bambu Studio fork, and the implementation across all three slicers is nearly identical. The one big addition Orca tacked on later is the Inset parameter, which landed in early 2025 via PR #7969 after a long community thread (Discussion #2778). I'll get to why that one matters in the settings section. Here's the honest part. Ironing isn't magic. It only works on flat, horizontal, top-facing surfaces. It adds time to your print (anywhere from a couple of minutes for one small flat top to twenty or thirty percent on a print that's mostly stacked flat panels). It exposes every flaw in the layers underneath it, which means you actually have to get top shell layers, top surface pattern, and your flow rate calibration right first or the iron pass will just make a bad surface shinier. Same print, same filament. Top has ironing on, bottom doesn't. The difference is exactly the kind of surface ironing exists to produce. When to use ironing, and when to skip it Use ironing when your model has a flat, horizontal, top-facing surface that someone is actually going to look at. The classic candidates: maker plaques, name signs, badges, base plates for assemblies, decorative panels, coasters, the lid of a parametric box, signage prints, dashboards. Anywhere you want a finish that reads as "manufactured" rather than "additive manufactured", ironing is the cheapest single thing you can do in software to get there. Also use ironing when you're going to glue something. Prusa's documentation calls this out and it's true. Two flat ironed surfaces sit closer together than two ridged ones, so adhesion on glued assemblies improves noticeably. Skip it when: The top is curved, angled, or organic. Ironing only flattens flat. It does nothing useful on a sphere, a face, a terrain tile, a miniature, anything sloped. You're printing a vase or a single-wall ornament. There's no top solid surface to iron. You're printing TPU or soft TPU. The material doesn't melt smooth, surface tension doesn't behave the same way, and the iron pass either does nothing or makes the surface rougher because the nozzle drags. Don't waste the time. You're printing wood-filled, metal-filled, or particle-loaded filaments. The particles don't re-melt cleanly. Prusa's documentation calls these "generally unsuccessful" with ironing and that matches what I've seen. You're printing PLA-CF, PETG-CF, or any carbon-fiber loaded filament. The fibers prevent the smoothing effect. The 3DBite guide calls this out specifically and the result is that you get a still-rough surface plus all the time penalty of ironing for nothing. The top surface is tiny. Ironing's acceleration and deceleration penalty makes a 5mm square not worth the time. You're doing a multi-color print where pigment bleed at color boundaries would be visible. The iron pass drags a small amount of plastic horizontally and on a sharp two-color logo that smear shows. The five second rule: is the top of your part flat and visible? If yes, iron. If no, don't. Every ironing setting in OrcaSlicer, explained Menu path: Process settings > Quality > Ironing. All these settings are tagged Advanced mode in the OrcaSlicer UI, so if you're in Simple mode you won't see most of them. Switch to Advanced or Expert mode in the top right of the process panel. The Ironing section in OrcaSlicer's Quality tab. Type, Pattern, Flow, Line spacing, Inset, and Angle are all here. Type (ironing_type) Four options here, and getting this one right saves you a lot of print time. The internal enum in the OrcaSlicer source confirms there are exactly four: No ironing: feature off. This is the default, and ironing is something you opt into rather than the other way around. All top surfaces: irons every top-facing surface on the model, including the step tops on multi-tier shapes like terraced bases or stepped pyramids. This is what you want when the model has more than one flat plane the viewer will see. Topmost surface only: irons only the absolute final top layer of the print. Faster than "All top surfaces" because it skips intermediate flat planes. Use this when only the very top of the model matters and you don't care about ironing the lower steps. All solid surfaces: irons every solid layer including internal ones. Niche. People use this for translucent display pieces printed at 100% infill where every internal layer is going to be visible. Massive time penalty, don't enable casually. For 90% of plaque, sign, and base-plate work, "Topmost surface only" is the right choice. It's faster than "All top surfaces" and the result is identical for single-tier flat tops. I switch to "All top surfaces" only when I have stepped or multi-tier geometry. Pattern (ironing_pattern) Two options. Only two. I'm spelling this out because there's a persistent myth in forum threads that there's a third option for the iron pattern itself. There isn't. The Monotonic option you've seen in OrcaSlicer is a Top Surface Pattern (different setting, different menu, different effect on the print). For the iron pass itself, OrcaSlicer gives you exactly: Concentric: the iron path follows the outer shape of the surface inward in rings. Good for round, oval, or irregular curved shapes because there's no diagonal pattern crossing the curve. Rectilinear: parallel straight lines back and forth across the surface. Good for square plaques, rectangular plates, signs, anything with hard corners. Faster than the ring pattern on [truncated] --- ## OrcaSlicer Seam Placement: Hide That Z-Seam (2026 Guide) URL: https://orcaslicer.net/orcaslicer-seam-placement/ I printed three vases for a friend's wedding and all three had a vertical scar running up the front. The seam was Aligned (the default) and it'd parked itself right on the most-photographed face of every one. Three reslices later I had it on the back, and on the fourth I tried Scarf Joint, and that's when the seam basically disappeared. Here's how to make any model do that. I've been hunting seams in OrcaSlicer since the V1.x days, and there's no single setting that hides the Z-seam on every model. There are about a dozen, and they interact. This guide walks every one, gives you a decision tree, covers the scarf joint properly, and tells you when to stop tuning and just rotate the model so the ugly side faces the wall. Table of contents What the Z-seam actually is Every OrcaSlicer seam setting explained Which seam position should you pick Scarf joint seam, the 2024 technique Painted seam workflow step by step Pressure advance is half the battle Wipe and retraction tuning for cleaner seams Scarf joint vs traditional aligned, the comparison Troubleshooting common seam problems When you just have to accept some seam Related OrcaSlicer guides What the Z-seam actually is Every closed perimeter loop on a layer has to start somewhere and end somewhere. When the nozzle finishes a loop, the slicer commands a stop, often a small retraction, sometimes a wipe move, then a travel, then a de-retraction and a fresh start on the next layer's perimeter. That stop-restart cycle leaves a tiny imperfection. There's usually a small blob from residual nozzle pressure at the stop, sometimes a small gap if pressure advance over-retracts, and almost always a visible discontinuity in the wall. Stack 200 layers on top of each other and the lift points form a vertical scar running the full height of the print. That scar is the Z-seam. It's structurally fine but cosmetically ugly, and it's worst on smooth curved surfaces (vases, helmets, figurines), single-perimeter prints, glossy or silk PLA where any imperfection catches the light, and PETG and TPU where flow control is just harder than with PLA. Three honest truths. First, you cannot fully delete the seam in standard FDM. You can hide it, move it, or smear it across layers, but you can't eliminate it without spiral vase mode, which only works for hollow single-wall objects. Second, an aligned seam stacked on a hidden facet beats a scattered random seam for most cosmetic models. Third, a clean Z-seam comes from tuning extrusion as much as placement. Pressure advance, retraction, and wipe matter as much as which seam_position you picked. If you're brand new to Orca's settings layout in general, the OrcaSlicer settings master guide covers the whole Process panel and where every tab lives. The seam options I'm about to walk through all sit inside Process settings, Quality, Seam. Every OrcaSlicer seam setting explained The Quality tab in Process settings holds every seam control you'll touch. Seam position (the main dropdown) Variable: seam_position. Options: Aligned, Aligned Back, Nearest, Back, Random, Painted. Default in the system preset is Aligned (VERIFY, since some printer presets ship with Aligned Back). Aligned stacks every seam on the same vertical line, preferentially on a hidden internal facet. It's the right pick for anything with corners. On a cube, the seam climbs one corner and you can run your finger up it and barely feel it. Aligned Back combines Aligned with a bias toward the rear of the print (negative Y in Orca's coordinate space). It still hunts for a hidden facet, but prefers the back over the front. This is my default for figurines and display models. Open issue #12017: Aligned Back doesn't behave correctly on mirrored parts. If you've mirrored a model and the seam still ends up on the front, switch to plain Aligned and rotate 180 degrees. Nearest puts the seam at the nearest starting point compared to where the nozzle stopped on the previous layer. Print time is shortest with Nearest, but the seam can spiral around the model in a visible diagonal. I avoid it for cosmetic prints. Back forces the seam to the rear (minimum Y point each layer). It ignores geometry hints, which makes it predictable but ugly on round and cylindrical models. Random distributes seams across the whole surface, replacing a continuous line with a stippled pattern. Good for mechanical parts where you'd rather spread weak points across many spots. Bad for cosmetic prints because the eye picks up the dot pattern as noise. Painted honors your manual Enforce/Block markup from the Seam Paint tool. The big gotcha: if you paint a seam line but leave seam_position on Aligned or Nearest or Random, the painted markup is treated as a soft hint, not an enforced location. This is the root cause of GitHub issue #5621. Switch the dropdown to Painted or your paint won't be honored. Seam gap Variable: seam_gap. Default around 10% (VERIFY). Recommended range 0 to 15%. This shortens the loop slightly so the perimeter start and end don't physically overlap and form a bulge. Larger gap means less bulge but higher chance of a visible split. Smaller gap means less split but higher chance of a blob. If your seam looks like a zit, gap might be too small or PA too low. If it looks like a crack, gap is too large or PA too high. When you turn on scarf seam, drop seam gap to 0. Staggered inner seams Variable: staggered_inner_seams. Default off (VERIFY). Moves the inner-wall seam start around across layers so internal seams don't stack into a structural weak line. Mostly a strength feature. Turn it on for parts that take stress along the seam axis. Wipe controls Wipe on loop (wipe_on_loops): pulls the nozzle inward at loop end to reduce the over-extrusion blob. On for traditional seam. Off for scarf joint. Wipe before external loop (wipe_before_external_loop): does de-retraction inside the model so leftover ooze lands on the inner wall. Default off (VERIFY). Turning it on is one of the highest-value changes for clean outer-wall seams. Role-based wipe speed (role_based_wipe_speed): syncs wipe motion with feature print speed instead of travel speed. Leave it on. Retract before wipe: set to 0 when transitioning inner wall to outer wall. Small change, real effect. Which seam position should you pick I get asked this constantly, so here's a rough flowchart. It assumes your pressure advance is calibrated. If it isn't, none of this matters and you'll get a blob no matter which option you pick. More on that below. Display figurine, vase, statue, helmet: Aligned Back, plus seam painting on the rear surface, and orient the model so the ugly side faces -Y. Mechanical part where strength matters more than looks: Random, to spread the layer-start weak points across the whole surface. Box, enclosure, anything with corners: Aligned. The seam will naturally hunt for a sharp concealed corner and hide on it. Tall round cylinder or pillar with no corner to hide on: Scarf joint seam (Contour, conditional on, length around 20 mm), or paint a back-facing line and use Painted. Random on cylinders looks like noise. Single-perimeter ornamental print: Spiral vase mode if the geometry allows it. Otherwise scarf seam with a low slope speed. Visible curved surface where nothing else works: Painted, following a panel break, hair line, or design seam line. Pick where the eye expects a join. Same Aligned setting, three geometries. The cube hides it on a corner, the cylinder shows it as a vertical line, and the sphere shows it as a faint diagonal because there's no flat facet for the slicer to lock onto. Honest caveat: on a smooth round model with single perimeters, there is no setting in OrcaSlicer that makes the seam invisible. Scarf joint reduces it to a faint diagonal smudge. Everything else leaves either a line or a stippled pattern. If you absolutely need an invisible seam on a round single-perimeter print, you sand it after, or you print it in vase mode if the geometry allows. Scarf join [truncated] --- ## OrcaSlicer Infill Patterns Compared: Gyroid, Cubic, Honeycomb (2026) URL: https://orcaslicer.net/orcaslicer-infill-patterns/ I've slung the same Benchy through nine different OrcaSlicer infill patterns this month, weighing each one and timing it, and the answer to "what's the best infill" depends entirely on what "best" means for the part. Gyroid is the safe default. It's also wrong half the time. Here's the matrix that actually answers the question. The Sparse Infill Pattern dropdown in OrcaSlicer 2.3.2 lists more than twenty options. Most users only ever need three of them. The other patterns aren't filler, but they're tuned for edge cases (shock loading, resin-filled interiors, multi-color flushing, vase-mode shells) that most prints don't trigger. This guide is the part of the OrcaSlicer settings master guide that tells you which pattern to pick and, just as important, when to ignore the popular advice. Table of contents The 30-second answer How OrcaSlicer's infill actually works Decision matrix: pick the right pattern for your job Every pattern in the dropdown, explained Strength tests: what CNC Kitchen actually found Density vs pattern: which one matters more? OrcaSlicer-specific settings around the pattern Multi-color and flush-into-infill Common pitfalls and fixes The bottom line The 30-second answer If you don't want to read the whole thing, here's the short version. I've been printing functional jigs and display props on the same machine for two years, and 95% of my parts use one of three patterns. Gyroid at 15 to 20% density for any functional part where the load direction isn't obvious. It's isotropic (roughly equal strength in every direction), it's reasonably fast, and it doesn't fail catastrophically when you push it. Lightning at 10 to 15% density for display models, cosplay shells, statues, and anything you don't plan to drop. It uses about a third of the material of Cubic and prints in roughly half the time. It carries no useful structural load. Cubic at 25% for "I have no idea what this part will be used for." Cubic is the safe coward's pick. Genuinely isotropic, fast to slice, and it's been the default in PrusaSlicer-derived slicers forever for a reason. That covers the vast majority of real-world prints. The other 25-plus patterns exist because edge cases exist: shock loading needs TPMS-FK, multi-color flush-into-infill needs big internal cavities, vase-mode shells use Concentric, and so on. We'll get to all of them. If you came here because your slicer is taking forever or your top surface is pillowing, scroll to the pitfalls section. The fix is almost never the pattern itself. It's a setting next to the pattern. How OrcaSlicer's infill actually works Infill lives under Process > Strength in the prepare panel. The Sparse Infill Pattern dropdown is one of three knobs on that row. The other two (Sparse Infill Density and Sparse Infill Direction) are bigger levers than the pattern in most situations, and we'll get to that. But for now, this is the screen you're looking at. The Strength tab in OrcaSlicer 2.3.2. Density and pattern sit on the same row, but they don't have equal weight in the final result. OrcaSlicer actually has three different infill systems running in parallel, and a lot of people miss this: Sparse Infill is the patterned interior at less than 100% density. This is the dropdown everyone argues about. Internal Solid Infill is the dense fill the slicer drops at 100% density under top layers, above bottom layers, and in transition zones. It has its own pattern dropdown (internal_solid_infill_pattern), which defaults to Monotonic for wide areas and Concentric for narrow strips. Top Surface Pattern and Bottom Surface Pattern are separate dropdowns again, controlling only the visible outer faces. Monotonic is the modern default and it's the right answer almost always. If your top surface looks bad, you don't fix it by changing the sparse infill pattern. You fix it by changing the top surface pattern, the number of top layers, or the surface flow. If your part feels weak, that's the sparse pattern conversation. Keep them straight. Under the dropdown row sit several settings that matter more than which pattern you pick: Infill Anchor, Infill Anchor Max, Infill Wall Overlap, Infill Combination, and Sparse Infill Direction. We'll cover those in the OrcaSlicer-specific settings section, but flag this now: tuning anchor and overlap fixes more "weak part" complaints than swapping Gyroid for Cubic ever will. Decision matrix: pick the right pattern for your job Here's the matrix I keep open on my second monitor when I'm slicing. I built it by cross-referencing the OrcaSlicer Patterns wiki with CNC Kitchen's empirical crush-test rankings and about a year of my own failed prints. JobPrimary pickBackupWhy Functional, unknown load directionGyroid at 20 to 30%TPMS-DTruly isotropic, no in-layer crossings, interconnected channels Functional, known unidirectional loadAligned Rectilinear with direction set to load axisCubic at 25%One strong axis is cheaper than full isotropy when you know the loading Maximum stiffness, time isn't a concernHoneycomb at 30%Triangles at 30%Highest kg-to-density per CNC Kitchen rankings; pay the print-time cost Speed and material economy with structural dutyGrid at 15 to 20%RectilinearFastest of the strong patterns; watch for nozzle-drag failures at intersections Display model, no structural dutyLightning at 10 to 15%Adaptive CubicLowest material and time; pair with 5 to 6 top layers to fix pillowing Hollow vase or shellConcentricn/aVase mode bypasses infill anyway; Concentric matches visually for translucent walls Multi-color with flush-into-infillCubic, Gyroid, Grid, Triangles at 15 to 20%Tri-hexagonBig interior cavities absorb purge; avoid Lightning and Concentric Tall thin part loaded along its long axisAligned RectilinearHoneycombBeam loading benefits from fiber-aligned infill Flexible filament (TPU)Gyroid or ConcentricLateral LatticeSmooth curves deform uniformly; rigid intersections crack flex parts Resin-fillable interiorGyroid or Archimedean ChordsHilbert CurveContinuous interconnected voids let resin flow throughout Shock or impact loadTPMS-FK (Fischer-Koch S)GyroidBone-like topology designed for energy absorption Large brackets, big interior volumesAdaptive CubicSupport CubicDensifies near walls only; real material savings without strength loss Two notes on how to read this. First, "primary pick" doesn't mean the only answer. If you've been printing Gyroid at 20% for the last six months and your parts hold up, don't change anything. Second, if your job isn't on this list, the safest fallback is Gyroid at 20%. I've never had Gyroid fail in a way that surprised me. Every pattern in the dropdown, explained OrcaSlicer's wiki lists roughly 28 dropdown options, though the live UI in 2.3.0 was missing a few (issue #10414 documented this). The current 2.3.2 build exposes the full set as far as I can see. Here's every pattern, organized by family. I've kept each entry tight: what it does, when to use it, when not to. The Sparse Infill Pattern dropdown in OrcaSlicer 2.3.2. There's the full menagerie. We're going to walk through all of it. Linear patterns Rectilinear. Parallel lines that alternate 0 and 90 degrees between layers. Cheap, fast, weak in Z. Use it for prototypes where you'll throw the part out anyway. The wiki itself suggests considering Zig Zag instead for slightly better layer bonding. Aligned Rectilinear. Same lines, but every layer points the same direction. Strong only along the line axis. Set infill_direction to your load axis (a long bracket loaded vertically wants vertical lines). I use this for beam-like parts and almost nothing else. Line. Slightly rotated parallel lines. Same complaints as Rectilinear. There's no use case where Line beats Rectilinear cleanly, so I skip it. Monotonic / Monotonic Line. Rectilinear with smarter path planning so the extruder always moves the same direction across each surface. Sounds boring, but it produces the cleanest top surfaces in the slicer. Use Monotonic as your top [truncated] --- ## OrcaSlicer Tree Supports & Organic Supports: Complete 2026 Guide URL: https://orcaslicer.net/orcaslicer-tree-supports/ I scrapped a Spider-Man bust last month because I'd left supports on Normal and then spent 90 minutes with a knife and pliers carving them off the back of his head. The model itself printed in four hours. Tree supports would've taken twenty minutes to remove and saved me 18g of filament. The slicer's default isn't the right default for figurines, and that's the whole reason this article exists. OrcaSlicer ships with support_type set to normal(auto). That's a defensible choice for square mechanical parts and flat overhang shelves, but it's the wrong choice for almost everything else. If you print figurines, miniatures, busts, organic shapes, or anything tall and narrow with overhangs scattered up the side, the tree and organic styles waste less filament, leave smaller witness marks, and peel off in seconds instead of minutes. They're also slower to slice, RAM-hungry on large models, and have a real failure mode if your nozzle is too small or your branches are too thin. I'll cover all of it. Every default value below was pulled from the OrcaSlicer source (src/libslic3r/PrintConfig.cpp, lines 5853 to 6345 on the main branch). Where the source key has both a classic-tree and an organic variant, I'll show both because the defaults differ in ways that matter. Table of contents Normal vs Tree vs Organic: what's actually different Where to find tree support settings in OrcaSlicer Every tree support setting, explained with defaults When to use tree, normal, or organic: decision matrix Tuning workflow: a step-by-step first pass Common failures and how to fix them Soluble supports: PVA, BVOH, and HIPS Tree supports plus multi-color printing FAQ Wrap-up Normal vs Tree vs Organic: what's actually different There are three families. Inside each family there are sub-styles. The dropdown in OrcaSlicer calls them all "support styles", which doesn't help when you're trying to figure out what does what. Normal supports (Grid and Snug) Normal supports are what most slicers used for the first decade of FDM printing. They build vertical pillars in a grid (or follow the perimeter snugly) under every overhanging face. They're dense, predictable, fast to slice, and they hold well during the print because they're effectively a solid prism of plastic under each overhang. The trade-off is removal and surface quality. A grid pattern leaves a flat-grid impression on the underside of every supported face. With support_top_z_distance at the default 0.2 mm, the contact layer fuses just enough to the model that you'll need pliers, a knife, and patience to get it off without scarring. Snug is a little better than Grid for surface finish because the support hugs the perimeter instead of running a uniform pattern, but it's still a contiguous wall of material against your model. Normal is the right choice when the geometry is simple, the supported face is hidden, and you want predictable print times. Classic tree (Slim, Strong, Hybrid) Classic tree supports grow upward from the bed (or the model surface) like a real tree. The trunk splits into branches, each branch ends in a tip that contacts the overhang only where needed. The point is to use less material, leave smaller marks, and avoid building a useless tower under empty space. OrcaSlicer's classic tree comes in three flavors: Slim: narrow trunks, low material use, easiest cleanup, fragile under heavy overhangs. Strong: fatter trunks, more perimeters per branch, better for big overhangs but slower to slice and harder to remove. Hybrid: classic tree under sparse overhangs, normal-grid under flat shelves. Most flexible. A community tester in GitHub Discussion #10609 ranks Hybrid worst on contact-layer quality among the four styles because the grid sections inherit normal's scarring problem, which matches what I've seen on shelf-heavy prints. Classic tree slices noticeably slower than normal. The branch-routing solver has to figure out where each trunk goes and which model walls to avoid. Strong is the slowest of the three because it carries more perimeters per branch. Organic supports Organic is the newest style and it's now OrcaSlicer's default when you pick tree(auto). The algorithm came from Thomas Rahm's improved port of Cura's tree support, was refined into PrusaSlicer 2.6, and was ported into OrcaSlicer shortly after. As of the current 2.3.x release line, OrcaSlicer keeps expanding the organic path (the 2.3.2 release added rectilinear, grid, and honeycomb base-infill patterns for organic trees, which were previously hollow-only). Branches are shorter, smoother, and circular in cross-section. The solver merges multiple tips into a shared trunk when it can, which is why organic shines on figurines: a face with four separate overhangs (chin, hair, ear, shoulder) gets four tips off one or two trunks instead of four full towers. Organic uses the least material of any style. Prusa's own organic-supports page is the canonical write-up and credits the algorithm as an evolution of Thomas Rahm's improved Cura tree-supports port. Community benchmarks on figurine-style models commonly show organic cutting support material by 30 to 50 percent versus normal grid, with the biggest savings on tall sparse geometry. Print times are typically lower than classic Slim because branches are shorter and fewer perimeters get printed. Removal is the easiest of any style because the tip diameter (default 0.8 mm) is the contact patch. The downside: it's RAM-hungry. PrusaSlicer recommends 16 GB or more for large models with organic supports. On 8 GB systems with big plates, the slicer can hang or out-of-memory crash (issues #7507 and #3871 both document this). The comparison table AspectNormal GridNormal SnugTree Slim / Strong / HybridOrganic Material useHighestHighMediumLowest Print timeMedium-highMediumMediumLower Slice timeFastFastSlow (Strong slowest)Slow, RAM-heavy Removal easeHardestMediumMedium-easyEasiest Surface scarringWorst (grid pattern)ReducedSmall isolated dotsSmallest dots Mid-print stabilityExcellentGoodVariableGood Best forSquare mechanical partsMid-size functionalTall sparse modelsFigurines, default Reading the table the lazy way: if you don't know what you want, pick Organic. If your model is a flat-bottom mechanical bracket with a clean overhang shelf, pick Normal Grid. Don't pick Hybrid unless you specifically want classic tree under your sparse overhangs and a grid under your shelves and you're willing to accept worse surface marks. For more on which print settings fit which model class, the OrcaSlicer Settings Master guide walks through the full Process panel including supports, infill, walls, and quality. Where to find tree support settings in OrcaSlicer Process panel on the right. Click the Support tab. Make sure your settings mode is set to Advanced or Developer in the bottom-right of the panel, otherwise most of the tree-specific fields are hidden. The screenshot above shows my usual setup for assessing whether a model needs tree supports. The Buddha bust is the test piece because it's got everything: smooth curves, undercuts on the chin, overhangs at the shoulders, and tight detail on the face. If your support style works on a Buddha, it works on most figurines. Toggle support on first. Then set support_type to tree(auto). The Style dropdown will switch its meaning: when type is normal, "default" means Grid; when type is tree, "default" means Organic. You can also pick a specific style explicitly (Tree Slim, Tree Strong, Tree Hybrid, Organic) which is what I usually do because explicit beats implicit when you're tuning. Once you're in tree mode, an extra block of settings called "Tree supports" or "Organic supports" appears further down the Support tab. Some fields are duplicated between the two: branch angle, branch distance, and branch diameter all have separate config keys for the classic-tree variant and the organic variant. The slicer auto-applies whichever one matches your style. Every tree support setting, explained with [truncated] --- ## Best OrcaSlicer Printer Profiles in 2026: Complete Master Hub URL: https://orcaslicer.net/orcaslicer-printer-profiles-hub/ I've added eight printers to OrcaSlicer in two years, and exactly two of them appeared in the dropdown without me hunting down a JSON. The other six required some combination of vendor fork, community profile, or hand-built machine settings, and the difference between a smooth setup and a six-tab nightmare came down to one question: which brand made the printer. That's the thing nobody tells you when you switch slicers. OrcaSlicer ships with profiles for more than sixty brands out of the box, but "supported" is a sliding scale. A Bambu X1 Carbon is a ten-second setup. A Snapmaker U1 needs the Snapmaker fork to behave properly. A Prusa MK4 technically works, but the pressure-advance numbers have drifted from PrusaSlicer's official defaults and you'll fight ghosting until you recalibrate. A Kobra S1 will print, but the change-filament G-code in upstream Orca is missing pieces that Anycubic Slicer Next gets right. So this is the master map. I'll walk through what an OrcaSlicer profile actually contains, where each brand falls on the support spectrum, which vendors maintain their own forks, and how to add a printer manually when yours isn't in the list. Seven per-printer guides are linked through this article, plus a brand-by-brand matrix you can scan in thirty seconds. What's in this guide What an OrcaSlicer printer profile actually is Where printer profiles come from in 2026 The brand-by-brand support matrix Brand notes: Bambu, Creality, Prusa, Anycubic, Elegoo, Voron, Sovol, QIDI, Flashforge, Snapmaker The "I have printer X" decision tree How to add a printer profile manually Where profile files live and how to back them up Common gotchas across brands After you install the profile: calibrate Multi-color profiles and AMS FAQ What an OrcaSlicer printer profile actually is OrcaSlicer doesn't store "a printer profile" as one thing. A working configuration is four JSON objects stacked on top of each other, and once you understand the stack the brand matrix makes a lot more sense. The four layers: Printer Model (type: "machine_model"): the family. Brand, family name, default nozzle, default bed STL. Example: "Bambu Lab P1S". Printer Variant / Machine (type: "machine"): a specific hardware combination. Nozzle diameter, tool count, plate type. This is where printable_area, retraction, max accelerations, and kinematics live. Filament (type: "filament"): temperatures, fan curves, flow ratio, max volumetric speed. Lists compatible_printers, so a "@BBL X1C" PETG-CF profile won't appear in a Creality dropdown. Process (type: "process"): layer height, perimeters, infill, support, speed presets like "0.20mm Standard @BBL X1C 0.4 nozzle". All four use the inherits field to chain. A user PETG profile inherits from "Generic PETG @BBL X1C", which inherits from "fdm_filament_pet". Edit a parent and every child changes. Delete a parent and the children vanish from the UI, even though the JSON files are still on disk. I've hit this personally after pruning what I thought were duplicate presets. A working setup is the printer profile, filament profiles for each AMS slot, and the process preset on the right. Three layers of the stack, one screen. System presets ship with the installer and live under resources/profiles/. User presets (anything you create or import) live in the per-OS user data folder; exact paths are in the where-files-live section. You can't edit a system preset in place; the first change forces a "Save as new user preset" prompt. Profile bundles export as .orca_printer, .orca_filament, or .orca_process files (zipped JSON). Import via File → Import → Import Configs. Where printer profiles come from in 2026 Four channels. You'll use at least two of them in a typical setup, and most of the brand-specific advice below boils down to "which channel should you start with for this brand." Channel 1: bundled / official upstream. Anything in resources/profiles/<vendor>/ ships in the installer. The current bundle covers more than sixty vendor folders including Anycubic, BBL (Bambu), Creality, Elegoo, Flashforge, Prusa, Qidi, Raise3D, Snapmaker, Sovol, UltiMaker, Voron, plus the OrcaFilamentLibrary for materials. The list grows monthly, so confirm by opening the wizard and looking. Channel 2: vendor forks. Three printer makers ship their own customized OrcaSlicer build with deeper integration: Snapmaker Orca: a Snapmaker-maintained fork. The U1 tool-changer is first-class here and rough-around-the-edges in upstream Orca. Artisan, 2.0, J1, and J1s support is "in active development" per Snapmaker's own page. Adds device management and Snapmaker-tuned filament defaults. Orca-Flashforge: Flashforge's fork, optimized for the Adventurer 5M, 5M Pro, AD5X, and Guider 3 Ultra. Adds FlashMaker app integration for wireless print and remote monitoring. Anycubic Slicer Next (Orca version): Anycubic's fork built on the OrcaSlicer codebase. First-class for the Kobra 3 family and Kobra S1. Some printer-specific G-code (filament-change macros, the Kobra 3 model-size fix) was only correct in Anycubic Slicer Next for a while; upstream Orca's Kobra S1 profile was missing the change-filament G-code until December 2025, when PR #11650 ported it from Anycubic Slicer Next (issue #11632). If you're on an older build, update before swapping colors mid-print. I'll be blunt about this: if you own a U1, a 5M Pro, or a Kobra S1, install the vendor fork. The OrcaSlicer team isn't trying to lock you out, the vendors just maintain those edge cases better than upstream does. Channel 3: community. User-uploaded bundles on MakerWorld, Printables, and GitHub. The repos that come up most often in Reddit threads and forum links: mjonuschat/OrcaSlicer-Profiles for Klipper and Voron, shyblower/Shyblowers-OrcaSlicer-Profiles for a multi-printer collection, R3tr0_D's Printables uploads for the Creality K1 / K1C / K1 Max, MPC561/Anycubic-Kobra-S1-Orcaslicer-Profiles and ditschi/OrcaSlicer-user-profiles for the Kobra S1 Combo, and Ace3DPrints/sv06plus for the Sovol SV06 Plus running Klipper. There's also ratdoux/OrcaSlicer-FullSpectrum, a niche fork for U1 multi-color tool-changer users that adds layer-blending features beyond the base Snapmaker Orca. Channel 4: manual import from PrusaSlicer or Bambu Studio. Both are codebase ancestors of OrcaSlicer, so their .ini configs and .zip config bundles can be brought in via File → Import → Import Configs. Filament and process settings transfer almost cleanly. Printer settings need a sanity check on bed shape and start G-code. We cover the practical differences in OrcaSlicer vs PrusaSlicer and OrcaSlicer vs Bambu Studio. The brand-by-brand support matrix This is the keystone of the article. Read across to find your brand, the support level, the path I'd take if I were setting it up tomorrow, and the gotchas that bite people most often. The "bundled models" column is what I see in the wizard at the time of writing; vendors add new printers monthly so check the picker on your install before assuming a model is missing. BrandBundled models (sample)Support levelRecommended pathKnown gotchas Bambu Lab A1, A1 mini, P1P, P1S, P2S, X1, X1 Carbon, X1E, H2D family, H2S, X2D Full, first-class Upstream OrcaSlicer + LAN-only mode + access code Cloud-bound prints route through Bambu Connect; firmware updates still need Bambu Studio; H2D / H2S profiles are recent Creality (FDM) CR-10 SE / V2 / V3 / Max, CR-M4, Ender-3 family, Ender-3 V3 (V3 / SE / KE / Plus), Ender-3 S1 line, Ender-5 line, Hi, K1, K1 SE, K1 Max, K1C, K2, K2 Plus, K2 Pro, Sermoon V1 Full for V3 line and K-series; community-tier for older Ender 3 derivatives K-series via Klipper / Moonraker; bedslingers via standard wizard K1 family runs Klipper, connect via Mainsail port 4409 or Fluidd port 4408; leave root-level Klipper settings to the vendor Prusa Research MINI (standard + IS), MK3S, MK3.5, MK4, MK4S, MK4S HF, CORE One (standard, L, L HF, HF), XL, XL 5T Full for FFF; MMU3 not present in u [truncated] --- ## OrcaSlicer Wipe & Prime Tower: Cut Multi-Color Waste 50%+ (2026) URL: https://orcaslicer.net/orcaslicer-wipe-prime-tower/ I sliced a 4-color phone stand last weekend and OrcaSlicer told me the prime tower would burn through 84g of filament. The model itself was 22g. After two hours of settings work I had it down to 31g of purge with zero color bleed, and almost all of the saving came from three settings that don't show up in any tutorial I've watched. If you're printing multi-color on an AMS, an MMU2 or MMU3, an IDEX rig, or you're a power user running M600 swaps, you already know the prime tower is a tax. The question is how big a tax, and the honest answer is that the slicer's defaults are wildly conservative because they have to assume the worst about your nozzle, your filaments, and your color order. Most people accept the default 280 mm³ per swap because changing it feels risky. It isn't, if you do it in the right order. This guide walks through every setting that controls the wipe tower, every place purge volume can be redirected, and the tuning sequence I use to get a 50% to 80% reduction without color bleed. Table of contents What the wipe and prime tower actually is Why every multi-color print wastes filament Every OrcaSlicer prime tower setting, decoded The flush volume matrix (the headline win) Calibrating flush volumes the proper way Twelve ways to cut waste, ranked Show me the math: a worked savings example AMS, MMU, IDEX: printer-specific quirks Common pain points and fixes When to disable the prime tower entirely FAQ What the wipe and prime tower actually is The wipe tower (PrusaSlicer's name), prime tower (Bambu Studio and most OrcaSlicer menus), prime pillar (Simplify3D), or purge tower (community shorthand) is a sacrificial structure printed beside your main object on multi-material and multi-color jobs. Every time the printer swaps filament, the new color enters a hot end that still has melted residue from the previous color sitting in the melt zone. If you don't dump that contaminated transition material somewhere, the next visible feature on your model prints streaky. The tower is that "somewhere." It absorbs the dirty transition extrusion into a place you'll throw out. It also stabilizes nozzle pressure after the swap so the first wall extrusion of the next color lands at the right flow rate, which matters more than people realize because under-pressured walls show as stripes on the next layer. OrcaSlicer's slicing panel makes the flush total easy to ignore. On a small model the per-swap volume can outweigh the part itself, which is the whole reason this article exists. There are exactly three places that purge volume can go in OrcaSlicer: Into the prime tower. The default. Always works, wastes the most filament, because every gram you purge is a gram you throw out. Into the object's infill. Purge is dumped into infill walls where you can't see it. Saves a lot of filament if the object has enough hidden volume at the right layer heights. Into a sacrificial flush object. A separate "trash" model absorbs the purge instead. Best when your main object doesn't have enough hidden volume to soak up the swaps. One terminology note before we go further. OrcaSlicer is genuinely inconsistent. The print-settings menu uses "Prime Tower" but most underlying variables are named wipe_tower_*. The wiki page is called multimaterial_settings_prime_tower but lists settings like wipe_tower_rotation_angle and wipe_tower_no_sparse_layers. PrusaSlicer's documentation uses "Wipe Tower" everywhere. Treat them as the same thing. I'll use both names interchangeably below, the way the program does. One more thing worth flagging if you're new to multi-color: there's a separate waste source on Bambu printers that has nothing to do with the prime tower. The "purge poop" ejected through the chute on the P1 and X1, or the on-bed brick on the A1, comes from the cutter upstream of the hot end. That's controlled by Bambu firmware, not the slicer. The prime tower is downstream of the chute purge, and it's the part you actually have control over from inside OrcaSlicer. We'll come back to this in the AMS section. Why every multi-color print wastes filament This is the part most tutorials skip and it matters because if you don't understand it, you'll over-tune and get color bleed, or you'll under-tune and waste spools. The hot end isn't a clean instant-swap mechanism. It's a melt zone. When you push white PLA through a nozzle that just printed black, you're not magically replacing every molecule of black with white. You're shoving white in behind a slug of black, and the first few centimeters of extrusion come out as a gradient. The exact volume needed to flush that gradient out depends on three things: Color contrast. Black in white shows immediately. White in black shows even more. Yellow in red barely registers. The slicer knows this and the auto-calculation uses color delta. Material chemistry. PLA into PLA flushes faster than PLA into PETG, because the melt temps and viscosities differ. The auto-calc doesn't always weigh material, which is why PLA to PETG transitions need bumping by 1.2 to 1.5 times the auto value at minimum. Nozzle and hot end geometry. A hot end with a steep transition cone holds less residual than a long melt zone. This is why a fresh hardened-steel CHT-style nozzle behaves differently from the stock brass on the same printer. The slicer can't measure your specific nozzle's residue, so it picks safe defaults. Bambu Studio sits at around 280 mm³ per swap on stock profiles per multiple community references. That's a lot. For comparison, a single-perimeter wall on a 0.4 mm nozzle at 0.2 mm layer height is about 0.08 mm³ per millimeter of travel, so 280 mm³ is roughly 3.5 meters of wall extrusion thrown out per swap. Multiply that by every color change in a 3-hour print and you're suddenly in "the tower weighs more than the model" territory. Tuning is mostly about calibrating that mm³-per-swap number for the specific filament pairs you actually use, then redirecting whatever you can't avoid into places that don't need to look pretty. We'll do exactly that in a few sections. Every OrcaSlicer prime tower setting, decoded Menu path varies by Orca version, which is annoying. In Orca 2.x the simple-mode controls live in Process settings panel, "Others" tab, Prime Tower section. Switch the settings panel to Advanced (top-right of the panel) and the rest appear under Process settings, "Multimaterial" tab. Older Orca docs and some forks like Anycubic put the toggle under a "Multiple Extruders" group instead. If you don't see what I'm describing, switch the mode dropdown to Advanced and check both tabs. Process-settings level (per-print) SettingVariableWhat it does Enable prime towerenable_prime_towerMaster on/off. Defaults ON for any multi-material printer profile. Prime tower widthprime_tower_widthDiameter of the rectangular footprint. Community threads cite a default around 60 mm; Snapmaker U1 users have dropped this to 20 to 25 mm with stable results. Prime volumeprime_volumeVolume of priming extrusion per tool change in the tower itself, separate from the per-pair flush volume in the matrix. Brim widthprime_tower_brim_widthBrim around the tower base for adhesion. Default sits around 3 mm; raising to 4 to 8 mm stops most topples. Wipe tower rotation anglewipe_tower_rotation_angleRotates the tower's purge-line direction relative to the X axis. Maximal bridging distancewipe_tower_bridgingMaximum distance between supports on sparse infill sections inside the tower. Purge lines spacingwipe_tower_extra_spacingSpacing of the purge lines themselves. Extra flow for purgewipe_tower_extra_flowMultiplier for thicker or thinner purge lines vs normal extrusion. Maximum wipe tower print speedwipe_tower_max_purge_speedDefault sits around 90 mm/s per the wiki. Raising it requires verified bridging behavior. Wall typewipe_tower_wall_typeRectangle (default), Cone (fillet at base), or Rib (four ribs for stability). Stabilization cone apex anglewipe_tower_cone_angleApex angle of the cone wa [truncated] --- ## OrcaSlicer M600 Filament Change: Multi-Color Without AMS (2026) URL: https://orcaslicer.net/orcaslicer-m600-manual-filament-change/ Half the multi-color tutorials online assume you bought an AMS. I'm running an Ender 3 V3 SE that cost less than the AMS Lite alone, and I do clean two-color prints every week using nothing but a single line of G-code. M600 is the trick, but the firmware story is messier than anyone tells you, and OrcaSlicer has its own quirks on top. This guide is the long version of what I wish someone had handed me two years ago. I'll walk through what M600 actually is, which printers respond to it out of the box, which ones need a macro or a firmware flash, and exactly where to click in OrcaSlicer to make it all work. I'll also be honest about when an AMS is genuinely the better answer, because for some workflows it is. Table of contents What is M600 and what does it actually do Does your printer even support M600? OrcaSlicer setup: pointing the slicer at your M600 Technique A: a single color change at one layer Technique B: full multi-color with painted regions Klipper M600 macro, line by line Enabling M600 on Marlin (Ender 3 etc.) What happens at the printer when M600 fires Common mistakes and how to fix them M600 vs AMS vs prime tower Bambu A1 / A1 Mini without AMS: the special case FAQ What is M600 and what does it actually do M600 is a Marlin G-code that triggers a filament change pause. The official Marlin documentation puts it plainly: "the basic procedure will move the print head away from the print, eject the filament, wait for new filament to be inserted and the user to confirm, load and prime the filament, and continue with the print." That's the whole pitch. One line in your sliced file, and the printer stops cleanly so you can swap colors. The catch is that M600 is a Marlin command. Other firmwares either reimplement it, ignore it, or replace it with something completely different. That's where most of the pain comes from. People drop M600 into a gcode file, watch their printer plow straight through it, and conclude that OrcaSlicer is broken. It isn't. The slicer is doing its job. The firmware just doesn't speak that dialect. The Marlin recipe, step by step When a properly configured Marlin board hits an M600 line, here's what happens in order: Retract by PAUSE_PARK_RETRACT_LENGTH (or by the E parameter you pass) so molten filament doesn't ooze. Park the head, moving to the X/Y coordinates set by NOZZLE_PARK_POINT with a configurable Z lift on top. Unload by FILAMENT_CHANGE_UNLOAD_LENGTH, which is typically around 100 to 120mm for Bowden setups and roughly 50mm for direct drive. Beep and prompt on the LCD: "Insert filament and press to continue." The number of beeps comes from FILAMENT_CHANGE_ALERT_BEEPS. Wait for you to load the new spool and click the encoder. Load by FILAMENT_CHANGE_FAST_LOAD_LENGTH, then prime by FILAMENT_CHANGE_SLOW_LOAD_LENGTH. Purge loop: the printer asks "More purge?" until you say no. Return to the original X/Y/Z and resume extrusion. You can override almost any of those values inline. M600 X10 Y15 Z5 parks at X10 Y15 with a 5mm Z lift, for example. The full parameter list (per the Marlin docs) covers T (tool index), E (initial retract), U (unload length), L (load length), X/Y/Z (park position), B (beep count), and R (resume temperature). For 95% of single-extruder workflows you'll never touch any of them. Bare M600 just works, assuming the firmware supports it. How Klipper handles it (it doesn't, until you add the macro) Klipper has no native M600 because M600 is Marlin's invention. Drop it into a gcode file on a stock Klipper printer and you'll get an "Unknown command" error in the console while the print continues underneath it. To make it work you need to define a [gcode_macro M600] block in printer.cfg, and you also need [pause_resume] enabled. The Klipper team ships a sample in config/sample-macros.cfg in the official repo. I'll walk through that whole file later in the article. How Bambu handles it (it doesn't, period) Bambu firmware is the awkward one. The X1, P1, A1, and A1 Mini all use Bambu's own AMS-flavored toolchange codes (community-reverse-engineered as M620 and M621) and they don't reliably respond to bare M600. Multiple Bambu Lab forum threads confirm this: people insert M600, the printer ignores it, the print continues. If you're on a Bambu without an AMS you have to use the slicer's pause workflow with custom Bambu-specific gcode. There's a section near the end of this article that covers that path. Does your printer even support M600? Before you spend an hour configuring OrcaSlicer, run a five-second sanity check on your firmware. The matrix below is what I've confirmed against firmware docs and forum posts. FirmwareM600 native?Setup needed Marlin 2.x stock (Ender 3 etc.)Usually noUncomment ADVANCED_PAUSE_FEATURE in Configuration_adv.h and reflash Marlin 2.x with ADVANCED_PAUSE_FEATUREYesNone beyond the flash. LCD recommended for the press-to-continue prompt KlipperNoAdd [pause_resume] and a [gcode_macro M600] to printer.cfg RepRapFirmware (Duet)Yes, configurableCreate /sys/filament-change.g, otherwise it falls back to pause.g Prusa firmware (MK3, MK4, Mini, XL)YesNothing. Best-supported implementation out there Bambu firmware (X1, P1, A1, A1 Mini)NoUse Bambu Studio / Orca pause feature with community workaround gcode Stock Creality Klipper (K1, K1 Max, K1C)NoInstall Creality Helper Script or write your own M600 macro Stock Creality Marlin (Ender 3, V2, Pro 4.2.7)NoFlash a community Marlin build with the pause feature enabled Ender 3 V3 SE / V3 KE / V3NoCreality-modded Klipper. Same fix path as standard Klipper, or use the slicer's PAUSE The fastest test I know is to write a five-line gcode file by hand: G28 G1 Z20 F1500 M600 G1 Z40 F1500 M104 S0 Send that to your printer. If it homes, lifts, then pauses and beeps, your firmware understands M600. If it homes, lifts, lifts again, and turns the heater off without ever stopping, your firmware doesn't. No need to slice an actual model and waste filament finding out. OrcaSlicer setup: pointing the slicer at your M600 OrcaSlicer doesn't generate M600 by itself. It generates whatever string you put in the "Change filament G-code" field. Once you've confirmed your firmware speaks M600, you're going to tell Orca to inject that string at every color boundary. The Machine G-code panel inside Printer Settings is where Orca stores all the firmware-level snippets that get baked into your sliced files. Open the printer profile by clicking the small wrench icon next to your printer name in the right panel, then switch to the Machine G-code tab. You'll see a column of fields: Machine start G-code, Machine end G-code, Before layer change G-code, and a few rows down, Change filament G-code. That last one is the field you care about for this whole article. For Marlin and Prusa printers, a single M600 in this field is all you need. Klipper users typically put M600 here too, with the supporting macro in printer.cfg. What you type depends on your firmware. For Marlin with the pause feature enabled, Prusa, and Klipper with a macro defined, just put: M600 For Klipper without an M600 macro, you can use the bare PAUSE command instead: PAUSE The Marlin docs confirm M600 is the right thing to inject for any Marlin-flavored firmware that has ADVANCED_PAUSE_FEATURE enabled. The Klipper sample-macros.cfg confirms PAUSE works on its own once the [pause_resume] module is loaded. Enable Manual Filament Change Still in Printer Settings, switch to the Multimaterial tab. Find the Manual Filament Change toggle and turn it on. This single switch tells OrcaSlicer that your printer doesn't have an AMS or a toolchanger, so any color transitions should fire your Change filament G-code instead of trying to issue an automated tool swap. Without it enabled, the slicer can either skip the color changes entirely or insert the wrong codes. While you're there, decide whether you want a prime tower. For a single-color change at one layer it's overkill. For full multi-color prints with painted regions, a [truncated] --- ## Multi-Color Printing in OrcaSlicer: Complete 2026 Guide (AMS, MMU, M600) URL: https://orcaslicer.net/orcaslicer-multi-color-guide/ The first time I tried multi-color in OrcaSlicer I burned through 60g of filament on a wipe tower that toppled over at layer 14 and dragged half the print with it. I'd queued a four-color logo plaque overnight, woke up to a beige-and-black puddle welded to the build plate, and spent the morning chiseling PETG off PEI. Three months and a lot of forum reading later, I'm running 4-color prints with under 8g of purge per swap, and the trick wasn't a magic setting. It was understanding which path to multi-color my hardware actually wanted. OrcaSlicer in 2026 is genuinely the most flexible multi-color slicer out there. It's not the smoothest (Bambu Studio's AMS UX still wins on pure polish) and it's not the most stable for every multi-material rig (PrusaSlicer remains the official path for the MMU3, more on that later). But it covers more printers, gives you finer control over flush volumes and tower geometry, and exposes the per-filament knobs that the locked-down slicers hide. That's why it's worth learning properly, and that's what this guide is for. I'm going to walk through every realistic path to multi-color, from a $0 manual M600 setup to a 24-slot AMS chain, with the settings and gotchas I've actually hit. If you're brand new to OrcaSlicer entirely, start with the beginner's first print walkthrough first. Comfortable with single-color slicing already? Read on. Need the installer? Grab it from the downloads page before you start clicking through screenshots that won't match your version. What's in this guide The five paths to multi-color in OrcaSlicer The multi-color hardware landscape in 2026 Setting up filaments and color slots The painting tool, end to end Color change by height with the layer slider Wipe tower and prime tower mechanics Cutting waste from multi-color prints AMS workflow from setup to first print Prusa MMU and OrcaSlicer in 2026 IDEX and toolchanger workflows Verifying the slice in Preview Common multi-color failures and fixes Where multi-color is heading next Wrap-up and next steps The five paths to multi-color in OrcaSlicer Before we touch a single setting, you need to know which of these five paths fits the printer on your desk. Pick the wrong one and nothing else in this guide will save you. Pick the right one and the rest is just tuning. Path 1: Manual M600. Single nozzle, one filament loaded at a time, you swap by hand when the printer pauses. Costs nothing. Works on any Marlin, Klipper, or RepRapFirmware printer. Good for color-by-layer prints (a green base, a yellow body, a red top) but tedious for anything with frequent swaps. I covered this in detail in the M600 manual filament change guide, which is the cluster sibling to this pillar. Path 2: Bambu AMS family. Auto-feed boxes with 4 spools each, chainable up to 24 slots on H-series printers. The most plug-and-play option if you already own a Bambu printer, and OrcaSlicer treats them as first-class. Full setup walkthrough lives in the Bambu AMS guide. Path 3: Prusa MMU2S/MMU3. Five filaments via a selector head on a Prusa printer. Stable and mature on PrusaSlicer, semi-supported on OrcaSlicer through community profiles. I'll be blunt about this in the MMU section below. Path 4: IDEX printers. Two heads on the same gantry. Snapmaker J1/J1S, the older Sovol SV04 line, and a handful of others. Multi-color without purge waste because each color gets its own nozzle. Path 5: Toolchangers. Multiple complete tool heads that swap in and out. Snapmaker U1 (4 tools), Prusa XL (up to 5 tools). Faster swaps than AMS, no purge tower, but pricier hardware. Most readers here are on Path 1 or Path 2. If that's you, the AMS section and the layer-slider section are the two you'll actually use. The other paths are covered for completeness and so you can compare against what you might upgrade to. The multi-color hardware landscape in 2026 Here's where the market sits as of mid-2026. I've kept the table to what OrcaSlicer actually supports, and I've noted the support level honestly, because there's a real difference between "ships a profile" and "works without community workarounds". HardwareSlotsApprox price (USD)OrcaSlicer supportNotes Bambu AMS (original)4 per unit, up to 16 chained on X1/P1~$349Full, first-classRFID auto-detect on Bambu spools, bowden buffer manages hand-off Bambu AMS Lite4 (single-stage open feeder)~$249Full, A-series onlyA1 and A1 mini only, one Lite per printer Bambu AMS 2 Pro4 per unit, chainable to 24 with AMS HT~$369FullActive 65C drying, brushless servo motor, ceramic inlets Bambu AMS HT1 spool~$199FullSingle-spool drying companion to AMS 2 Pro Prusa MMU2S5~$299 usedCommunity profiles onlyNo official profile in OrcaSlicer Prusa MMU35~$369Community profiles onlyIssue 8876 open since March 2025 Prusa XL toolchangerUp to 5 tools~$3,499 to $4,999Multi-tool support, some open issuesTreated as multi-tool extruder printer Snapmaker J1 / J1S (IDEX)2~$1,099 to $1,299Full, profile shipped since v1.6.2-betaCopy, Mirror, Backup, Dual-color modes Snapmaker U1 (toolchanger)4 heads~$899 typicalNative Snapmaker Orca fork plus upstream profileRoughly 5 to 10 second swaps per Snapmaker numbers Bambu H2D / H2D Pro2 nozzles plus AMS chain$1,899 / $3,799Not supportedBambu Studio is the only official slicer Co Print ChromaSet4 or 8 inputs~$299 to $499Works via SEMM workflowMarlin compatible, no Klipper required Manual M600Unlimited (limited by patience)$0FullUniversal, requires user attention at every change Bambu AMS family The original AMS still works fine and you can find them used. The AMS Lite is an open-feed cube that only works on the A1 series and tops out at one unit per printer. AMS 2 Pro is the current flagship: brushless servo motor (Bambu claims roughly 60% faster feeding versus the original, marketing copy so take it with a pinch of salt), active drying to about 65C, ceramic inlets that should last longer against abrasive filaments, and a top tube that lets you swap spools without dragging the printer out of its enclosure. AMS HT is a single-spool sidekick aimed at high-temp materials like PA, PC, and CF/GF composites, with active drying up to 85C (versus 65C on AMS 2 Pro). Chain limits matter. Bambu's H2D supports the maximum configuration: 4 AMS 2 Pro plus 8 AMS HT in series for 24 total slots, where the AMS HT is a single-spool drying unit rather than a 4-bay AMS. On an X1C or P1S with original AMS the practical ceiling is 16. AMS Lite doesn't chain at all. If you're shopping for an X1C or P1S setup, the X1C/P1S guide walks through the AMS configuration in OrcaSlicer for those specific machines. Prusa MMU2S and MMU3 Prusa's MMU3 is the most refined non-Bambu multi-material option on the market. It's also where OrcaSlicer's universal-slicer ambition runs into a wall. As of v2.3.2 there's no official MMU3 profile shipped, the GitHub feature request (Issue 8876) has been sitting open since March 2025 with no PR or assignment, and the path forward for Prusa MMU users is honestly still PrusaSlicer. I cover the trade-offs more in the OrcaSlicer vs PrusaSlicer comparison, but the short version is: if MMU3 is your only multi-material hardware, don't fight the slicer. IDEX printers IDEX means two independent heads on one gantry. The Snapmaker J1/J1S has had an OrcaSlicer profile since v1.6.2-beta, and the v2.2.0 release expanded the IDEX feature set noticeably. Each head holds its own filament, so a color change is a head swap with retraction and a Z-hop, not a 700mm3 purge into a tower. That alone explains why IDEX prints look so much cleaner at the seams: there's nothing to purge. Toolchangers Snapmaker U1 carries 4 full tool heads and parks them on a dock when not in use. Per Snapmaker's own numbers, SnapSwap toolhead exchanges take about 5 seconds, and roughly 10 seconds once you include nozzle prime. Either figure is dramatically faster than an AMS swap, and there's no purge waste either. Take the marketing figures with a healthy pinch of salt, but the structural argument [truncated] --- ## OrcaSlicer Settings Master Guide: Every Setting Explained (2026) URL: https://orcaslicer.net/orcaslicer-settings-master-guide/ I keep the OrcaSlicer Process panel pinned to a second monitor while I slice. Not because I tweak everything, but because the panel has six tabs and somewhere around four hundred individual settings, and the answer to "where does that thing live" is genuinely not always obvious. The tab labels are accurate, but tabs sometimes overlap. Quality holds layer height and seam, but Speed holds the speeds that affect quality. Support sits between Strength and Multimaterial. Brim is in Others, not Strength. This guide is the panel reference I wish I'd had when I first opened OrcaSlicer. It walks through every Process tab, names every setting that matters, explains what it actually does in one or two plain sentences, and points out the thirty most-misunderstood toggles. Written against OrcaSlicer v2.3.2, the current stable build with 13.7k GitHub stars, current as of writing. It's organised by Process panel tab, with deep dives on the settings beginners and intermediate users actually search for. At the end is the profile-hierarchy explainer (Printer settings vs Filament settings vs Process settings, which override which) and a frequently-asked-questions section. How the Process panel is laid out The Process panel sits below the Printer and Filament selectors in OrcaSlicer's right sidebar. It has six tabs across the top, left to right: Quality, Strength, Speed, Support, Multimaterial, Others. You'll spend most of your time on Quality, Strength, and Speed. Two important details before diving in. First, the panel has a mode toggle at the bottom-left labelled Simple, Advanced, or Developer. Many of the settings below are only visible in Advanced mode. A handful are Developer-only. If you can't find a setting, switch modes. Second, the colored chip in the panel header tells you whether the current settings match the saved profile. When the dot's showing, your tweak is unsaved; it'll print, but won't survive switching profiles. Quality tab The Quality tab is the geometry-of-extrusion tab. It's where you tune how the slicer actually puts plastic down: layer height, line width, where seams go, and the precision settings that affect dimensional accuracy. The Quality tab top half holds layer height, all eight line width fields, and the seam controls. Layer height is the single most impactful slider in the entire slicer. Each line width field defaults to a percentage of nozzle diameter. Layer height and first layer height Layer height is the single most impactful slider in the entire slicer. OrcaSlicer's wiki recommends 20% to 80% of nozzle diameter; for a 0.4mm nozzle that means 0.08mm to 0.32mm. The 0.20mm default is the sweet spot for most prints. Drop to 0.12mm for fine detail (figurines, miniatures), bump to 0.28mm for fast prints where surface quality doesn't matter. First layer height is typically larger than the rest of the print, around 0.20-0.24mm. A larger first layer is more forgiving to bed level imperfections. Line width (eight separate sliders) Line width has eight independent fields: default, first layer, outer wall, inner wall, top surface, sparse infill, internal solid infill, and support. They all default to percentages of nozzle diameter. The wiki's recommended starting points: Outer wall: 105 to 120% (slightly wider than nozzle for surface quality) Inner wall: 120% or higher (filling space between perimeters cleanly) Top surface: 100 to 105% Sparse infill: 115% Internal solid infill: 110% Support: 100% Most users never touch these. They're worth tuning when you have specific issues like top-layer pinholes (raise top surface line width) or thin internal walls separating from outer walls (raise inner wall). Seam settings Where each layer's start point lives. Six options for seam position: Aligned (default, all seams stack on one vertical line), Aligned Back (same but biased to the back of the model), Nearest (per-layer-shortest-travel), Back (always at the back), Random (every layer different), and Painted (you brush where you want it in Prepare). For most prints, Aligned plus painting where the seam should hide is the sweet spot. Random looks worse than people expect; the seam stays visible just spread out everywhere. Staggered inner seams (Advanced mode) rotates the inner-wall seam start positions across layers so they don't pile up directly under the outer seam. This is a free quality bump. The "smoosh" of stacked inner seams telegraphs through to the outer wall otherwise. Scarf joint seam is OrcaSlicer's signature feature. Instead of starting and ending each loop at a single Z-seam point, the slicer ramps the first and last segment over itself across multiple layers, like a wood scarf joint. The result is that on curved or organic surfaces, the seam essentially disappears. Slope length defaults to 20mm. Outer wall speed must be below 100mm/s for clean execution. Toggle "applied to inner walls" off unless you specifically want it; it adds slicing time without visible benefit. Precision settings The precision block is for parts that need to fit something else. Slice gap closing radius: how aggressively to close small gaps in the input mesh. Default 0.049mm. Touch only if a model has gaps the slicer is missing. Resolution: how finely to slice arc segments. Default 0.0125mm. Lower values produce smoother curves at the cost of larger G-code files. Arc fitting: replaces small straight segments with arc commands (G2/G3) when the firmware supports them. Smaller files, smoother curves on Klipper / Marlin 2. X-Y hole compensation: enlarges or shrinks holes only. Use when M3 holes won't accept screws (typically +0.1mm to enlarge). X-Y contour compensation: changes outside profile dimensions. Different from hole comp. Elephant foot compensation: shrinks the bottom 1-5 layers inward to counteract first-layer squish bulge. 0.10-0.20mm typical. Crank if your prints look fat at the base. Precise wall: zeros the overlap between outer and inner walls (default has slight overlap). Improves dimensional accuracy by about 0.05mm. Enable for parts that need to fit something. Strength is unaffected per Orca's docs. Known interaction bug with "only one wall on top". Wall generator: Classic vs Arachne This is the big architectural choice. Classic uses fixed line width perimeters; if a feature is thinner than 2× line width, you get a gap. Arachne dynamically varies line width to fit thin features without leaving gaps. Arachne is the default for most modern profiles. Pick Arachne for detailed or organic models where thin features matter. Pick Classic for boxy mechanical parts where predictability matters more than fit-to-shape adaptation. Arachne adds slicing time. Walls and surfaces Only one wall on top surfaces: replaces the standard 2-loop top with a single loop, freeing surface area for top infill to fill cleanly. Best for prints with small letters, logos, or thin top features. Only one wall on first layer: similar idea for the bottom. Walls printing order: Inner/Outer/Inner gives the cleanest exterior (requires 3+ walls). Outer/Inner is fastest. Inner/Outer is the overhang-friendly default. Alternate extra wall (sandwich mode): adds an extra wall every other layer for stronger functional parts without doubling print time. Detect thin wall: for very thin features that don't fit Arachne's full line, prints them as gap-fill instead of a true wall. Bridging How the slicer handles overhangs of more than 45 degrees. Thick bridges sets the bridge layer's flow ratio higher (1.0 default; some users set 0.95-1.05). Extra bridge layer (Advanced) adds a second bridge layer at higher density. Don't filter out small internal bridges keeps tiny bridges that the slicer would otherwise replace with infill. Ironing Smooths top surfaces by re-running the nozzle across them with minimal extrusion. Type options: no ironing (default), top surfaces, topmost surface only, all solid layers. Pattern, flow (about 10% by default), line spacing (0.10mm default), inset, angle, speed (typically 30 [truncated] --- # Section: Bambu Lab printers ## OrcaSlicer for Bambu Lab H2D: Setup, IDEX Modes and AMS URL: https://orcaslicer.net/orcaslicer-bambu-h2d/ The first time I tried to send a sliced plate from OrcaSlicer to my Bambu Lab H2D, nothing happened. I'd forgotten the printer was still sitting in cloud mode, the LAN+Developer toggles were both off, and my first AMS HT had only been clipped to the back of the chamber an hour earlier. That fumble cost me about forty minutes of menu hunting, so this guide is basically the cheat sheet I wish I'd printed before unboxing the thing. OrcaSlicer and the Bambu H2D in 2026: where we actually stand Here's the short version, because you don't need another 800-word preamble before you can slice. As of late October 2025, OrcaSlicer has native H2D and H2D Pro profiles in upstream. They aren't a fork, they aren't a community side-load, they're sitting right inside resources/profiles/BBL/machine alongside the X1C and P1S files. PR #10780, contributed by Noisyfox, merged into main on 24 October 2025 and brought ten H2D-family JSONs with it. If you grab OrcaSlicer 2.3.x or newer today, you'll see the H2D in the Configuration Wizard without doing anything weird. The catch isn't profile support anymore. It's Bambu Connect. Since Bambu rolled that authorisation layer out across the X1, P1, A1 and H series, OrcaSlicer can't push a plate through Bambu Cloud. The legitimate path that Bambu themselves describe for "advanced users" is to flip the printer into LAN-Only Mode plus Developer Mode, then disable "Use legacy network plugin" inside OrcaSlicer. That's the entire trick, and we'll walk through each switch. Laser and cutter workflows, by the way, still live inside Bambu Studio / Bambu Suite. OrcaSlicer doesn't drive those modules. This guide is about 3D printing on H2D, nothing else. Table of contents H2D at a glance: what makes it different from X1C and P1S Confirming OrcaSlicer has your H2D profile Bambu Connect, LAN Mode and Developer Mode First connection: adding the H2D to OrcaSlicer over LAN Dual-nozzle setup and build volume awareness AMS strategy for H2D Calibrations to run before your first serious print Common gotchas H2D Pro: what changes FAQ H2D at a glance: what makes it different from X1C and P1S If you've been running a P1S or X1C, the H2D feels familiar for about five minutes and then completely doesn't. It's a bigger machine, it's heavier (about 31 kg net), and the external footprint is 492 x 514 x 626 mm, so it's not the kind of thing you tuck under a desk shelf. The headline change is the toolhead, which carries two direct-drive extruders instead of one. Bambu officially calls this their "Dual-Nozzle" or "Independent Dual Extruder" system. There's a lively debate in reviewer circles about whether that counts as "true IDEX" in the Prusa XL or Snapmaker J1 sense, since both nozzles share one toolhead with a nozzle-lifting rail rather than two independent X carriages. The Bambu Wiki has a whole maintenance page on the "Nozzle Lifting Rail," which tells you what's physically going on. I'll dodge the IDEX vocabulary war and just use Bambu's own wording for the rest of this article. The thermal envelope is the other huge jump. You're looking at a 350 °C hotend, 120 °C bed and 65 °C actively-heated chamber, which puts engineering polymers like PA, PC, PEI blends and carbon-filled variants firmly on the table. Quoted toolhead speed is 1,000 mm/s at 20,000 mm/s² acceleration, with 50 µm motion accuracy via Bambu's Vision-Assisted Encoder System. Whether you'll print that fast in practice is another conversation, but the chassis is rated for it. Build volume is where the dual-nozzle reality bites. The single-nozzle printable volume is 325 x 320 x 325 mm. When both nozzles are in use, the common print area where either nozzle can reach the same part is 300 x 320 x 325 mm. The total X-axis envelope (the entire reach across both nozzles, alternating) is 350 x 320 x 325 mm. Per Bambu's printable-range wiki, the left nozzle's max printable height is 320 mm, the right is 325 mm, and the common dual area is 300 x 320 mm at 320 mm tall. You'll see those constraints surface in OrcaSlicer the first time you load a tall plate. H2D dual nozzle toolhead close-up showing both hotends and lifting mechanism One more thing worth knowing before you dive in: the H2D base model launched on 25 March 2025 at $1,899, with the H2D AMS Combo at $2,199 and the Laser Full Combos at $2,799 (10 W) and $3,499 (40 W). These are different SKUs with different bundled hardware. I'll talk about the base H2D unless I say otherwise. If you've got the AMS Combo you already have one AMS 2 Pro in the box. If you have a base H2D you start with the original AMS, and we'll deal with that below. Confirming OrcaSlicer has your H2D profile Before you fuss with any network stuff, make sure your OrcaSlicer install actually carries the H2D files. The fastest sanity check is the directory listing on GitHub. Go to resources/profiles/BBL/machine and you'll see ten H2D JSONs: a base Bambu Lab H2D.json, four Bambu Lab H2D 0.2/0.4/0.6/0.8 nozzle.json files, and the same set for H2D Pro. If those files are present, the H2D will appear in OrcaSlicer's Configuration Wizard. Inside OrcaSlicer, the workflow is the standard one: Configuration Wizard, Bambu Lab, H2D or H2D Pro, then pick the nozzle size. If you've already run the wizard once and the H2D isn't there, you're almost certainly on a version older than 2.3.x. Grab the latest stable from the canonical OrcaSlicer GitHub releases page. We don't host binaries here, and you shouldn't trust any third-party mirror for a slicer that's about to talk to a 350 °C hotend. If you're a dev-build user testing PR builds, there's one extra step Noisyfox flagged in the PR: delete the system/ folder inside your Orca config directory before launching the new build, so the latest profile changes actually load. For anyone on stable, that step isn't necessary anymore now that PR #10780 has been merged into main. For background on the upstream feature request, see Issue #9254, and for the H2D Pro request specifically, Issue #12458. Both are good context if you want to see how the support landed. One housekeeping note. If you're managing multiple Bambu printers in OrcaSlicer (I've got an X1C and the H2D paired up), each printer gets its own profile entry. There's a separate guide on that pattern over at our OrcaSlicer printer profiles hub if you want the broader walkthrough. OrcaSlicer printer selection wizard with H2D and H2D Pro entries visible Bambu Connect, LAN Mode and Developer Mode: the three switches This is the part that trips up almost everyone moving from X1C / P1S to H2D, because Bambu Connect changed the rules in 2025. The short story: third-party slicers can't push print jobs through Bambu Cloud anymore. Bambu's own position, stated in their blog about third-party integration with Bambu Connect, is that "advanced users" can keep MQTT, live stream and FTP open by enabling Developer Mode on the printer. That's the door OrcaSlicer walks through. Three switches need flipping, and they have to all be on at the same time. Switch 1: LAN-Only Mode on the H2D itself. Open Settings on the printer's touchscreen and find the network section. Enabling LAN-Only Mode disconnects the printer from Bambu Cloud, which is the trade-off you're making. You won't get cloud monitoring, remote start from the Handy app over the internet, or cloud print history. You will get a printer that responds directly to OrcaSlicer over your local network. Switch 2: Developer Mode on the H2D. Per Bambu's wiki on enabling Developer Mode, this option arrived on the H2D in firmware 01.01.00.01. If your printer hasn't been updated since unboxing and you can't see Developer Mode in the settings panel, that's why. Run the firmware update first, then come back and toggle Developer Mode on. Bambu shows a confirmation dialog warning that you're stepping outside the standard cloud workflow. Accept it. H2D screen showing LAN-Only Mode toggle in network settings Switch 3: Disable "Use legacy network plugin" in OrcaSlic [truncated] --- ## OrcaSlicer Bambu AMS Setup & Tips: Complete 2026 Guide URL: https://orcaslicer.net/orcaslicer-bambu-ams-setup/ My first multi-color print on a Bambu AMS was a four-color Pokeball that came off the plate looking like a tie-dye experiment. The wipe tower was nearly as tall as the model, half my purge was wasted on a transition I didn't actually need, and I'd spent twenty minutes trying to figure out why OrcaSlicer kept showing slot 2 as "Generic PLA" when I'd loaded Bambu Matte. That was my crash course in how the AMS actually talks to OrcaSlicer, and where the conversation tends to break down. I'm writing this guide on May 13, 2026, and the AMS family is now four distinct units instead of one. Mixing them up is the single fastest way to buy the wrong hardware or generate a print that fails on the first filament swap. If you've been Googling "orcaslicer bambu ams" and getting a mess of conflicting advice from 2023 forum posts, this is the current map: what each AMS variant does, how to pair it with OrcaSlicer in either cloud or LAN mode, how to actually configure slots, how to tune the wipe tower so you stop wasting half your spool on purge, and the quirks (TPU, abrasives, RFID, the Bambu Connect middleware drama) that nobody warns you about until you've already wasted a roll. This is a third-party guide. We're not the official Bambu site, and we don't host the slicer. If you need the binary, grab it from SoftFever's GitHub releases. Everything below is what I've learned running OrcaSlicer with an X1C plus standard AMS, a P1S with an AMS 2 Pro, and an A1 mini with an AMS Lite. I'll flag what's verified, what's known-broken, and what the official docs are deliberately vague about. Table of contents The four AMS units in 2026 (compatibility matrix) Connecting your Bambu AMS to OrcaSlicer Slot configuration and filament sync Multi-color workflow: painting, wipe tower, purge tuning Limitations and gotchas (TPU, abrasives, humidity, spool size) AMS firmware quirks and troubleshooting Bambu-specific OrcaSlicer features worth knowing FAQ Final CTA The four AMS units in 2026: don't conflate them Before I touch OrcaSlicer settings, I have to draw a line under what Bambu actually sells right now, because the comment sections of every AMS video are still full of people insisting the AMS Lite works on their P1S. It doesn't. There are four distinct units, plus a hub accessory, and they're not interchangeable. Bambu's own firmware-release page is the only place I trust for AMS compatibility dates. Wiki versions move; community posts don't. AMS (original) The unit Bambu launched with the X1 in 2022. Four slots, enclosed plastic enclosure, passive desiccant tray that you swap when the indicator changes color. Spool spec is width 50 to 68 mm and outer diameter 197 to 202 mm. RFID reads any Bambu-branded spool automatically. It's compatible with the X1, X1C, X1E, P1S, P1P (after the early-2024 firmware update), and the H-series. It is not compatible with the A1 or A1 mini. AMS Lite The open-air four-spool unit that ships with the A1 series. No enclosure, no humidity control, no desiccant. Spool spec is width 40 to 68 mm and inner core 53 to 58 mm. And here's the part people keep missing: AMS Lite is locked to the A1 and A1 mini. It physically uses a different connector and a different firmware path. It does not work on the X1C, P1S, P1P, X1E, or anything in the H-series. If you're shopping for an AMS for your X1C, you don't want this one. One AMS Lite per A1-series printer gets you four colors. The A1-series AMS Hub lets you stack more, but with a catch I'll cover below. AMS HT (high-temperature, 2025) The 2025 single-spool unit aimed at engineering filaments. It's not a four-slot box; it's one spool with active heating up to 85°C for drying, a brushless servo feeding motor, and ceramic filament inlets (Bambu lists them around Vickers 1200) so it doesn't chew itself apart on abrasive carbon-fiber or glass-fiber blends. Each AMS HT needs its own 220 V external power brick because it draws too much for the printer to feed it directly. Compatibility rolled out via OTA during 2025: X1 and P1 series got it earlier in the year, A1 and A1 mini got support later in 2025. I'm being deliberately vague on exact firmware numbers because Bambu's release notes shifted twice during the year and I'd rather you check your printer's update page than trust a number I copy-pasted three months ago. AMS 2 Pro The four-slot unit that launched alongside the H2D. It looks like the original AMS but adds active drying (heater plus air circulation built in) and upgraded feeding motors that are quieter and noticeably faster on slot changes. It's compatible with X1C, X1E, P1P, P1S, H2D, H2S, H2C, X2D, P2S, and (via the late-2025 OTA) the A1 and A1 mini. One caveat I want to be honest about: the drying function on the X1E was still listed as "under testing" in Bambu's docs late last year, so if you're on an X1E, verify in your current firmware before assuming drying works. AMS Hub (and the A1-series hub) There are two separate hub SKUs: the standard AMS Hub for X1, P1, and H-series, and the AMS Hub A1 Series for A1 and A1 mini. Either hub lets you chain up to four AMS units onto one printer, which gets you a maximum of 16 logical filament slots. On the X1/P1/H hub you can mix any combination of AMS, AMS 2 Pro, and AMS HT. On the A1-series hub you can also mix AMS, AMS 2 Pro, and AMS HT. What you cannot do on either hub: add an AMS Lite. AMS Lite is firmware-isolated and won't enumerate alongside the other variants. Compatibility matrix (quick reference) UnitSlotsX1 / X1C / X1EP1S / P1PA1 / A1 miniH-seriesActive drying AMS (original)4YesYesNoYesNo (passive desiccant) AMS Lite4NoNoYes (only here)NoNo (open-air) AMS HT1Yes (OTA 2025)Yes (OTA 2025)Yes (OTA 2025)YesYes, up to 85°C AMS 2 Pro4Yes (X1E: verify)YesYes (OTA late 2025)YesYes For the full official picture, the Bambu multi-model compatibility guide is the source of truth and is updated more often than any third-party page. If you're picking which printer + AMS combo to start with, I've written more on the slicer side of each printer in our X1C and P1S guide and the A1 series guide. Connecting your Bambu AMS to OrcaSlicer This is where most of the support threads come from, and honestly, it's not OrcaSlicer's fault. It's the post-2025 Bambu Connect situation. Let me explain what happened and then walk through both pairing paths. The Bambu Connect middleware (yes, it matters) In January 2025, Bambu rolled out a new authorization-and-authentication layer for X1-series firmware. The official blog post (still live at blog.bambulab.com) explains it as a security measure against unauthorized access to printer hardware. The practical effect for OrcaSlicer users is that anything that touches the printer remotely (sending a job, querying AMS slots, controlling cameras) now passes through a piece of middleware called Bambu Connect. The independent OrcaSlicer-BambuLab fork that briefly re-enabled the bypassed features got hit with a legal threat and was shuttered, which Tom's Hardware covered at the time. Bambu's January 2025 firmware authorization announcement. This is the change that reshaped how OrcaSlicer talks to AMS in cloud mode. What this means in practice: in cloud mode, OrcaSlicer can still send prints to your printer through Bambu Connect, and the AMS panel still populates. But some interactions are now read-only or gated. The good news is that LAN-only mode bypasses most of this, at the cost of having to manually re-enter the access code and IP when things change. Path A: Bambu Cloud pairing This is the path most people land on by accident, and it's the easiest for a first-time setup. You sign into your Bambu account inside OrcaSlicer (the top-right account icon), and any printer bound to that account shows up in the Device tab automatically. There's no IP or access code to copy. The setup flow: Open OrcaSlicer, click the account icon, sign in to your Bambu Cloud account. Make sure your printer is online and bound to the same account (you can verify on [truncated] --- ## OrcaSlicer for Bambu Lab P1S / P1P - Complete Setup Guide (2026) URL: https://orcaslicer.net/orcaslicer-bambu-p1s-p1p/ Bambu officially retired the P1P in early 2026 (production end on February 10), but there are still tens of thousands of them in active workshops, mine included. The P1S replaced it as the entry-level Bambu, and the two share more DNA than the marketing suggests. They also share a thorny problem: OrcaSlicer's old LAN-only access code path stopped working sometime in 2026, and most online guides haven't caught up. I've been running both printers through OrcaSlicer for the better part of a year, so here's what actually works in 2026, what's changed, and what to do about it. If you also own an X1C alongside, our X1C and P1S setup guide already covers that overlap. This piece is for the P1S and P1P specifically, with the material capability split that the X1C guide skipped over because both X1C and P1S are enclosed boxes. Table of contents P1S vs P1P at a glance Installing OrcaSlicer profiles for P1S and P1P Network setup: three working paths in 2026 Material capability: what each printer can actually print AMS integration on P1S and P1P First print and calibration OrcaSlicer-specific tweaks vs Bambu Studio Common pain points and fixes When to upgrade: P1P to P1S, or skip to X1C Wrap-up P1S vs P1P at a glance I'll start with the comparison most guides bury at the bottom. The P1S and P1P are the same printer underneath. Same steel CoreXY frame, same 256 mm cube build volume, same all-metal hotend, same 500 mm/s top speed, same 20,000 mm/s² acceleration ceiling. The differences are bolted onto the outside, and they're the differences that decide what you can actually print. Here's the side-by-side I wish I'd had when I bought my P1P: SpecP1SP1P FrameEnclosed (glass top, side panels, front door)Open (snap-on plastic side panels) Build volume256 x 256 x 256 mm256 x 256 x 256 mm Max speed500 mm/s500 mm/s Max accel (extruding)20,000 mm/s²20,000 mm/s² Hotend300 °C all-metal, stainless 0.4 stock300 °C all-metal, stainless 0.4 stock Heated bed100 °C, textured PEI100 °C, textured PEI Aux part-cooling fanYesNo Chamber regulator fanYes (expels heat for PLA)No Carbon air filterYesNo AMS portSame 4-pin connectorSame 4-pin connector Lidar (auto Flow Dynamics)No (X1-only)No (X1-only) NetworkWi-Fi 2.4 GHz onlyWi-Fi 2.4 GHz only Status (May 2026)Current entry-level BambuDiscontinued February 10, 2026; P1S and P2S replaced it That's the headline. The P1S is a P1P with a box around it, plus an aux fan, plus a carbon filter, plus the chamber regulator fan that pushes hot air out so PLA doesn't heat-soak past 60 °C in the chamber. Bambu also still sells the official P1P-to-P1S enclosure kit (around $150) for owners who want to convert without buying a whole new printer. Quick verdict by buyer: Buying new in 2026: get the P1S. The price gap to a used P1P closed once the P1P went out of production, and the enclosure unlocks ABS, ASA, and most CF blends day one. Already own a P1P: keep printing PLA, PETG, and TPU. If ABS becomes a need, the official enclosure kit is cheaper and quieter than DIY panels, though the DIY route (ARC, Vision, Voxelpla) works if you can borrow a friend's enclosed printer to make the parts. ABS or ASA on the shopping list: P1S, not P1P, full stop. I've watched too many P1P-with-cardboard-box experiments end in cracked layers. Multi-color PLA shop: either is fine. The AMS works identically on both. Installing OrcaSlicer profiles for P1S and P1P Both printers are first-class citizens in OrcaSlicer's BBL profile bundle. I checked the upstream resources/profiles/BBL/machine/ folder before writing this and confirmed the current set: Bambu Lab P1S.json (parent) plus 0.2, 0.4, 0.6, and 0.8 nozzle variants Bambu Lab P1P.json (parent) plus the same four nozzle variants Here's how to install them on a fresh OrcaSlicer setup: Launch OrcaSlicer. On a first run the Configuration Wizard pops up automatically. Otherwise hit Configuration → Configuration Wizard from the top menu. Tick Bambu Lab in the vendor list. In the model picker, tick Bambu Lab P1S, Bambu Lab P1P, or both. They coexist without conflict, which is handy if you run both like I do. Pick the nozzle sizes you actually own. The 0.4 mm is the default; if you've got a 0.2 or 0.6 hotend kit, tick those too. They'll install as separate machine profiles in the dropdown. On the filament step, keep the Bambu Basic and Generic PLA/PETG presets. Skip the exotics until you actually load them. Finish the wizard. You'll see entries like "Bambu Lab P1S 0.4 nozzle" and "Bambu Lab P1P 0.4 nozzle" in the printer dropdown at the top right of the Prepare view. An OrcaSlicer Prepare tab with a Bambu profile loaded; the workflow is identical whether you're on P1S, P1P, or A1 mini. Switch printers from the top-right dropdown and the plate, kinematics, and start G-code update automatically. One detail worth knowing if you ever pop the JSONs open. The P1S profile inherits its filament tunes from the X1C set (the @BBL X1C suffix). The P1P has its own dedicated @BBL P1P filament set, with slightly slower max volumetric speeds and lower bed temps for ABS to acknowledge that the open frame will fight you on warping. That's why a generic Bambu PLA Basic on the P1P will print slightly differently than the same spool on a P1S, even with no other settings touched. It's not a bug. It's the team being honest about what an open-frame printer can sustain. The kinematics line up exactly between the two on the 0.4 mm variant: 20,000 mm/s² acceleration on extruding moves, 9,000 on travel, 500 mm/s top speed, jerk capped at 9, retraction 0.8 mm at 30 mm/s in and out. That's identical Bambu firmware-side, and OrcaSlicer mirrors it. Network setup: three working paths in 2026 This is the part everyone's googling, and it's where most older guides are flat-out wrong now. The short version: OrcaSlicer cannot bind directly to your P1S or P1P using just the LAN-Only Access Code anymore. Bambu's January 2025 firmware (01.08.02.00 on the P1 series, with later patches into 2026) introduced an "authorization control system" that broke the simple direct connection that OrcaSlicer used for years. There are three paths that actually work today. Pick based on how you feel about Bambu accounts and Developer Mode warnings. PathAccount required?Trade-off LAN + Developer ModeNoManual toggle on the printer; warranty/support nuance Bambu Connect helperYes (Bambu account)Telemetry; account-tied; smoothest experience MicroSD sneakernetNoNo remote control; immune to firmware lockouts Path 1: LAN + Developer Mode This is the path I use, because I don't want a cloud account in the loop and I'd rather toggle a setting than install a helper app. On the printer's knob LCD, go to Settings → General and enable LAN Only Mode. Scroll down in Settings. Developer Mode only appears once LAN Only Mode is on. Enable it and read the warning prompt. Open Settings → WLAN and note the printer's IP address and the eight-character Access Code shown there. In OrcaSlicer, switch to the Device tab, click Add Printer, and enter the IP and Access Code. The chamber camera (port 6000), AMS panel, status, and file transfer all light up. Print initiation works. Path 2: Bambu Connect helper Bambu Connect is the official helper app Bambu introduced in 2025 to bridge OrcaSlicer (and other third-party slicers) to authorized printers. It needs a Bambu account but skips the Developer Mode toggle. Bambu Connect is Bambu's official answer to "I want third-party slicers to keep working." It's a small helper app you install alongside OrcaSlicer that brokers print sends through Bambu's authorization layer. Sign in once with your Bambu account, OrcaSlicer detects it via the network plugin, and Send to Printer routes through it. No Developer Mode, no manual IP entry, and it survives firmware updates that break the direct path. Trade-off: account-tied, some telemetry, and you're trusting Bambu to keep the helper alive. For someone who wants to print without thinking about it, this is the smoo [truncated] --- ## OrcaSlicer with Bambu AMS, AMS 2 Pro & AMS Lite: 2026 Setup Guide URL: https://orcaslicer.net/orcaslicer-bambu-ams/ I bought my X1C with the AMS expecting OrcaSlicer support to be a one-click toggle. The first time I sliced a 4-color print, OrcaSlicer didn't see the AMS at all, and the LAN-only mode access code I'd ignored for months turned out to be the whole problem. That was a couple of firmware revisions ago. Since then, Bambu has rolled out Authorization Control, launched the AMS 2 Pro alongside the H2D, pushed the AMS HT to nearly every printer in the lineup via OTA, and OrcaSlicer has had to chase all of it. If you landed here in 2026 trying to figure out why your AMS works in Bambu Studio but not in OrcaSlicer, or whether the AMS Lite on your A1 will let you run a 4-color print without melting your TPU, this is the guide I wish I'd had. I'll be honest about where Bambu Studio still wins, where OrcaSlicer is the better tool, and where both are still actively buggy. Nothing here is marketing. If you want a broader view across every printer brand, our wider multi-color guide is the pillar that ties this all together. Table of contents Which AMS do you have, and what does OrcaSlicer do with it AMS hardware comparison table Setting up the AMS itself (hardware first) Pairing OrcaSlicer to your Bambu printer Your first multi-color print Flush volumes and purge towers RFID, filament profiles, and generic spools Multi-color workflow tricks Common AMS plus OrcaSlicer problems and fixes AMS for non-Bambu printers AMS 2 Pro and AMS HT in OrcaSlicer specifically When Bambu Studio still wins FAQ Which AMS do you have, and what does OrcaSlicer do with it Bambu sells four units that all get called "the AMS" in casual conversation, and they're not interchangeable. The original AMS is the boxy 4-slot enclosure that sits on top of an X1 or P1. The AMS Lite is the open-frame 4-arm spool holder that ships with A1 and A1 Mini bundles. The AMS 2 Pro launched alongside the H2D and added active drying. The AMS HT is the single-spool drying-focused unit that's now compatible with most of the lineup via OTA updates rolled out across 2025. OrcaSlicer treats all four roughly the same way at the slicer level. You bind the printer, the slicer reads the AMS state over MQTT, and each slot shows up as a tray with a color, a filament type, and a remaining percentage. Color mapping, RFID readout, and slot-aware slicing all work. What differs by hardware isn't the slicing, it's what filaments each AMS can physically push without jamming. The big asterisk is Authorization Control. Recent Bambu firmware versions introduced a stricter handshake between the printer and any third-party software trying to send a print job. OrcaSlicer can still slice, map, and prepare your AMS print perfectly, but the actual "send to printer" call may need to hand off through Bambu Connect or you can run the printer in LAN-only Developer Mode to bypass the check. I'll walk through both paths in the pairing section. Quick orientation before we dig in: if you have an X1C, X1E, P1S, P1P, or P2S, you almost certainly have the original AMS or an AMS 2 Pro. If you have an A1 or A1 Mini, you have an AMS Lite. If you have an H2D, you started with an AMS 2 Pro and may have added AMS HT units for engineering filaments. Each of these has its own quirks in OrcaSlicer, and we'll cover them in turn. AMS hardware comparison table This table is the cheat sheet I keep open when someone in our Discord asks which AMS they should buy for what printer. Spec sheets change with every Bambu refresh, so verify pricing and OTA compatibility on the Bambu compare page before you put money down. The functional rows below have been stable through 2025 and into early 2026. SpecOriginal AMSAMS LiteAMS 2 ProAMS HT Slot count4 per unit4 per unit4 per unit1 per unit (single spool) Compatible printersX1C, X1E, P1P, P1S, P2SA1, A1 Mini onlyX1, X1E, P1P, P1S, P2S, A1, A1 Mini, H2D (A1 series via OTA)Most current Bambu printers (X1, P1 added in 2025; A1 added later in 2025 via OTA) Max units chained4 (16 spools) via AMS Hub1 (no chaining)Multi-AMS chaining supported with mixed AMS HT unitsChains in mixed setups with AMS 2 Pro PLA, PETGYesYesYesYes ABS, ASAYesCompatible but warned against (open frame, no enclosure)YesYes TPUOnly "Bambu TPU for AMS" (stiffer formulation), no generic flexible TPUSame restriction; FAQ recommends external spool"TPU for AMS" onlySingle spool path tolerates flexible TPU more freely PA, PC, carbon fiberHardened nozzle required for CF blendsNot recommended (abrasive plus open environment)Yes, full supportDesigned for engineering filaments and drying RFID auto-detect (Bambu spools)YesYesYes (also stores drying timestamps)Yes Drying capabilityNone (passive desiccant only)None (open air)Active drying, sealed chamberActive drying, higher temperature ceiling and stronger heater Sealed and humidity controlledSealed lid plus desiccant pouchesNo (open frame)Sealed, humidity sensorSealed, humidity sensor OrcaSlicer support level (2.3.1, late 2025)Full multi-color, RFID, mapping. Print initiation may route through Bambu Connect on auth firmware.Full multi-color, RFID, mapping. Same auth caveat.Full support added in 2.3.1.Full support added in 2.3.1; intermittent P1S print-stall reported in Issue #11498. Two things to take away from this table. First, if you own an A1 or A1 Mini, your AMS Lite is the only Bambu AMS rated to push generic flexible TPU at all, and even then via the external spool path rather than through the Lite itself. Second, the 24-channel mixed setup people quote on YouTube is achieved by chaining multiple AMS 2 Pro units with AMS HT units on the H2D. That's a lot of money and a lot of desk real estate. For 95 percent of users, one AMS or one AMS 2 Pro is enough. If you're still deciding between printer plus AMS combos, our X1 Carbon and P1S setup guide covers the original-AMS workflow in depth, and the A1 and A1 Mini guide covers the AMS Lite specifics. Setting up the AMS itself (hardware first) I'm going to assume you've already gotten your printer assembled and printing single-color. The AMS install is its own little project, and skipping the physical setup steps is the most common reason "OrcaSlicer can't see my AMS" later turns out to be a mechanical problem. Mounting. On an X1C, X1E, P1S, P1P, or P2S, the AMS or AMS 2 Pro sits on top of the printer or on a shelf next to it, with its bowden output routed into the back of the printer's filament-in port. On A1 and A1 Mini, the AMS Lite mounts on the printer's left side using its dedicated 4-arm bracket and uses 4 short PTFE tubes, one per slot, no buffer. PTFE routing. This is the single biggest cause of "slot 3 keeps jamming" complaints. Slots 3 and 4 use longer tubes than slots 1 and 2, so they have more bend and more resistance. Keep the AMS within roughly 600 mm of the printer's back panel, route tubes with no kinks, and push the AMS as close to the printer as you can spare the desk space for. Failure rates on slot 3/4 swaps drop to near zero with clean routing. Loading your first spool. Open the lid, drop the spool in, cut the leader at a 45-degree angle, and push it into the slot intake until the auto-load motor grabs it. For Bambu spools, the RFID tag is read automatically and the AMS knows the brand, type, and color within a few seconds. For generic spools, mark the type and color manually, either on the printer screen or in OrcaSlicer's Device tab. Desiccant and drying. The original AMS has a desiccant slot at the bottom with orange beads that turn dark green when saturated. Replace or oven-recharge them if you're somewhere humid. AMS 2 Pro and AMS HT use active drying instead. AMS Lite has nothing by design, so overnight moisture absorption on the open arms is just part of the deal. Pairing OrcaSlicer to your Bambu printer This is the section where I get the most questions. The pairing flow has changed twice since OrcaSlicer 2.2 because Bambu kept moving the goalposts on third-party authentication. Here's where it stands as of early 2026. LAN-only mode walk [truncated] --- ## OrcaSlicer with Bambu Cloud and LAN-Only Mode: A Complete 2026 Setup URL: https://orcaslicer.net/orcaslicer-bambu-cloud-lan/ Last February my P1S stopped accepting OrcaSlicer prints in the middle of a perfectly normal Tuesday. I'd just let the printer pull down firmware 1.7.x the night before, hit Print on a sliced plate the next morning, and watched OrcaSlicer's Device tab spit back an authorization error I hadn't seen before. Ten minutes of re-logging into Bambu Cloud, restarting the slicer, and re-binding the printer got me nowhere. Then I flipped LAN-Only Mode on, toggled Developer Mode underneath it, pasted in the access code, and the print started immediately. That's the moment I stopped fighting cloud auth on this machine. If you've landed here, you're probably staring at the same wall. The honest 2026 answer is that LAN-Only with Developer Mode is the only fully reliable OrcaSlicer-to-Bambu path on stock firmware right now, and cloud printing through OrcaSlicer either bounces you through Bambu Connect or only works if you've stayed on pre-auth firmware. This guide walks both paths end to end, with the per-printer landmines and the actual error codes you'll see when something breaks. What's in this guide The 2026 reality of OrcaSlicer plus Bambu Cloud vs LAN-Only at a glance Why Bambu locked things down Developer Mode and what it unlocks LAN-Only setup on the printer side LAN-Only setup on the OrcaSlicer side Cloud printing through OrcaSlicer in 2026 Per-printer notes Common errors and the firewall fix Bambu Studio vs OrcaSlicer for Bambu owners After connection works, what to calibrate Frequently asked questions The 2026 reality of OrcaSlicer plus Bambu Here's the short version of what changed. On January 16, 2025, Bambu shipped firmware called Authorization Control. The X1 series got it first at version 01.08.03, and the P1 and A1 lines picked it up over the rest of 2025. The wrapper sits in front of every privileged action: starting a print, jogging the toolhead, binding a printer, pushing AMS commands, firmware upgrades, remote video. Anything that touches motion or media now needs a Bambu-signed software signature. OrcaSlicer doesn't have that signature, and that's not an oversight. Bambu opened PR #8103 to wire in a Bambu Connect handoff. SoftFever marked it draft within three days, and on January 26, 2025, the public position landed: OrcaSlicer would not ship Bambu Connect support. A few weeks later Pawel Jarczak's OrcaSlicer-BambuLab fork got hit with a cease-and-desist and went dark. By early February 2025, Noisyfox's PR #8256 had merged native LAN support that didn't depend on Bambu's network plugin at all. That's the lineage running today's LAN-Only path. So when you read "OrcaSlicer works with Bambu," it means one of three things now. LAN-Only plus Developer Mode on stock firmware (what most people run). Cloud printing on a printer still on pre-auth firmware (rare, and you can't roll back once you upgrade). Or X1Plus on an X1C, which voids warranty. Each is a different deal with the device. Cloud vs LAN-Only at a glance Before the walkthrough, this table is the thing I wish someone had handed me before I started. It's the actual capability matrix on stock 2026 firmware, not the marketing version. Capability Bambu Cloud LAN-Only Mode Send print directly from OrcaSlicerHands off through Bambu Connect on FW 01.08.03+Works directly with Developer Mode on Bambu Handy mobile appYesNo, Handy is cloud-only MakerWorld auto-syncYesManual download to .3mf, slice locally Live camera in OrcaSlicerRemoved for third parties on auth-control FWYes if "LAN Only Liveview" is on Lidar first-layer scan (X1C/X1E/H2D)YesYes, lidar is local hardware Remote print from outside your LANYes via Bambu CloudNo, unless you VPN or run Obico AMS slot mapping in OrcaSlicerYesYes, AMS data syncs over MQTT in both modes Print history across devicesCloud-storedLocal only, lost on reinstall Files stay on your networkNo, routed through Bambu's cloudYes Works with no internetNoYes Works on locked-down school or office networksOften blockedWorks without WAN Voids warrantyNoNo (LAN-Only is sanctioned, X1Plus is the warranty-killer, not LAN-Only) The decision usually shakes out by what you actually need. If you live in Bambu Handy, ride MakerWorld every weekend, and your printer's in a friendly home network, cloud is fine and the Bambu Connect detour is one extra confirmation click. If you're in a workshop with patchy internet, on a school or hospital network where IT blocks outbound MQTT, or you read the August 2023 cloud-outage story and it stuck with you, LAN-Only is the obvious move. Why Bambu locked things down The backstory matters because it tells you why Developer Mode exists at all. On August 15, 2023, Bambu's cloud had a partial outage. During recovery, a cloud-side issue caused some idle printers to start queued jobs without anyone hitting Print. Most ran into already-printed parts and made a mess. A few people came home to molten plastic on top of finished prints. The Bambu blog post afterward was honest about the failure mode, and the public reaction was the first real spike in LAN-Only adoption. The story moved from "concerned hobbyists" to "policy" when third-party slicers started shipping deeper integrations. By late 2024, OrcaSlicer was driving Bambu hardware almost as fluidly as Bambu Studio itself, which made any future cloud incident a bigger surface. Authorization Control was the response. The community immediately filed OrcaSlicer issue #8063 ("FW 1.08.03.00 from Bambu WILL BREAK ORCASLICER for X, P and A series"). Developer Mode is the part of the policy a lot of coverage skipped. Bambu's own wiki documents it as the carve-out: if you're on LAN-Only and you toggle Developer Mode on, the printer keeps the MQTT and FTPS channels open without auth wrapping. That's not a leak, it's the design. Your printer, your local network, your control. The trade-off is that you have to stay local. Developer Mode and what it unlocks The thing nobody tells you on day one is that LAN-Only by itself isn't enough on auth-control firmware. You also need Developer Mode toggled on. Without it, OrcaSlicer can see the printer, pull MQTT status, even browse the FTPS share, but the moment you click Print, the auth wrapper rejects the unsigned command and you watch a spinner until it times out. With Developer Mode on, print initiation works because the printer stops requiring the signed-software check on the LAN side. Camera streaming is a separate toggle called "LAN Only Liveview" in the same Settings group. Turn it on if you want live view in OrcaSlicer's Device tab. If you skip it, the camera tile stays blank but everything else works. Worth saying out loud: Developer Mode doesn't mean "developer-only." Bambu's wiki labels it that way because it was originally for power users running scripts, but in 2026 it's just the standard switch every LAN-Only OrcaSlicer user flips. No warranty implication, no firmware modification. LAN-Only setup on the printer side This part is the same on every modern Bambu (A1, A1 Mini, P1S, P1P, X1C, X1E, with H2D being its own situation we'll cover later). The menu wording shifts slightly between firmware revisions, but the path is consistent. The home screen on a P1/A1 touchscreen. Tap the Setting tile to start the LAN-Only flow. From the home screen, tap the Setting tile. On X1 hardware the same thing lives behind the gear icon at the bottom right. You'll land on a multi-page settings screen that splits into Account, WLAN, SD Card, Device, and a few others. The Settings landing page. WLAN is where you'll grab the IP. The LAN-Only toggle itself sits under the General/Network area depending on firmware. Find the LAN Only Mode toggle. On P1/A1 it's under General. On X1 it's on the third Settings page. Flip it on. The printer will warn you that turning this on disables Bambu Studio and Bambu Handy connectivity, which is exactly what you want. The commitment moment. Yes here means cloud is off and OrcaSlicer becomes the primary driver. The wording mentions Bambu Studio bec [truncated] --- ## OrcaSlicer for Bambu Lab A1 & A1 Mini: Setup, AMS Lite, Calibration URL: https://orcaslicer.net/orcaslicer-bambu-a1-mini/ I bought a Bambu A1 in 2024 because I wanted a CoreXY-ish bedslinger that didn't cost X1C money. Then the January 2025 firmware lockdown happened. Then I read three different blog posts that contradicted each other on whether OrcaSlicer would still work on the A1, and one of them claimed the A1 Mini had a 4.3-inch touchscreen which it does not. So I wrote my own. This is the OrcaSlicer setup guide for Bambu Lab A1 and A1 Mini owners in 2026, post-firmware-lockdown. It covers the LAN-Mode-plus-Developer-Mode walkthrough (different menu paths from the X1 series and P1 series), how AMS Lite mapping works in OrcaSlicer (different from regular AMS), the eddy-current Flow Dynamics calibration that's specific to A1, and the single biggest "Orca won't connect to my A1" mistake (it's not what you think). Written against OrcaSlicer v2.3.2, current stable. Per-printer spec reference SpecBambu A1Bambu A1 Mini Release2024 (Jan/Feb)2023 Q4 Build volume256 × 256 × 256 mm180 × 180 × 180 mm MotionBedslinger CoreXY-style with linear rails on XSame architecture, smaller frame Max toolhead speed500 mm/s500 mm/s Max acceleration10,000 mm/s²10,000 mm/s² Max nozzle temp300°C300°C Max bed temp100°C (textured PEI rated)80°C (lower-power bed) Stock nozzle0.4mm stainless steel quick-swap0.4mm stainless steel quick-swap Optional nozzles0.2 SS, 0.4 / 0.6 / 0.8 hardened steelSame range ToolheadEBB01 (eddy-current force sensor on filament path)Same EBB01, physically interchangeable AMSAMS Lite only (4 open spool slots, rotary RFID)AMS Lite only LidarNoneNone Auto-cal sensorsEddy current (Flow Dynamics) + load-cell-style nozzle Z-probeSame Wi-Fi2.4 GHz only (ESP32, internal antenna)2.4 GHz only EthernetNoneNone Touchscreen4.3" capacitive color2.4" capacitive color Default plateBambu Textured PEIBambu Textured PEI EnclosureOpen frameOpen frame Power1300W150W (low-power bed) Two important corrections I see misreported elsewhere: The A1 Mini has a 2.4-inch screen, NOT 4.3-inch. Bambu's own spec sheet confirms this. Many third-party sites copy each other and get this wrong. The 4.3-inch touchscreen is on the full A1 only. The A1 series uses AMS Lite ONLY. Original Gen-1 AMS (the enclosed unit on X1 / P1) does not work on A1 hardware. AMS Lite has 4 open slots and rotary RFID; it's a fundamentally different mechanical design. The firmware lockdown context (read this first) On January 17, 2025, Bambu pushed firmware introducing the Authorization Control System (ACS). This affects all current Bambu printers including A1 / A1 Mini. The ACS requires third-party tools to route through Bambu Connect (proprietary middleware) for direct print initiation, AMS RFID auto-detect, and cloud features. OrcaSlicer's maintainers declined to integrate Bambu Connect (it would mean shipping closed code inside an AGPL project). The community workaround that's stable and supported on A1 / A1 Mini: enable LAN Only Mode plus Developer Mode on the printer touchscreen. This keeps OrcaSlicer talking to the printer over LAN. Cloud printing through Orca on A1 is gone. AMS RFID auto-detect is gone (you assign filament types manually now). The independent OrcaSlicer-BambuLab fork by Pawel Jarczak briefly restored cloud printing; Bambu issued a cease-and-desist in early 2026 and the fork is shuttered. Don't try to install copies floating online. The 2026 practical answer: A1 + OrcaSlicer = LAN Mode + Developer Mode. Or stay on firmware 1.07.x to keep older direct-print behaviour. Enabling LAN Mode + Developer Mode on A1 / A1 Mini The A1's touchscreen is 4.3-inch capacitive; the A1 Mini's is 2.4-inch capacitive. Menu structure is identical, only the touch targets are smaller on the Mini. Tap the Settings gear (top right corner) Scroll to WLAN (NOT "Network", since A1 series uses "WLAN" because there's no Ethernet option). Tap. Scroll past the SSID list to the toggle row labelled LAN Only Mode. Default OFF. Tap to switch ON. Confirm the warning ("printer will disconnect from Bambu Cloud") The screen now shows an Access Code (8 digits) and the printer's IP address. Photograph or write these down. Scroll further down in the same WLAN screen. A new entry Developer Mode appears (only after LAN Only is on) Tap Developer Mode. Read the multi-screen warning. Scroll to the bottom. Tap Enable. This is different from sibling printers. The X1C uses a 5" capacitive screen with a "Network" tile. The P1S uses a 2.7" mono LCD with a click-wheel and the toggles live under "General" not under "WLAN". Only the A1 series consolidates everything under one WLAN tile. Connecting OrcaSlicer to A1 / A1 Mini Once LAN + Dev Mode is on: Open OrcaSlicer. Switch to the Device tab. Click Add Printer (or the + icon) Enter the IP address (e.g., 192.168.1.62) and the 8-digit Access Code Pick the model: Bambu Lab A1 or Bambu Lab A1 mini Click Bind / Connect If the connection succeeds, the Device tab shows your printer's name, current temperatures, and a thumbnail of the AMS Lite slots if attached. Common A1 connection errors and fixes Printer not visible after enabling LAN-Only Mode (especially A1 Mini): Issue #8188. Power-cycle the printer after the toggle. Confirm 2.4 GHz Wi-Fi (no 5 GHz, no Ethernet). Ensure router AP isolation or guest-VLAN segmentation is OFF; mDNS broadcast doesn't traverse VLANs. "Failed to upload print file to FTP" error -4020: almost always a stale connection from another client. Bambu Handy or Home Assistant integration is hogging the single MQTT/FTP slot. The A1 firmware enforces ONE active client connection at a time. Disconnect Home Assistant, Bambu Handy, OctoEverywhere first, then retry from OrcaSlicer. "Cannot find printer" after firmware update: re-enable LAN Only + Developer Mode (toggles get reset by some updates). Re-enter Access Code in OrcaSlicer. If it still fails, reinstall the Bambu Network Plugin via OrcaSlicer Preferences → Network plugin → Reinstall. AMS Lite handling in OrcaSlicer (different from regular AMS) AMS Lite is a rotary 4-spool open carousel. It's NOT enclosed (no humidity control). Same RFID protocol as the original AMS, so OrcaSlicer auto-syncs Bambu-branded filaments via the green "Sync with AMS" button on the Filament panel. The A1 Device tab with AMS Lite slots visible (A1 through A4 labelled with current filament types). The Preferences dialog shows the Auto-refill and Bed leveling toggles that control print initiation behaviour. Important AMS Lite differences from regular AMS: Hard-capped at 4 slots. Multi-AMS daisy-chaining (which X1C / P1S can do up to 16 colors with 4 AMS units) is NOT supported on A1 series. No drying chamber. The Lite is open. Warn yourself: PETG and TPU left in AMS Lite for 2-3 days will start absorbing humidity. Use an external dryer if you care about print quality on those materials. External spool ("Ext"). The A1 Mini's right-side filament holder is a 5th source. Use this slot only for materials AMS Lite can't handle (TPU 95A, soft TPU, abrasive CF that you don't want grinding the rotary mechanism). Slot mapping workflow (post-firmware-lockdown) RFID is no longer auto-pushed in Dev Mode. Manual workflow: In OrcaSlicer's Device tab, confirm the A1 shows the AMS Lite panel with 4 slots (A1, A2, A3, A4) In the Filament panel (left rail in Prepare tab), click each slot icon and select the filament preset (e.g., "Bambu PLA Basic @BBL A1M") Set color manually. Click the swatch, enter HEX or use the eyedropper For multi-color prints, use the + button to add filaments up to 4 (no chaining on A1) Send-time AMS Mapping dialog: top row is sliced project filaments, bottom row is AMS slots. Reorder if material/color mismatch. Click "Use AMS" then Send. The Send print job dialog with AMS Lite slot mapping. The PETG A3 chip shows the assigned slot for this print. The Bed leveling and Flow Dynamics Calibration toggles let you enable or skip the per-print auto-routines. Calibration sequence: eddy-current changes everything This is where A1 / A1 Mini differ most from other Bambu pr [truncated] --- ## OrcaSlicer for Bambu Lab X1C, X1E and P1S: Setup, Profiles & Calibration URL: https://orcaslicer.net/orcaslicer-bambu-x1c-p1s/ I bought a Bambu Lab P1S in 2024 and an X1C six months later. Both shipped with Bambu Studio on the SD card. For a year I ran Bambu Studio because the AMS just worked and the cloud printing felt magical. Then I started caring about flow rate calibration, scarf seams, and being able to print from a Voron alongside my Bambus. So I switched the fleet to OrcaSlicer. This guide is what I wish I'd had when making that switch. It covers the X1C, X1E, and P1S specifically, with all the LAN-Mode-plus-Developer-Mode walkthrough that nobody warned me about, the AMS configuration changes since the January 2025 firmware lockdown, and the recommended calibration sequence for each printer. Written against OrcaSlicer v2.3.2, current stable. The article focuses on the X1 series and P1S because those are the printers most likely to drive an OrcaSlicer switch. The A1 and A1 Mini share the AMS Lite ecosystem and similar lockdown but use different printer profiles in OrcaSlicer; quick mention at the end. The P1P is referenced briefly since it's open-frame and only useful for low-temp filaments. Per-printer spec reference The X1C, X1E, and P1S share the same 256mm cubed build volume but differ on key features: SpecX1CX1EP1S Build volume256×256×256256×256×256256×256×256 Default plateTextured PEI / CoolEngineeringTextured PEI Hotend max temp300°C320°C300°C Heated chamberNo (passive enclosure)Yes, active 60°CNo (passive enclosure) Stock nozzleHardened steel 0.4Hardened 0.4Stainless 0.4 AMS supportYesYesYes Lidar (auto Flow Dynamics)YesYesNo Touchscreen5" capacitive5" capacitive2.7" mono LCD + knob EthernetWi-Fi onlyWi-Fi + Gigabit EthernetWi-Fi only Air filterActivated carbonG3 + H12 HEPAActivated carbon Camera port (LAN)3223226000 The X1E is the enterprise variant. The active heated chamber is what unlocks PPA-CF and PPS-CF printing; X1C and P1S can't reach the chamber temps those filaments need. The X1E also adds Gigabit Ethernet with WPA2-Enterprise (cert upload via SD card) and the H12 HEPA filter rating that makes it factory-floor-friendly. The P1S is the budget X1C analog. Same enclosed CoreXY, same AMS support, but no lidar, no auto Flow Dynamics, mono LCD instead of capacitive. The lack of lidar is the single biggest reason P1S owners switch to OrcaSlicer; you're on a manual calibration path either way, and OrcaSlicer's manual suite is dramatically deeper than Bambu Studio's. The big context: the January 2025 firmware lockdown Before any setup, you need to understand what changed. On January 16, 2025, Bambu Lab pushed firmware 1.08.02.00 (P-series) and 1.05.00.00 (A-series) introducing what they call the Authorization Control System. The firmware now requires third-party tools to route through Bambu Connect (a proprietary middleware) for direct print initiation, AMS RFID auto-detection, and cloud features. OrcaSlicer's maintainers publicly declined to integrate Bambu Connect (it would require shipping closed code inside an AGPL project). The community workaround that's stable and supported: enable LAN Only Mode plus Developer Mode on the printer touchscreen, which keeps OrcaSlicer talking to the printer over LAN. Cloud printing through Orca is gone. AMS RFID auto-detect is gone (you assign filament types manually now). An independent fork called OrcaSlicer-BambuLab by developer Paweł Jarczak briefly restored cloud printing by replicating Bambu's authorization handshake. Bambu issued a cease-and-desist in early May 2026; the fork is shuttered. Don't try to install copies floating online; they're unmaintained and may be modified in unsafe ways. So in 2026, the practical answer for Bambu plus OrcaSlicer is: LAN Mode + Developer Mode. Or stay on firmware 1.07.x to keep the older direct-print behavior (downgrade is possible on consumer X1C and P1S via SD card; X1E enterprise units are signed-locked tighter). Enabling LAN Mode + Developer Mode (verified menu paths) The toggle paths differ between the X1 series 5" capacitive touchscreen and the P1 series 2.7" mono LCD with click-wheel. X1C and X1E (5-inch capacitive) Tap the gear / settings icon in the top-right of the home screen Scroll the left-side rail to WLAN (or Network on X1E, which adds a separate Ethernet sub-page) Toggle LAN Only Mode ON. A confirmation dialog warns that cloud features (Bambu Handy, Studio cloud project sync, MakerWorld direct) will be disabled. Confirm. Scroll down in the same WLAN screen. A new entry appears: Developer Mode. (It only appears after LAN Only is ON.) Tap Developer Mode. Read the warning page. Scroll to the bottom (the Enable button is greyed until you reach the bottom). Tap Enable. Back out to General → Device Info / Network tab. Write down the IP address (e.g., 192.168.1.41) and the 8-digit Access Code. X1E with Ethernet: plug in the cable, set the WLAN kill switch OFF and the LAN kill switch ON via the physical toggles next to the rear ports. WPA2-Enterprise certs upload via SD card. P1S and P1P (2.7-inch mono LCD with click-wheel) The P1 menu structure differs. There's no nested WLAN page; LAN Only sits under the general Settings list. From home screen, click the wrench / settings icon Scroll down to General Inside General, scroll to LAN Only Mode, click, toggle ON. Confirm warning. Back out to Settings. Scroll down. Developer Mode entry now appears (hidden until LAN Only is enabled). Click. Read the scrolling warning. Scroll to the bottom. Click Enable. Back out to WLAN. The screen now shows IP address and Access Code on a single page. (P1S puts IP/Access Code under WLAN, not under Device info like X1.) The Access Code resets if you regenerate it. If OrcaSlicer ever gets stuck with a stale code (issue #6169 documents this happening), tap or click the Access Code value on the printer to regenerate it, then re-enter in OrcaSlicer. Connecting OrcaSlicer to your Bambu printer Once LAN + Dev Mode is on: Open OrcaSlicer. Switch to the Device tab Click Add Printer (or the + icon) Enter the IP address (e.g., 192.168.1.41) and the 8-digit Access Code Pick the model: Bambu X1C, X1E, or P1S Click Bind / Connect OrcaSlicer's printer-selection wizard lists every Bambu Lab model. Pick your specific machine and the nozzle sizes you own (0.4mm by default; tick others if you've swapped nozzles). If the connection succeeds, the Device tab shows your printer's name, current temperatures, and (for X1C/X1E) the camera feed. AMS slot count and currently-loaded filament type appear in the AMS panel on the left. If it fails, the most common causes are: wrong Access Code, IP changed because of DHCP rotation (set a static lease on your router), or the network plugin needs reinstalling. Open Preferences → Network plugin → Reinstall to fix the third one. What works and what doesn't in LAN + Dev Mode This is the practical capability matrix. As of the firmware in May 2026: Works: Sending sliced 3MF / G-code to printer over LAN Print-job initiation via Bambu Connect helper app (OrcaSlicer hands the file to Bambu Connect, which executes the authorized print start) Reading AMS slot count and currently-loaded slot identity Plate selection (Cool / Engineering / Textured / High-Temp / PEI Plus) for slicing presets Camera live view in LAN (port 322 X1, port 6000 P1) Printer status telemetry (temps, position, fan speeds, AMS slot temps) LED control Firmware updates via SD card Broken: Direct print initiation from Orca without Bambu Connect (the network plugin is signed-only) RFID auto-detect of Bambu spools propagating filament info into Orca's AMS mapping (you mirror the choice manually) Live AMS slot calibration commands from Orca Cloud project sync, MakerWorld push, Bambu Handy notifications Any motion control, nudging, first-layer-tweak from inside OrcaSlicer The growing minority workaround is to pair Orca with OctoEverywhere or Obico for remote functionality, while using stock LAN+Dev for local printing. Bambu Connect installed and running in the system tray is also a common path; Orca → Send hands the file [truncated] --- # Section: vs PrusaSlicer ## OrcaSlicer for Prusa MK3S+ / MK3S / Mini+: Profile Tweaks Guide URL: https://orcaslicer.net/orcaslicer-prusa-mk3s-mini/ I've been running an MK3S+ in the office and a Mini+ in the garage for almost three years, and for most of that time PrusaSlicer was the only slicer either machine ever saw. The switch to OrcaSlicer started the night I dropped a Bambu P1S onto the same network and didn't want to juggle two slicer windows just to keep my Prusas in the loop. If you're in that same boat, this guide walks through which Prusa profiles OrcaSlicer ships, where the firmware boundary sits between the MK3 family and the Mini+, how the start G-code differs, where to drop your Linear Advance K-factor, and how to send files over PrusaLink without breaking what Prusa already does right. Table of contents Why a Prusa MK3S+ or Mini+ owner picks OrcaSlicer Which Prusa profiles ship in OrcaSlicer Build volumes and the firmware boundary Start and end G-code that ships in the Prusa profiles Mesh bed leveling: G80 on MK3S+, G29 on Mini+ Linear Advance on MK3S+ versus Pressure Advance on Mini+ PrusaLink upload workflow from OrcaSlicer Recommended calibration order for a Prusa in OrcaSlicer Common gotchas and profile drift versus PrusaSlicer FAQ Why a Prusa MK3S+ or Mini+ owner picks OrcaSlicer Let's be honest up front: PrusaSlicer is canonical for Prusa printers. Prusa's engineers maintain the profiles, the start G-code, and the firmware. If you only own a Prusa and your prints are fine, there's no quality reason to swap. I still keep PrusaSlicer installed alongside Orca, and you probably will too. What pulled me across is what OrcaSlicer adds on top. The Calibration menu hands you a one-click test suite: Temperature, Volumetric Speed, Pressure Advance, Flow Ratio, Retraction, Tolerance, Cornering, and Input Shaping with VFA. Eight prebuilt calibration entries in the menu bar. In PrusaSlicer I was hand-rolling temperature towers in Notepad. In Orca I pick a range, hit slice, and the tower's done. The other draw is the unified Device tab. OrcaSlicer talks PrusaLink, OctoPrint, Klipper, Obico, and Bambu's native protocol from the same dialog. If your household has a Mini+ in the garage and a Bambu upstairs, Orca becomes the common slicer instead of the third tab you keep open just for one printer. For a side-by-side on where each slicer shines, my OrcaSlicer vs PrusaSlicer comparison goes through the trade-offs. I'm not pitching Orca as "better" for Prusa work. The Prusa profiles inside OrcaSlicer are derived from PrusaSlicer upstream, and there's an open community argument (Issue #6037, closed as not planned) that some Orca defaults drift from Prusa's published values. We'll get to that. For now, treat OrcaSlicer as a second slicer that brings a calibration suite and a unified device manager to the printers you already trust. OrcaSlicer printer selection dialog showing the Prusa vendor section with MK3S and MINI profiles listed Which Prusa profiles ship in OrcaSlicer When you run the Add Printer wizard and expand the Prusa vendor row, OrcaSlicer offers a specific set of machine profiles. For our three printers, here's exactly what ships in the current main branch. Prusa MK3S profiles (five files): a base Prusa MK3S.json template plus dedicated 0.25 mm, 0.4 mm, 0.6 mm, and 0.8 mm nozzle variants. That's the full nozzle ladder, matching PrusaSlicer's coverage. Prusa MINI profiles (five files): the same structure. Base template plus 0.25 mm, 0.4 mm, 0.6 mm, and 0.8 mm nozzle profiles. Prusa MINIIS profiles (five files): the Mini Input Shaper variant gets its own profile family. If your Mini+ is running the IS firmware build, you'll want this one rather than the plain MINI profile. The fact that MINIIS now ships partly resolves Discussion #2491, which originally asked for a dedicated Input Shaper variant. Here's the thing that trips up MK3 (non-S) and MK3S+ owners: there's no separate MK3 profile, and no separate MK3S+ profile. OrcaSlicer ships only Prusa MK3S for the entire MK3 / MK3S / MK3S+ / MK2.5S / MK2.5 cluster. This isn't a bug. PrusaSlicer upstream does the same thing because the MK3S to MK3S+ upgrade is hardware-only (SuperPINDA probe, improved Y-axis hardware, no firmware or profile changes). MK3 and MK2.5 owners pick the MK3S profile too. The printer pops up a one-time prompt that reads roughly "This was made for other printer, do you want to continue?", and you say yes. If you've heavily modded your Prusa and the stock MK3S profile no longer fits, the printer profiles hub covers cloning and customising vendor profiles without breaking the start G-code. Useful trivia: PrusaSlicer labels print presets "SPEED" and "QUALITY". OrcaSlicer relabels them "Speed" and "Standard". Same lineage, different vocabulary, source of some confusion in Issue #6037. Build volumes and the firmware boundary This is the section I wish someone had handed me on day one, because mixing up the firmware lineage leads you down the wrong rabbit hole on every other setting. MK3S+ and MK3S build volume: 250 x 210 x 210 mm. The OrcaSlicer machine JSON confirms it: printable_area: ["0x0", "250x0", "250x210", "0x210"], printable_height: "210". Both printers share that volume because the "+" is a hardware revision. Mini+ build volume: 180 x 180 x 180 mm. The OrcaSlicer Mini JSON has "printable_area": ["0x0", "180x0", "180x180", "0x180"] and "printable_height": "180". There's a Prusa KB FAQ that shows a different number; we'll cover that in the gotchas section. Use 180 cubed. Now the firmware boundary, which matters when you're editing G-code or talking to the printer. MK3S+, MK3S, MK3, MK2.5S, and MK2.5 all run prusa3d/Prusa-Firmware. The repo README describes it as "the source code and the development versions of the firmware running on the Original Prusa i3 MK3S/MK3/MK2.5S/MK2.5 line of printers" and adds that it's "proudly based on Marlin 1.0.x by Scott Lahteine (@thinkyhead) et al." It runs on the 8-bit Einsy board. Call it Original Prusa Firmware. The Mini+ runs prusa3d/Prusa-Firmware-Buddy. A separate Marlin-based codebase, 32-bit, that covers the MINI/MINI+, MK3.5, MK3.9, MK4, XL, and CORE One. Both firmwares share Marlin lineage but they're separate codebases. They don't share binaries, and not every G-code dialect. In practice: when you read a tutorial about MK4 settings, don't assume it applies to your MK3S+. Different firmware. The reverse is also true. The Mini+ sits awkwardly in the middle, sharing a firmware family with the MK4 but a price tier closer to the MK3. Side-by-side build-volume comparison of Prusa MK3S+ (250x210x210 mm) and Prusa Mini+ (180x180x180 mm) Start and end G-code that ships in the Prusa profiles The fastest way to convince yourself OrcaSlicer will behave on your Prusa is to open the Machine G-code tab. Here's the verbatim start block from the Prusa MK3S 0.4 nozzle profile, trimmed for the bits that matter. M862.3 P "MK3S" ; printer model check M862.1 P[nozzle_diameter] ; nozzle diameter check M115 U3.13.0 ; tell printer latest fw version G90 ; use absolute coordinates M83 ; extruder relative mode M104 S[first_layer_temperature] ; set extruder temp M140 S[first_layer_bed_temperature] ; set bed temp M190 S[first_layer_bed_temperature] ; wait for bed temp M109 S[first_layer_temperature] ; wait for extruder temp G28 W ; home all without mesh bed level G80 ; mesh bed leveling That's the Prusa house style. M862.3 P "MK3S" is the printer-model check (the line that produces the "made for other printer" prompt when you slice for an MK3S and print on an MK3). M862.1 verifies nozzle diameter. M115 U3.13.0 compares firmware version; per Prusa KB, "if a provided version (U parameter) exceeds the current firmware, the printer pauses for 30 seconds and requests a firmware upgrade." G28 W homes without running the mesh, then G80 kicks off mesh bed leveling. The MK3S end G-code is just as Prusa-aware: G1 X0 Y200 F3600 ; park M221 S100 ; reset flow M900 K0 ; reset LA M104 S0 ; turn off temperature M140 S0 ; turn off heatbed M107 ; turn off fan M84 ; disable motors The M900 K0 resets the Linear Adva [truncated] --- ## OrcaSlicer vs SuperSlicer vs PrusaSlicer: 3-Way Showdown 2026 URL: https://orcaslicer.net/orcaslicer-vs-superslicer-vs-prusaslicer/ I've run all three of these slicers as my daily driver at different points over the last four years, and I've watched every "which one is best" thread devolve into tribal warfare. So let me say up front: I'm not picking a winner here, because there isn't one. There's only the right tool for the printer sitting on your bench, the materials you actually print, and how much time you want to spend tuning versus printing. I'll save you the suspense if you came here hoping I'd tell you OrcaSlicer crushes the other two. I'm not going to. I genuinely like all three. I think each one has earned its place in my workflow at some point, and I'd argue each one has a list of things it does better than the others. I want to give you the honest comparison I wish someone had handed me back when I first switched off Cura. Table of contents The three-way dilemma in 2026 Lineage: how three slicers descended from one Calibration suite head to head Support strategy comparison Multi-material: AMS vs MMU3 vs the wipe tower rework Printer compatibility and profile ecosystem UI and UX: polish vs power vs depth Profile interchange and switching cost Performance and slice speed Where each slicer wins The verdict, and what the verdict isn't FAQ The three-way dilemma in 2026 I'll lay out the shape of the decision in 2026 as I see it after four years of bouncing between these three. PrusaSlicer is the most polished upstream, maintained by a hardware company that owns the print path from filament to firmware. OrcaSlicer is the feature-rich community fork that started life as a Bambu Studio derivative and now ships the most aggressive built-in calibration tooling of any free slicer. SuperSlicer is the long-running PrusaSlicer fork by a single developer (supermerill) that exposes the most granular per-feature controls but moves at the slowest cadence of the three. I see Bambu owners, Creality K-series owners, and generic Klipper users almost always end up on OrcaSlicer because it ships the profiles and the built-in calibration menu that match those printers out of the box. I see Prusa owners stay on PrusaSlicer because MMU3 integration, scarf seams, multi-bed slicing, and Printables sign-in are all first-party features there. I see the power-user crowd printing weird single-extruder workflows, exotic materials, or just wanting more knobs exposed land on SuperSlicer for a reason. I'm going to walk through this in the order that actually matters: who made what, what each one calibrates, how they handle supports and multi-material, what printers each one supports, and what the UI actually feels like to live with. I'll tell you who should pick what at the end, and I'll be honest about which slicer wins each round on the way there. Lineage: how three slicers descended from one I think you've got to understand the family tree to understand why the comparison shakes out the way it does. Every one of these slicers traces back to Slic3r, the original open-source slicer Alessandro Ranellucci wrote with the RepRap community. PrusaSlicer is a direct fork of Slic3r maintained by Prusa Research (the prusa3d organization on GitHub). SuperSlicer is a fork of PrusaSlicer maintained by supermerill. OrcaSlicer is a fork of Bambu Studio, which is itself a fork of PrusaSlicer. So OrcaSlicer is two forks deep from PrusaSlicer, while SuperSlicer is one fork deep. I want to dwell on this because it matters: PrusaSlicer is the upstream where most genuinely new features land first. Arachne, the variable-width perimeter generator, originated in PrusaSlicer 2.5.0 alongside Lightning infill, the pressure equalizer, and STEP file import. The 2.5.0 release notes from prusa3d are unambiguous: "Arachne perimeter generator, Lightning infill, Pressure equalizer, STEP file import and various smaller features and bugfixes." Organic Supports landed in PrusaSlicer 2.6.0, with the first public alpha in February 2023 and the final release in July 2023. Scarf seams, multiple beds, paint-on fuzzy skin, and Printables integration all arrived in PrusaSlicer 2.9.0 in December 2024. I made the mistake of assuming OrcaSlicer invented all those features back in my first year, and I see other people make the same mistake constantly. They came from Prusa, and OrcaSlicer ported them later. Who maintains each project PrusaSlicer is maintained by Prusa Research, a Czech hardware company with a paid engineering team and a release cadence backed by a printer business. OrcaSlicer is maintained by SoftFever, a Singapore-based developer working with a community contributor base under the AGPL-3.0 license. SuperSlicer is essentially supermerill's project. He went full-time on the slicer per his own statement in discussion #3902: "hi I left my job, now i'm full time on the slicer." That's verifiable, and it's the cleanest way to put to bed the recurring "is SuperSlicer dead" question. It isn't. All three projects are AGPL-3.0 licensed, which means the source is open and any fork has to publish its source too. That's how OrcaSlicer was even possible to build, and it's why this whole ecosystem keeps moving forward. Release cadence and current versions Here's where the differences get sharp. PrusaSlicer ships polished point releases roughly every couple of months. The current stable as of this writing is 2.9.4, released November 7, 2025, which added support for junction deviation for Marlin firmware. 2.9.3 (September 12, 2025) introduced a "Consistent Surface" cooling strategy and short-travel acceleration controls. 2.9.0 in December 2024 was the big-feature release. OrcaSlicer ships the 2.3.x line, with the most recent stable being v2.3.2 released in March of the 2.3.2 cycle (the GitHub tag and some scrapes have conflicted on the year, so I'd just check the live OrcaSlicer releases page on GitHub if the exact date matters to you). v2.3.2's headline changes are configurable wipe tower type selection (the new variant is "Type 2"), an accordion-style main tab UI that collapses to icons when space is tight, a security fix for path traversal in 3MF import, and improved MMU detection via the Moonraker database. SuperSlicer is the slowest of the three. The latest stable per the project docs is 2.5.59.13 from July 1, 2024, with the docs noting "There are no plans for any further development for the 2.5.59.x version." The active line is the 2.7.61.x beta series, with 2.7.61.0 beta released March 24, 2025 and 2.7.61.7 released July 21, 2025 as "the seventh bug fix release for the 2.7.61.X Beta and most likely the last Beta version before 2.7.61.X becomes the new stable branch." So no, SuperSlicer is not abandoned. The cadence is just slower than the other two, and you have to be comfortable running a beta if you want recent features. I should mention one sad note here, since it explains some of the slowdown in 2025. supermerill noted in September 2025 that John Beima, who managed the SuperSlicer website, had passed away. I think that kind of loss in a small project run mostly by one developer plus a handful of volunteers shows up in release pacing, and I'd treat SuperSlicer's cadence with a little context rather than a snap judgment. Three-slicer feature comparison table Calibration suite head to head I'd argue this is where the daylight between these three is widest, and it's the single biggest reason most people end up picking OrcaSlicer for a new printer they bought without a pre-tuned profile. I'll be straight: nobody else ships a calibration menu like OrcaSlicer's. I haven't found a comparable native menu in any other free slicer, not even close. That's a real OrcaSlicer win. I want to flag the framing though, because PrusaSlicer's lack of one isn't an oversight, and SuperSlicer's set is older but still useful. OrcaSlicer's nine-test menu I'll list out OrcaSlicer's Calibration menu, which sits in the main menu bar and ships nine separate tests organized into a top-level group and an Advanced submenu. The top-level entries are Temperature, Volumetri [truncated] --- ## OrcaSlicer for Prusa MK4 / MK4S / Core One: 2026 Setup Guide URL: https://orcaslicer.net/orcaslicer-prusa-mk4/ PrusaSlicer is genuinely excellent for Prusa printers, and the only reason I switched my MK4 over to OrcaSlicer was the calibration test integration. Then I found a half-dozen other reasons to stay. Then I found out about the bgcode thing, which almost sent me back. Here's the honest story of running OrcaSlicer on a Prusa MK4, MK4S, or Core One in 2026, written from someone who's been bouncing between both slicers for the better part of a year. I'll cover the setup, the settings I actually override, the real pain points (network printing on Windows, MMU3 in Orca, the bgcode situation), and the spots where PrusaSlicer is still the smarter call. If you own a Prusa and you've never strayed, this is a useful map. If you already use OrcaSlicer for a Bambu or a Voron and you just bought a Core One, this is the shortcut that saves you a weekend of forum hopping. Table of contents Should you actually switch from PrusaSlicer? What works in OrcaSlicer 2.3.2 with the MK4 family today MK4 vs MK4S vs Core One: which profile should you pick? Setting up OrcaSlicer for your Prusa, step by step Best settings to override on the MK4 MK4S-specific tweaks Core One specific tweaks Multi-color (MMU3) in OrcaSlicer in 2026 Calibration sequence for the MK4 family Common pain points and fixes When PrusaSlicer is still the right call Wrap-up Should you actually switch from PrusaSlicer? Let me be blunt before we get into the setup details. PrusaSlicer is excellent on Prusa hardware. It's the official slicer, it gets Prusament profile updates first, the start G-code is maintained directly by Prusa engineers, and the MMU3 painting workflow is the most polished in the industry. If you only own Prusa printers and you don't care about feature breadth, you don't need OrcaSlicer. That said, here's why I keep Orca installed on every machine I run. The built-in calibration tools are the killer feature. Pressure Advance, flow rate, temperature tower, retraction tower, input shaper frequency sweeps, and junction deviation tests all live one click away in the Calibration menu. PrusaSlicer makes you generate these manually or rely on Prusa's printer-side wizards. The OrcaSlicer fuzzy skin painting tool, the per-line speed control, the modifier mesh painting, and the Color Painting tool are all genuinely better than what ships in PrusaSlicer right now. If you've ever wanted to slow down just the front face of a part because the layer changes are showing, OrcaSlicer makes that trivial. The flip side is real. You're now responsible for syncing Prusament profile updates manually. When Prusa pushes a tweak to start G-code or a new filament profile, PrusaSlicer pulls it automatically. In Orca, you wait for the upstream profile maintainers to merge it, then you update the slicer, then you re-import. The MMU3 workflow in Orca is technically there but it's nowhere near PrusaSlicer's level of polish, and bgcode (the binary G-code format Prusa printers prefer for fast Wi-Fi uploads) is not native in OrcaSlicer at all. You'll need a post-processing script, which I'll cover below. The realistic answer for most people I talk to: run both. Use OrcaSlicer for calibration, single-color jobs, and anything that benefits from its painting tools. Keep PrusaSlicer for MMU3 painted prints and for anything where you want the official, hands-off Prusa experience. We have a deeper PrusaSlicer vs OrcaSlicer comparison if you want to dig into the feature-by-feature breakdown. What works in OrcaSlicer 2.3.2 with the MK4 family today Here's the state of play in May 2026. OrcaSlicer's current stable is 2.3.2. The Prusa MK4, MK4S, MK4S HF (high-flow), and Core One profiles all ship in upstream Orca, sitting in the resources/profiles/Prusa/machine/ directory. Core One landed in 2.3.1 (released October 5, 2025), so anything older than that won't see the printer at all. The Core One+ and Core One L variants are also in the upstream tree, paired with current Prusa Buddy firmware 6.5.3 on the printer side. You don't need a community profile pack to get started. Open the Configuration Wizard, expand Prusa Research, and check the boxes. That's the supported path now, which wasn't true even a year ago. Network printing works through both PrusaLink (local LAN, no internet required) and Prusa Connect (cloud). PrusaLink is rock-solid on every OS I've tried it on. Prusa Connect through Orca works reliably on macOS but has some flakiness on Windows 11 (a "TEST passes but upload fails" pattern that several forum threads document). PrusaLink is the workaround if you hit it. Once paired, the OrcaSlicer Device tab embeds the PrusaLink web UI directly inside the slicer, so you can monitor temps, browse files, and start prints without opening a browser tab. Pressure Advance is handled exactly like in PrusaSlicer, through M572 commands embedded in the start G-code conditional, with smoothing time W=0.04s baked in. Input Shaper is factory-calibrated by the printer firmware (introduced in Buddy 5.0.0 for the MK4, and refined since). OrcaSlicer doesn't need to know anything about either of these, the slicer just sends G-code and the printer applies its own resonance compensation. The bgcode caveat is the one ugly bit. Prusa printers prefer .bgcode (binary, compressed) for faster Wi-Fi uploads, especially on chunky multi-color jobs. OrcaSlicer doesn't export bgcode natively. The workaround is the orca_bgcode post-processing script from the bwees GitHub repo, which we'll add in the setup section. Plain .gcode still prints fine, it just uploads slower. MK4 vs MK4S vs Core One: which profile should you pick? If you've upgraded an MK4 to an MK4S, this matters. The MK4 profile in OrcaSlicer ships with conservative volumetric speed limits matching the original V1 nozzle. The MK4S profile loosens those limits to match the high-flow CHT nozzle. Picking the wrong profile caps your real-world throughput by roughly 50% on PLA, even if everything else looks fine. Here's the spec comparison that drives the choice. Spec Prusa MK4 (V1) Prusa MK4S Prusa CORE One ArchitectureCartesian bedslingerCartesian bedslingerCoreXY, fully enclosed Build volume250 x 210 x 220 mm250 x 210 x 220 mm250 x 220 x 270 mm Hot endNextruder + Prusa Nozzle V1Nextruder + High-flow CHT 0.4Nextruder + High-flow CHT, quick-swap Max nozzle temp290 °C290 °C290 °C (400 °C with HT upgrade) Max bed temp120 °C120 °C120 °C Active chamberNoneNone (optional Enclosure)Yes, up to 55 °C Volumetric flow ceiling (PLA)~16 mm³/s~24 mm³/s~24 mm³/s and up Machine max travel speedup to 200 mm/sup to 300 mm/sup to 600 mm/s Machine max accel (X/Y)4000 mm/s²4000 mm/s²higher (CoreXY) Native firmwareBuddy 6.5.3Buddy 6.5.3Buddy 6.5.3 Input ShaperYes (factory calibrated)Yes (factory calibrated)Yes + phase stepping OrcaSlicer profileYes (pre-2.3.x)Yes + HF variantsYes (since 2.3.1) The decision tree is simple. Original V1 hot end? Pick the Prusa MK4 profile. V2 CHT after the $99 upgrade? Switch to Prusa MK4S. Bought the optional higher-flow CHT-HF nozzle separately? Use Prusa MK4S HF. CoreXY enclosed? Pick Prusa CORE One. The Core One L variant has its own profile if you got the larger build version. One thing I want to flag: OrcaSlicer ships separate JSON files for each nozzle size (0.25, 0.3, 0.4, 0.5, 0.6, 0.8 mm) under each printer family. Pick the size you actually have installed. The defaults in fdm_machine_common_mk4s.json assume a 250 x 210 build area, retraction 0.7 mm at 35 mm/s, machine max accel 4000 mm/s², and a profile-default acceleration of 2500 mm/s². If those numbers look familiar from PrusaSlicer, that's because Orca's profile is a faithful translation. Setting up OrcaSlicer for your Prusa, step by step Step 1: install OrcaSlicer 2.3.2 or newer Grab the latest stable from the OrcaSlicer GitHub releases page. Windows, macOS, and Linux builds are all there. Don't bother with anything older than 2.3.1, you'll miss the Core One prof [truncated] --- ## OrcaSlicer vs PrusaSlicer: Detailed Feature Comparison (2026) URL: https://orcaslicer.net/orcaslicer-vs-prusaslicer/ I switched my workflow from PrusaSlicer to OrcaSlicer about eighteen months ago. I keep PrusaSlicer installed for a Prusa MK4S that lives in the production room and for occasional resin work on an SL1S that's gathering dust. Everything else now opens in Orca by default. The reasons aren't ideological; they're practical, and they don't apply to everyone. This is the honest comparison. Both slicers descend from the same lineage, both are AGPL-3.0, both are excellent. They just optimise for different users. If you've got a Prusa MK4S and you mostly print models from Printables, PrusaSlicer is genuinely the easier path. If you've got a Voron, a K1, or a multi-brand fleet, OrcaSlicer is the obvious pick. The middle cases are where it gets interesting. Written against OrcaSlicer v2.3.2 (current stable, 13.7k GitHub stars) and PrusaSlicer v2.9.4 (released November 2025, around 9k stars). PrusaSlicer v2.9.5 is in beta with a major SLA pipeline refactor. The lineage (it actually matters here) Both slicers share around 70% of their code DNA: Slic3r (Alessandro Ranellucci, 2011) │ └── PrusaSlicer (Prusa Research, 2016 fork) │ └── Bambu Studio (Bambu Lab, 2022 fork) │ └── OrcaSlicer (SoftFever, December 2022 fork) The interesting twist is that feature flow has reversed. Originally PrusaSlicer was the upstream and Orca was the downstream consumer of its features. By 2026 that's flipped: PrusaSlicer 2.9 explicitly back-ported scarf seams from OrcaSlicer (the release notes credit the Orca devs by name), added "multiple beds in one project" (an Orca-derived idea), and is rumoured to be back-porting more in the 3.0 refactor. So the lineage matters because it explains why so much of each slicer feels familiar. Both share the same project-based UI, the same plate-based workflow, the same parameter philosophy. They've diverged on UI density, calibration depth, and printer-vendor breadth. Side-by-side feature matrix Twenty-eight rows. The decisive ones are flagged. FeatureOrcaSlicer 2.3.2PrusaSlicer 2.9.4 Latest stable2.3.2 (Mar 2026)2.9.4 (Nov 2025) GitHub stars13.7k~9k LicenseAGPL-3.0AGPL-3.0 Built-in calibration tests9 (Temp, Flow Ratio, PA, Retraction, Max Vol, Cornering, Input Shaping, VFA, Tolerance)2 (Temperature Tower, Pressure Equalizer) Pressure Advance / K-factor calibrationBuilt-in tower (DDE + Bowden)None built-in; use community towers Pressure Equalizer (software speed-smoother)NoYes (PrusaSlicer-exclusive) Input Shaping calibration printYesNo (uses Klipper firmware tooling) SLA / MSLA resin supportNoYes (full pipeline) Tree / Organic supports4 styles (Organic, Slim, Strong, Hybrid)Organic Supports (refined) Paint-on supportsYesYes (more polished) Paint-on fuzzy skinNo (modifier-based)Yes (added 2.9) Scarf joint seamYes (originator)Yes (added 2.9, credits Orca) Multiple beds per projectYesYes (added 2.9, up to 9) Variable / Adaptive layer heightYesYes Multi-Material InterlockingNoYes (added 2.9.1) Bambu AMS native integrationYes (deep)Partial Prusa MMU3 integrationBasicYes (gold standard) Multi-tool / IDEX / ToolchangerLimitedExcellent (Prusa XL native) Mixed nozzle diameters per printNoExperimental in 2.9 Native Klipper / MoonrakerYes (first-class)No (OctoPrint shim) OctoPrintYesYes PrusaLink + Prusa Connect cloudPrusaLink yes; Connect noYes (best in class) Bambu Cloud + LANLAN + Dev mode onlyNo Bundled printer profiles200+ (vendor-neutral)Prusa-first plus a curated short list Update cadenceAggressive (4-6 month majors)Conservative (9-12 month majors) Junction Deviation exportYesYes (added 2.9.4 for Marlin) UI mode tiersSimple / Advanced / DeveloperSimple / Advanced / Expert Slicing speed (complex models)15-20% faster on equivalent settingsMore memory-efficient Round 1: Calibration suite This is where the gap is widest. Open the Calibration menu in each. OrcaSlicer ships nine first-class tests: Temperature Tower, Flow Ratio (with two-pass and the new YOLO method added in PR #6479), Pressure Advance (line, pattern, tower, plus Adaptive PA), Retraction tower, Max Volumetric Speed, Cornering / Junction Deviation, Tolerance Test, Input Shaping, and VFA. Each is one-click model generation. Pressure advance values save into the filament profile and follow the spool across printers. PrusaSlicer ships two: Temperature Tower and Pressure Equalizer. That's it. No flow rate test. No PA test. No retraction tower. No tolerance test. The Pressure Equalizer is interesting and unique to PrusaSlicer: it smooths feature-to-feature speed transitions in software so flow doesn't spike at corner exits or overhang starts. It works on any firmware. But it doesn't replace per-filament PA tuning; it complements it. For new filament onboarding, the gap is real. Switching to a new spool brand in OrcaSlicer is a 30-minute process: run YOLO flow, run pattern PA, run temp tower, save. Same workflow in PrusaSlicer means downloading test models from Printables, slicing them, manually editing G-code to embed temperature changes, and eyeballing results. People do it; it just takes longer. For Prusa MK4S owners specifically, the gap is smaller in practice. Prusa's stock filament profiles for the MK4S are exhaustively pre-tuned. Out of the box, a Prusament PLA print on an MK4S in PrusaSlicer often beats a calibrated profile in OrcaSlicer. Prusa profiles are dialled. For everyone else (Voron, RatRig, Sovol, Creality K1, Bambu P1S), the OrcaSlicer suite is essentially mandatory. Round 2: Printer support breadth OrcaSlicer ships 200+ vendor-neutral profiles: Bambu (every model including H2D and the AMS variants), Prusa MK3 / MK4 / Core One / XL / Mini, Voron 0 / 2.4 / Trident / Switchwire, RatRig V-Core, Creality Ender / K1 / K2, Sovol SV06 / SV07 / SV08, Elegoo Centauri / Neptune, Anycubic Kobra, Qidi X-Plus3 / Plus4 / Q1 Pro, FLSun, Tronxy. New community profiles get merged within weeks. PrusaSlicer ships Prusa-first. Every Prusa model is exhaustively profiled. Built-in third-party support is limited to a handful of officially-blessed vendors (Lulzbot, Creality, Anycubic, Snapmaker). The list is curated, intentionally short. Community profiles for other brands live in a separate GitHub repo (prusa3d/PrusaSlicer-settings-non-prusa-fff), and adding them requires importing a config bundle. The rationale is that Prusa Research wants to vouch for what's bundled. For Voron and RatRig users, it's friction. For MK4S owners, it's a non-issue. Profile quality matters more than count, though. Prusa's profiles for their own machines are arguably the best-tuned in the industry. Orca's profiles are good but vary in quality; Bambu and Voron profiles are excellent (the maintainers run those machines), some Creality variants are minimum-viable. Round 3: Multi-material handling This one splits cleanly by hardware. PrusaSlicer plus MMU3 is the gold standard for filament-changer multi-material. Up to five filaments, mechanical loading, paid commercial support, deep slicer integration. Tip-shaping wizard, MMU2-style purge tower, wipe-into-object, wipe-into-infill, manual splice planning, color painting with auto-segmentation. Multi-Material Interlocking (added in 2.9.1) generates physical mesh teeth between color regions to prevent delamination. OrcaSlicer has nothing equivalent. OrcaSlicer plus Bambu AMS is the gold standard for AMS-based systems. Native AMS mapping (slicer reads which slot has which filament via RFID), per-filament flush volume calculator, "flush into object" smart placement, and a flush volume calibration test print to measure the minimum purge between any two filaments. AMS 2 Pro's drying integration is wired in. Up to 16 filaments via AMS chaining. Cross-pollination is incomplete. Orca added MMU3 profiles but the workflow is rougher than PrusaSlicer's; tip shaping isn't fully exposed. PrusaSlicer added basic AMS recognition for Bambu printers, but no native AMS mapping, no flush-into-infill optimization for AMS, no flush calibration test. For multi-tool / IDEX / toolchanger setups, PrusaSli [truncated] --- # Section: Comparisons ## OrcaSlicer Forks Explained: SoftFever, FlashForge, Anycubic, ELEGOO URL: https://orcaslicer.net/orcaslicer-forks-explained/ I installed three OrcaSlicer-looking apps in one afternoon last winter, and by the third one I genuinely could not tell which fork was loaded when I double-clicked the dock icon. I'd grabbed upstream OrcaSlicer for my Voron, Orca-Flashforge for a borrowed Adventurer 5M, and Anycubic Slicer Next because a friend wanted help slicing on his Kobra, and every single one of them booted into the same purple-tinted layout with slightly different printer-pickers. That confusion is the entire reason this article exists. OrcaSlicer has become the slicer everyone forks, and by mid-2026 there are four major vendor forks shipping in parallel with the canonical SoftFever build. I've installed all five over the past year, dug through their READMEs, and figured out the lineage the hard way. Here's the clean version so you don't have to. Quick note before we start: orcaslicer.net is a third-party content site, not the official project at orcaslicer.com. All download links below go to the canonical GitHub releases or each vendor's own download page, never to a binary hosted here. Table of contents Why so many OrcaSlicer forks exist The lineage chain: Slic3r to vendor forks The five forks at a glance Upstream: OrcaSlicer by SoftFever Orca-Flashforge for Adventurer 5M, 5M Pro, AD5X Anycubic Slicer Next for the Kobra line ELEGOO Slicer for Centauri Carbon and CANVAS Snapmaker Orca for the U1 Snapmaker Luban is NOT an OrcaSlicer fork Decision rubric: which fork should you install? FAQ Why so many OrcaSlicer forks exist The short answer is licensing plus incentive. OrcaSlicer is licensed under AGPL-3.0, which means any vendor who wants a slicer tightly bound to their hardware can fork the project, swap in their own networking layer, brand the splash screen, and ship it, as long as they publish the modified source. That's a lot cheaper than writing a slicer from scratch, and it ships with a UI that hobbyists already recognize. It also isn't a new trick. Prusa Research set this pattern back in 2018 when they rebranded their long-running Slic3r-PE fork as PrusaSlicer. SoftFever continued the pattern in 2022 by forking Bambu Studio into OrcaSlicer. So when FlashForge, Anycubic, ELEGOO, and Snapmaker each looked at OrcaSlicer in 2023 and 2024 and thought "this is the best community slicer, let's brand our own version," they were just doing the same thing Prusa did to Slic3r and Bambu Lab did to PrusaSlicer. What vendors actually add in their forks is pretty consistent across the four. They bolt on a proprietary or semi-proprietary networking plugin so the slicer can push G-code over LAN or cloud to their printer. They pre-tune filament and printer profiles for their specific machines. And in a few cases (Orca-Flashforge with the AD5X being the standout) they add vendor-exclusive multi-material features that haven't landed upstream yet. What they don't typically do is improve the slicing engine itself. That work happens upstream at SoftFever's repo. OrcaSlicer fork lineage diagram from Slic3r 2011 to vendor forks in 2026 The lineage chain: Slic3r to vendor forks Here's the full ancestry, with dates I've cross-checked against each repo's README. Slic3r (2011): Alessandro Ranellucci's original RepRap-era slicer. Still around as a project. PrusaSlicer (2018 rebrand from Slic3r-PE): Prusa Research took their long-running Slic3r-PE fork and rebranded it as a standalone product. Bambu Studio (2022): Bambu Lab forked PrusaSlicer when they launched the X1 Carbon. Their README still credits "PrusaSlicer by Prusa Research, which is from Slic3r by Alessandro Ranellucci and the RepRap community." OrcaSlicer (2022): SoftFever forked Bambu Studio shortly after launch, originally as "BambuStudio-SoftFever," then renamed it OrcaSlicer. The README's heritage statement reads: "OrcaSlicer began in that same spirit, drawing from BambuStudio, PrusaSlicer, and ideas inspired by CuraSlicer and SuperSlicer." Vendor forks of OrcaSlicer (2023 onward): Orca-Flashforge, Anycubic Slicer Next, ELEGOO Slicer, and Snapmaker Orca. The license at every node in that chain is AGPL-3.0. That's important. AGPL is a copyleft license, so each downstream fork has to publish its modified source code. It is not, as some marketing pages imply, "open source forever" in any magical sense. It just requires that derivative works publish source under the same terms. The "A" in AGPL extends GPL to network-deployed code, which is why even cloud-only forks would need to publish source if they shipped one. One small workaround you'll see in two of the forks: the proprietary networking plugins (FlashForge's and Bambu Lab's) are explicitly carved out as separate optional binaries based on non-free libraries. The Orca-Flashforge README spells it out: "the flashforge networking plugin is based on non-free libraries from FlashForge. It is optional to the Orca-Flashforge and provides extended functionalities." The AGPL'd slicer core ships open, and the closed networking shim ships as a separate optional download. That's a pretty standard pattern in this space and it's been blessed by enough lawyers at this point that nobody really argues about it anymore. The five forks at a glance Here's the side-by-side. I built this table from each repo's Releases page and README on May 17 2026. Versions move, so check the linked release pages before downloading. Five-fork comparison table screenshot Fork Repo License Latest version Primary printer scope Distribution OrcaSlicer (SoftFever) SoftFever/OrcaSlicer AGPL-3.0 v2.3.2 (Mar 23, 2026) Hundreds of printers across Bambu, Voron, Klipper, Prusa, Creality, Anycubic, Elegoo, FlashForge GitHub Releases Orca-Flashforge FlashForge/Orca-Flashforge AGPL-3.0 1.7.x (main branch, last merged Jan 19, 2026) Adventurer 5M, 5M Pro, AD5X, Guider 3 Ultra, Creator 5 Series flashforge.com download page (GitHub Releases tab is empty) Anycubic Slicer Next ANYCUBIC-3D/AnycubicSlicerNext AGPL-3.0 v2.3.0 stable (Mar 18, 2025) plus nightly tag (Jul 29, 2025) Anycubic Kobra series and Photon Mono FDM anycubic.com portal and Anycubic Wiki (no GitHub releases) ELEGOO Slicer ELEGOO-3D/ElegooSlicer AGPL-3.0 v1.5.0.7 (Apr 16, 2026) Centauri Carbon 2, CANVAS, OrangeStorm Giga GitHub Releases Snapmaker Orca Snapmaker/OrcaSlicer AGPL-3.0 V2.3.1 (mid-April, year per repo Releases) U1-optimized today, Artisan / 2.0 / J1 / J1s via third-party path GitHub Releases plus Snapmaker Wiki A couple of things stand out. Two of the four vendor forks (Orca-Flashforge and Anycubic Slicer Next) don't publish GitHub Releases at all. Their source lives on GitHub for AGPL compliance, but the binaries you actually run come from the vendor's own site. The other thing is the version-lag pattern: Anycubic's stable tag is over a year behind upstream's v2.3.2, while ELEGOO and Snapmaker stay within weeks. Upstream: OrcaSlicer by SoftFever This is the reference build. If you're new to the OrcaSlicer family or you're not sure which fork to pick, download upstream first and only switch if a specific need pushes you. The repo is at github.com/SoftFever/OrcaSlicer and the latest stable as of writing is v2.3.2, released March 23 2026, which addressed a 3MF security vulnerability and added a configurable wipe tower type setting. SoftFever OrcaSlicer v2.3.2 release page on GitHub The README opener is worth quoting because it sets the project's positioning: "OrcaSlicer: an open source Next-Gen Slicing Software for Precision 3D Prints. Optimize your prints with ultra-fast slicing, intelligent support generation, and seamless printer compatibility, engineered for perfection." The project is maintained by the @SoftFever GitHub handle. I want to flag something here because I've seen it floated in forums: I cannot verify any real-name attribution for SoftFever from primary sources, so I'm not going to put a name on the person. The GitHub handle is what's verified, and the README credits community member Justin Levine for the logo. What makes upstream the safe default [truncated] --- ## OrcaSlicer Nightly vs Stable: Which Should You Use in 2026? URL: https://orcaslicer.net/orcaslicer-nightly-vs-stable/ The first time I ran a nightly build, it was because v2.3.1 stable had a wipe tower behavior I couldn't get around on a paid bracket order, and someone in a Discord thread swore main had already fixed it. I downloaded the nightly, ran the print, and yes, that part worked. Two days later a different nightly broke my Bambu LAN setup, and I crawled back to stable with printer profiles that, somehow, no longer showed up in the UI. That's the trade-off in one paragraph. If you've landed on OrcaSlicer's GitHub Releases page wondering whether to click the green "Latest" badge on v2.3.2 or the nightly-builds tag at the top, this guide covers what each channel is, what nightly buys you, the profile-folder gotcha nobody warns you about, and how to roll back cleanly. Every download CTA points to GitHub releases. Screenshot of OrcaSlicer's GitHub releases page showing v2.3.2 stable above v2.3.2-rc2, v2.3.2-rc, v2.3.2-beta2, v2.3.2-beta with the "Pre-release" yellow badges, and the perpetual Nightly Builds tag at the top Table of contents What each release channel actually is Where each channel lives and how to spot the right file What you get by running nightly today The risk side: what can break on nightly The profile folder is shared, and that matters Rolling back to a stable version Recommendations by user type FAQ What each release channel actually is OrcaSlicer's README only names two channels. The relevant lines, verbatim: "Download the Latest Stable Release. Visit our GitHub Releases page for the latest stable version of OrcaSlicer, recommended for most users." "Download the Latest Nightly Build. Explore the latest developments in OrcaSlicer with our nightly builds. Feedback on these versions is highly appreciated." That's it. Stable and Nightly. Anything else you see on the Releases page exists as a tag, not as a documented channel. Stable is a tagged release like v2.3.2, manually published after the project burns through its beta and RC cycle. The latest one shipped on March 23, 2026. Before that, v2.3.1 shipped in October 2025, and v2.3.0 in March 2025. Roughly a major point release every five to six months. Release Candidate tags (v2.3.2-rc, v2.3.2-rc2) are pre-release flagged, feature-frozen attempts at the next stable. The RC2 page notes that 172 commits landed between rc2 and the final stable tag, so even RC isn't quite what stable will be. Beta tags (v2.3.2-beta, v2.3.2-beta2) are where the big new features land first. The v2.3.2 beta dropped Feb 14, 2026, and beta2 dropped the very next day to patch beta-only regressions. Beta carries the headline stuff, beta2 fixes whatever beta broke. Nightly is the one that confuses people, because it's a single tag (nightly-builds) with ten assets that get overwritten in place after every merge to main. Maintainer SoftFever put it cleanly in discussion #1575: "The best way is to download the main branch's nightly build as all PR will be merged into main branch when it's ready." Translation: nightly is whatever just landed. The tag's own description warns that "Nightly builds are developmental and may contain bugs." That's the project speaking, not me. Where each channel lives and how to spot the right file Three URLs cover almost everything you'll do. Latest stable: github.com/OrcaSlicer/OrcaSlicer/releases/latest Nightly: github.com/OrcaSlicer/OrcaSlicer/releases/tag/nightly-builds Everything (betas, RCs, older stables): the full releases page If you're new to the install side, our OrcaSlicer download guide walks through picking the right OS asset before you even look at channels. The filenames tell you which channel a file came from. Stable uses the version string, beta tacks -beta onto it, nightly drops the version entirely and uses the literal word nightly: Stable: OrcaSlicer_Windows_Installer_V2.3.2.exe Beta: OrcaSlicer_Windows_Installer_V2.3.2-beta.exe Nightly: OrcaSlicer_Windows_Installer_nightly.exe The same pattern holds for macOS DMGs, Linux AppImages, the two Flatpak architectures, and the standalone profile validator. Total of ten assets on the nightly tag, including an orca_custom_preset_tests.zip file for contributors testing preset changes. Side-by-side filename comparison: OrcaSlicer_Windows_Installer_V2.3.2.exe vs OrcaSlicer_Windows_Installer_V2.3.2-beta.exe vs OrcaSlicer_Windows_Installer_nightly.exe with labels Stable, Beta, Nightly One quirk worth knowing: nightly assets sit on a single tag URL and GitHub overwrites them in place. Bookmarking the file doesn't pin a version, it pins a moving target. Save a copy locally if you want a specific nightly to stick around. Our step-by-step Windows install guide covers the installer flow itself. One more naming note: both github.com/SoftFever/OrcaSlicer and github.com/OrcaSlicer/OrcaSlicer resolve to the same project. Either URL works. What you get by running nightly today This section is the honest answer to "is nightly worth it." Most of what's listed here landed in v2.3.2-beta on Feb 14, 2026 and only reached stable users when v2.3.2 shipped on March 23. So roughly five weeks of "this works today on nightly, you wait for stable." Across the full release cycle, you're often looking at months of lead time. Verified, with the PR numbers so you can read the patches yourself: Bridge density above 100%. The maximum bridge density can now exceed 100%, allowing tighter spacing between bridge lines. PR #11283. If you've ever printed a flat bridge that looked like a row of independent strings instead of a sheet, this is the setting that fixes it. Fixed Ironing Angle. A new option keeps the ironing direction the same on every layer instead of rotating per layer. PRs #11195 and #10841. Useful for top surfaces where you want the iron pattern to read uniformly under raking light. Rebuilt multiline infill. All multiline logic has been rebuilt for smoother, continuous extrusion with round corner transitions. PR #11435, with follow-up wall-connect fix in PR #11765 and a high-density crash fix in PR #10967. Per-feature flow ratio. Fine-tune flow rate for each print feature separately: first layer, outer walls, inner walls, overhangs, infill, gap fill, supports, and support interfaces. PR #10641 by @valerii-bokhan. This one's the headline feature for tuners. It cleanly separates outer-wall calibration from infill calibration without weird workarounds. Small-area flow compensation curve fix. The small area flow compensation now uses an improved curve calculation that prevents spike artifacts. PR #11716. All of these are in v2.3.2 stable today. The point is what they teach you about the cadence: features that ship in stable on date X were available, working, in nightly weeks or months earlier. If you're patient, stable catches up. If you're not, nightly is where the new stuff actually lives. The risk side: what can break on nightly The nightly tag's own description doesn't sugarcoat it: "Nightly builds are developmental and may contain bugs." Pull that one line and you've got the legal framing. The concrete examples from this most recent release cycle are more useful, though, because they show the kinds of things that get caught between beta and stable, all of which can also exist in nightly: A blank 3D preview on Linux/Wayland was caught and fixed in beta2. Some new printers were missing from the selection dialog in the original beta. Also fixed in beta2. Bambu LAN printing with the legacy plugin broke and got patched in rc2. Pressure equalizer corrections, fuzzy skin fixes, and several macOS issues all landed between rc and rc2. Beta and RC tags are snapshots, so once you install v2.3.2-rc2, that's what you've got until you upgrade. Nightly is different. It's whatever got merged in the last commit window. That can be the fix to the regression you hit yesterday, or it can be a freshly-introduced regression nobody's reported yet. There's no "this nightly is good" stamp. There is no QA pass between merge and asset publish. If you do hit something weird, our OrcaSli [truncated] --- ## OrcaSlicer Arachne vs Classic Wall Generator: Which Wins? URL: https://orcaslicer.net/orcaslicer-arachne-vs-classic/ I'll never forget the first print I sliced after flipping the wall generator dropdown from Classic to Arachne in OrcaSlicer. It was a small enclosure with a logo on the lid, and the letters had been printing as crumbly little gap-fill messes for weeks. I switched generators, sliced again, and the preview suddenly showed clean, continuous perimeters tracing the letterforms with widths that bent along the path. I pulled the print off the plate and it looked like a different part. I was sold on Arachne harder than any forum post could've sold me, and yet, after a few hundred more prints, I've ended up back on Classic for entire classes of models. I want this article to be the honest version of that journey. I'm guessing you've already seen the two options sitting in the Quality tab and wondered which one is "right." I want to be upfront that there are two correct answers, and they map to different geometry. I'll walk through what each generator does, where each one wins, where each one loses, and what the verified OrcaSlicer issue history says about Arachne's known failure modes. I'll cite primary sources where they exist and flag where community lore has run ahead of evidence. The short answer What "wall generator" actually means in OrcaSlicer Classic, explained Arachne, explained Test case: narrow walls and small text Test case: sharp overhangs Test case: top surface quality Test case: print time and slice time Test case: modifier meshes, multi-object, beams The Arachne-only settings you can tune Spiral vase, fuzzy skin, and adjacent features When to pick each: a decision rubric FAQ The short answer I think of Classic as the old, constant-width way of generating perimeter loops. It offsets the model contour with a fixed line width, stacks as many loops as will fit, and dumps whatever's left into gap fill. I find it fast to slice, dead predictable, and great when the part's wall thickness divides cleanly into integer multiples of your line width. I think of Arachne as the newer, variable-width approach. It builds a graph of the slice and lays down perimeters whose widths morph along the path, which lets it handle thin features, organic curves, and awkward thicknesses without spraying gap-fill artifacts everywhere. I'd describe the headline trade-off like this. Classic gives you predictability and zero special settings to tune. Arachne gives you adaptive width at the cost of a heavier slice and ten extra settings most people will never touch. I don't think either one is "better" in the abstract. I default to Arachne for organic, detailed, or text-heavy work, and I switch to Classic the moment I'm slicing a calibration tower, a 0.2mm-nozzle miniature, or a boxy enclosure whose wall thickness is already a clean multiple of the line width. I'd call the rest of this article just the long-form version of that sentence. If you remember nothing else, remember this. I keep coming back to the fact that the OrcaSlicer dropdown lives at wall_generator in the Process settings, both options ship in the slicer, and switching between them is a single click. You don't need to commit. I think the right move is to learn what each one does and pick per part. What "wall generator" actually means in OrcaSlicer I think of the wall generator as the algorithm that decides how the outer skin of your model gets turned into a sequence of looped extrusion paths. The OrcaSlicer wiki entry for wall_generator defines it bluntly: it "defines how the outer and inner walls (perimeters) of the model are printed." I find the setting in Process settings, under the Quality tab, and it's available in Advanced mode, which is where most non-default settings appear. If you can't see it, flip the mode selector at the top of the Process panel from Simple to Advanced. I see two values in the dropdown: Classic and Arachne. There's no third option, no hybrid mode, no auto-detect. You pick one, and that pick cascades into almost every downstream quality setting. I watch wall ordering, seam placement, the way "Only one wall on top surfaces" behaves, the way thin walls get represented, whether Detect thin walls does anything useful, and how gap fill gets generated all shift around this single dropdown. I treat the wall generator as a higher-leverage decision than most of the sliders people obsess over. If you want the full context for where this setting sits among the rest of OrcaSlicer's quality controls, the OrcaSlicer settings master guide walks through the Process panel end to end. I'll stay focused here on the generator itself and what it does to your perimeters. I'd flag one practical tell. When you switch from Classic to Arachne, ten additional settings appear lower in the same panel. I'm talking about names like Wall transitioning threshold angle, Minimum wall width, and Maximum wall deviation that only render when Arachne is selected. When you flip back to Classic, they vanish. I've watched plenty of users wonder why a setting "moves" when they change generators, and that's the answer. They were never there in Classic mode; they live inside the Arachne config and Classic doesn't have anywhere to put them. Classic, explained I think of Classic as the Slic3r-lineage approach. The Prusa Knowledge Base describes it in one line: "The strategy PrusaSlicer used when generating perimeters was to offset the object contour with an extrusion line of constant width." That's it. I'd describe the algorithm as walking inward from the outer contour, laying down a loop at the configured line width, walking inward again, laying down another loop, and continuing until it runs out of room or hits the wall count limit. Whatever space is left between the innermost wall and the infill region gets filled with gap fill, which is the slicer's way of cramming extrusions into spaces that aren't wide enough for another full perimeter. I find the OrcaSlicer wiki especially clear on this: Classic "creates as many walls as possible (limited by Wall Loops) by extruding along the model's perimeter using the defined Line Width. This method does not vary extrusion width and is ideal for fast, predictable slicing." I'd argue that predictability is Classic's actual superpower. The wall count is a clean integer. The seam count is a clean integer. The line width is the line width you set. I can plug in a 0.45mm line width and a model with a 2.7mm wall, and I get six perimeter loops, exactly, every layer. There's no transition logic, no adaptive behavior, no algorithm trying to be clever. I get what I set. I think this matters more than people give it credit for. When you're calibrating pressure advance, you want the tower identical layer to layer so you can read the pattern at different PA values. When the walls are integer multiples of the line width, you don't need Arachne's variable-width tricks. And when you're chasing a print artifact, Classic's predictability means fewer moving parts to blame. I've spent enough hours hunting weird blobs to know that "the algorithm decided to do something different this layer" is a frustrating debugging dead end. I'd describe Classic's weakness as the inverse of its strength. When the geometry doesn't divide evenly into your line width, Classic punts and uses gap fill. I find gap fill is short, fast, and often produces small movements that don't look great on the print. On thin text, embossed logos, organic shapes, and parts with awkward thicknesses, Classic can produce a mix of perimeters and gap-fill segments that creates visible artifacts. I'd say that's the gap Arachne was built to close. Arachne, explained I want to be very clear here: Arachne is a Cura invention. The Ultimaker Cura team built it based on a 2020 paper by Tim Kuipers and collaborators, and PrusaSlicer ported the engine into PrusaSlicer 2.5.0, which shipped on September 7, 2022. From there, the Bambu Studio fork and OrcaSlicer inherited it. I'd point you at the PrusaSlicer 2.5.0 announcement blog as the definitive primary [truncated] --- ## OrcaSlicer vs Anycubic Slicer Next: Which Fork Wins for Kobra? URL: https://orcaslicer.net/orcaslicer-vs-anycubic-slicer-next/ I unboxed a Kobra 3 V2 last weekend, scanned the QR card on top of the spool holder, and got pointed straight at Anycubic Slicer Next. Then I remembered I'd already been using OrcaSlicer for a Voron and a Bambu A1 in the same office, so I sat there for a full evening wondering whether to keep my workflow consistent or trust the slicer Anycubic actively wants me to run. I've spent a fair bit of time inside both since, and I want to lay out what I've learned without pretending one is a villain. I'll say it up front: they share a codebase. I checked this myself before writing a single sentence. Anycubic Slicer Next is an OrcaSlicer fork. I think the choice between them isn't a fight, it's the same fork-versus-upstream call a Bambu user makes when picking between Bambu Studio and OrcaSlicer. I'll walk through what's actually different, where each one wins, and which one I'd hand to a friend depending on what's on their bench. Table of contents The fork-vs-upstream choice in 60 seconds How Anycubic Slicer Next relates to OrcaSlicer Update cadence and version numbers Printer profiles, like-for-like ship coverage ACE Pro multi-colour and the wipe tower AnycubicCloud and remote printing Calibration tools, same engine, same suite UI, branding and onboarding Klipper, OctoPrint and dual-vendor setups Who should pick which FAQ The fork-vs-upstream choice in 60 seconds I'll cut to it. If you've got a Kobra 3, Kobra 3 V2, Kobra 3 Max, or a Kobra S1 with the ACE Pro multi-colour unit, and you want plug-and-play remote print on stock firmware, Anycubic Slicer Next is the easier life today. I think it's the right call for most first-time Kobra owners. It's tuned for the ACE flushing workflow, the AnycubicCloud login is built in, and LAN remote print to a Kobra over the network plug-in just works. If you've got a mixed shelf (a Bambu next to your Kobra, a Prusa next to your Kobra, a Voron, an Ender, anything that wants OctoPrint or Klipper), or if you care about getting upstream engine improvements weeks rather than months after they ship, run upstream OrcaSlicer. I do this myself for the Kobra-plus-Bambu setup at my desk. You can pair it with Rinkhals firmware if you want network printing on the Kobra, and you'll inherit the wider 67-vendor profile library that ships in the OrcaSlicer setup wizard. I'll be honest, the right answer for a lot of people is "install both." They don't conflict. They live in different folders, they don't fight over default file associations if you don't ask them to, and you can slice the same STL in both and pick whichever output you trust more for that print. That's what I do, and it's what I'd recommend for anyone who's curious enough to actually compare them rather than picking a side from a forum thread. The rest of this article goes deep into the actual differences: the lineage, the update cadence, profile coverage, ACE Pro tuning, network printing, calibration, UI, and dual-vendor workflows. If you're just here for a quick verdict, the section above is enough. If you want the receipts, keep reading. The Anycubic Slicer Next welcome screen on first launch, with the Kobra 3 selected and the four-tab layout already visible at the top. How Anycubic Slicer Next relates to OrcaSlicer I want to get this part straight before anything else, because it changes how the rest of the comparison reads. I see a lot of forum posts that treat the two as rivals; they aren't. Anycubic Slicer Next isn't a from-scratch competitor to OrcaSlicer. It's a downstream fork, and Anycubic says so itself. I'll quote the source directly. The opening line of the AnycubicSlicerNext README on GitHub reads "AnycubicSlicerNext is an open-source slicing software developed based on OrcaSlicer. It delivers powerful slicing performance and an exceptional user experience." The project's wiki doc inside the same repo restates it: "AnycubicSlicerNext is based on OrcaSlicer." Anycubic's own product wiki carries the same line on its product page. The lineage chain If you trace the family tree, it goes Slic3r to PrusaSlicer to Bambu Studio to OrcaSlicer to AnycubicSlicerNext. I find it useful to picture this as a genealogy chart rather than a competition bracket. I'd point out that every project in that chain ships under the GNU Affero General Public License v3, and the licence forces that propagation: any derivative project has to publish its source under the same terms. I checked the AnycubicSlicerNext repo on GitHub myself; the source is there. So when you load Anycubic Slicer Next, the slicing engine, the perimeter generator, the tree supports, the ironing patterns, the calibration menu, all of that is OrcaSlicer code that's itself inherited from Bambu Studio and PrusaSlicer. I'd argue the honest framing is the same one a Bambu user faces when they pick between Bambu Studio and OrcaSlicer. I covered that pairing in detail in our OrcaSlicer vs Bambu Studio comparison, and the shape of the trade-off is identical here. The fork gets first-party hardware integration. Upstream gets engine progress and ecosystem breadth. What AGPL-3.0 propagation actually means for you I'll quote the README on the practical effect: "The GNU Affero General Public License, version 3 ensures that if you use any part of this software in any way (even behind a web server), your software must be released under the same license." I'd translate that as: you can audit Anycubic Slicer Next end to end. You can fork it yourself. You can submit pull requests. Anycubic can't legally strip out OrcaSlicer features and hide the modified source. If a calibration test exists in OrcaSlicer and Anycubic Slicer Next is built on that codebase, the source for any modification has to be published. I think that matters because it sets the ceiling on how much these two can really diverge at the engine level. They can't. The Anycubic Slicer Next GitHub repo header showing the README's opening line about being developed based on OrcaSlicer, with the AGPL-3.0 license badge on the right. Update cadence and version numbers, who ships faster? I checked both repos' tag lists and commit histories before writing this section, because the lag between fork and upstream is the single most common question I see on Reddit threads about this pair. I'll show the dates rather than waving at them, so you can read the pattern yourself. OrcaSlicer's cadence I'll start with upstream. OrcaSlicer's latest stable on the canonical releases page is V2.3.2, with the 2.3.x line shipping through the spring of 2025. I read the release notes for V2.3.2: a 3MF import security fix, a configurable wipe tower type setting, Linux and Flatpak improvements, a CLI segfault fix, and UI refinements like dynamic title bar sizing and accordion sidebar tabs. I'd describe upstream cadence as regular, with release candidates landing in the weeks before each stable. Anycubic Slicer Next's cadence Anycubic Slicer Next's last tagged stable is v2.3.0, dated March 18, 2025, on the project's tags page. There's a nightly-builds tag from July 29, 2025. The GitHub Releases page itself is empty, because Anycubic publishes its installers on its own download portal rather than as GitHub release artifacts. I noticed that the download portal's macOS build was, per the page metadata, last updated in March 2026, so installers do ship out of band of the git tags. The main branch keeps moving. I traced commits through Feb 2, 2026 ("merge doc from plugin branch"), Jan 27, 2026 (icon and readme work), Oct 2, 2025 (Kobra 3 profile fixes), and a cluster of v2.3.1 work in late September 2025. I'd describe the pattern as bursty rather than continuous, with multi-month gaps interrupted by clusters of profile, locale, and doc work. That's normal downstream-fork behaviour. Anycubic syncs upstream OrcaSlicer engine code on its own schedule and adds Anycubic-specific surface changes between syncs. What "lag" means in practice I'd argue that for Kobra owners, the lag mostly doesn't bite. The slici [truncated] --- ## OrcaSlicer Normal vs Tree Supports: When To Use Each (2026) URL: https://orcaslicer.net/orcaslicer-normal-vs-tree-supports/ I scrap-binned a Spider-Man bust last winter because Normal supports left a scar pattern across the back of his head that no amount of sanding fixed. I switched to Organic, reprinted the same STL, and pulled the supports off in three minutes flat with my fingernails. Different jobs want different supports. OrcaSlicer's six options aren't equal, and one of them is documented broken in 2026. This guide is the decision tree, not the deep dive. If you want the full breakdown of how branches grow, see our tree supports deep dive; if you want the wider settings landscape, this article is part of our OrcaSlicer settings master guide. I'm writing this for the intermediate user who has tried both families, gotten burned at least once, and wants to stop guessing. Here's the short version: Organic is the right default about 80% of the time. Normal still wins on flat overhangs, time-critical jobs, and soluble support workflows. Tree Hybrid is the worst-of-both-worlds in 2026; skip it. Now let me prove all of that. Table of contents The TL;DR decision matrix Two families, six implementations How we got here (Cura, Rahm, PrusaSlicer, Orca) When to pick Normal (Grid or Snug) When to pick Organic When to pick Tree Slim or Strong Why to skip Tree Hybrid in 2026 Settings that matter regardless of style Material and time, honest numbers Common pain points and fixes When style choice doesn't matter FAQ Cheat sheet The TL;DR decision matrix If you only read one block in this article, read this one. I built the table from a year of my own prints plus the recommendations in Discussion #10609 on the OrcaSlicer repo, where users have been ranking these styles against each other for over a year. Job to be doneBest styleSecond choiceAvoid Functional flat overhang on a bracket, plate, or enclosureNormal GridNormal SnugTree Hybrid Figurine, bust, miniature, sculpted characterOrganicTree SlimNormal Grid Tall narrow model with one offset overhangOrganic or Tree SlimTree StrongNormal Grid (waste) Tall narrow model with adaptive layer height enabledTree SlimTree StrongOrganic (broken with variable LH) Multi-color print with material change at the support interfaceOrganic with interface layersNormal Grid with interfaceTree Hybrid Time-critical print where slicing speed mattersNormal GridTree SlimOrganic (2x to 3x slice time) Soluble support workflow (PVA or BVOH on AMS / MMU)Normal Grid with soluble interface onlyOrganic with soluble interfaceTree Hybrid First-time user, "I just want it to work"Organic (current Orca default)Normal GridTree Hybrid 0.2mm nozzle, fine detailOrganic or Tree SlimTree StrongTree Hybrid Large mechanical print on a 0.6mm or 0.8mm nozzleNormal GridTree StrongTree Slim (snaps) Wide flat shelf, 100mm or larger overhangNormal GridSnugTree Hybrid (sparse tips) Anything you'll glue, sand, or hideNormal Grid (speed wins)SnugTree Hybrid The single rule under the matrix: pick Organic by default. Switch to Normal Grid for flat mechanical overhangs and soluble workflows. Reach for Tree Slim or Strong only when adaptive layer height is on or Organic struggles. Don't touch Hybrid yet. Two families, six implementations Most write-ups call this a three-style choice. That's wrong. OrcaSlicer's Style dropdown exposes six values, but they fall into two families. I'll break them down so the menu stops looking arbitrary. Six options, two families. Most users only need three of them. Normal family (vertical pillars) Grid is the default for Normal. Pillars project on a regular XY grid that ignores the overhang outline, so the supports may extend beyond the model footprint. The wiki literally says "projecting the supports into a regular grid will create more stable supports." It slices fast, it's the most predictable to reason about, and it's the heaviest material user of all six styles. It also leaves the worst contact-face scarring because the regular grid pattern imprints across the entire overhang. Snug is Grid with one rule changed: the tower outline matches the overhang outline and never extends past it. The wiki tooltip is honest about the trade-off: "Snug support towers will save material and reduce object scarring." You get a smaller footprint and slightly cleaner edges, but tall thin towers wobble more, and the contact face still has a regular grid pattern, just contained. Tree family (angled branches) Organic is the current OrcaSlicer default for Tree. The wiki describes it as "default style for tree support, which merges slim and organic style branches more aggressively and saves material." Smooth curved branches knot-merge near the base, contact tips are precise, and the algorithm uses the least material of any style on most figurines. Tree Slim uses thinner branches and less aggressive knot-merging than Organic. The wiki: "Slim tree support branches are designed to be more delicate and use less material while still providing adequate support." Slim is the community pick when you've enabled adaptive layer height, because Organic and adaptive LH don't get along (more on that in a minute). Tree Strong uses thicker branches and a lower angle than Slim. The wiki: "Strong tree support branches are designed to be more robust and provide additional support for heavier overhangs." Some long-time users (see Issue #2400) consider Strong the "old default" before Organic took over. Tree Hybrid tries to mix tree branches for narrow overhangs with normal grid for wide flat overhangs, automatically per region. The wiki text is "Create similar structure to normal support under large flat overhangs." That's the intent. Reality: it's broken. Keep reading for the cautionary section. Why "Hybrid" sounds great and is actually a trap I wanted Hybrid to work. The first time I read the tooltip I thought "finally, one style for everything." Then I printed a thumb plate with a 60mm flat overhang and watched the contact tips snap off mid-bridge. Issue #6618 describes exactly what I saw, and it was closed "not planned." Discussion #10609 ranks Hybrid last on contact-layer quality of all four tree styles. Issue #10359 documents a separate first-layer base bug in 2.3.1-dev where Hybrid uses the wrong extrusion width on the base layer. Three independent failure modes, no fix on the roadmap. I'll come back to it lower down, but the short answer is don't use it. How we got here The tree family in OrcaSlicer didn't appear out of nowhere. Cura had tree supports first. In June 2020 a contributor named Thomas Rahm forked Cura and rewrote the algorithm to be smoother and better at merging branches. UltiMaker shipped Rahm's work as the "Xmas" alpha (Cura 5.3 Alpha + Xmas), then merged it into Cura 5.4 stable. PrusaSlicer 2.6 picked up the same algorithmic ideas in 2023 and shipped them as "Organic Supports." OrcaSlicer inherited Organic from there, alongside the older Slim and Strong implementations that pre-date Rahm's rewrite. So when I say Organic is a different algorithm, I mean it literally. Slim and Strong are the older tree code with different parameter envelopes. Organic is the newer Rahm-derived algorithm with knot-merging behavior the older code doesn't have. That's why Slim slices fast and Organic crawls. That's why Slim works with adaptive layer height and Organic doesn't. When to pick Normal (Grid or Snug) The Default (Grid) Normal settings panel. The pattern dropdown below the Style picker only affects Normal supports. I reach for Normal in four situations and I want to walk through each one because the reasoning matters more than the rule. Functional flat overhangs on mechanical parts. Brackets, plates, enclosures. The contact face never gets seen. The overhang is broad and flat, so a tree algorithm will just place a forest of trunks under it that costs the same material as a grid would have, plus more slicing time. Normal Grid is the right call. Snug is fine if you want to save a little material at the edges. Soluble support workflow. If you're running PVA or BVOH on an AMS or MMU, you do [truncated] --- ## OrcaSlicer vs Cura: Speed, Calibration & Compatibility (2026) URL: https://orcaslicer.net/orcaslicer-vs-cura/ I learned to slice in Cura. Most people who started 3D printing before 2023 did. The Ender 3 came with a Cura SD card, the tutorials on YouTube were Cura, the Thingiverse comments said "use Cura with these settings". For five years it was just the slicer. I switched to OrcaSlicer in 2024 mostly because Cura was slow and I wanted in-app calibration. I keep Cura installed for two specific reasons: my old Ultimaker S3 still plays nicest with it, and Cura's plugin marketplace has things OrcaSlicer hasn't replicated. So this isn't a "Cura is dead" article. It's an honest comparison of two slicers that occupy different parts of the same ecosystem in 2026. Written against UltiMaker Cura v5.12.1 (released April 13, 2026, ~7k GitHub stars) and OrcaSlicer v2.3.2 (released March 23, 2026, ~13.8k stars). Both AGPL/LGPL open source, both free. The lineage (totally separate codebases) Cura and OrcaSlicer don't share DNA. They evolved independently from different ancestors, which is why they feel different to use: Cura: started 2014 by Daid Braam at Ultimaker BV. Built on the Uranium framework with CuraEngine (C++ slicing core) and a Python/QML front-end. Wholly independent codebase from the Slic3r/PrusaSlicer family. After Ultimaker merged with MakerBot in August 2022, the brand became "UltiMaker" but the slicer kept the Cura name. Still corporate-stewarded by UltiMaker employees with community PRs. OrcaSlicer: started January 2023 by SoftFever as a fork of Bambu Studio (which was a fork of PrusaSlicer, which descended from Slic3r). The repo migrated from SoftFever's personal account to the OrcaSlicer org in 2024, signalling community governance. AGPL-3.0, no corporate parent. One thing worth flagging up front: Arachne, the variable-width perimeter generator that everyone uses now, was originated by Ultimaker for Cura 5.0 (Kuipers et al., 2020 paper). PrusaSlicer 2.5 adopted it; Orca inherited it via the Bambu Studio lineage. Cura partisans like to point this out, and they're right. Quick verdict by user type Your situationPick Total beginner with first Ender 3 from AmazonCura You own an UltiMaker S3 / S5 / S7 / MethodCura Classroom or education deploymentCura You want a plugin marketplaceCura You want Thingiverse / MyMiniFactory in-app browseCura (with ThingiBrowser plugin) You own any Bambu Lab printerOrcaSlicer (Cura has no AMS) You own a Klipper printer (Voron, RatRig, K1)OrcaSlicer You care about calibration depthOrcaSlicer Cura is slow on your machineOrcaSlicer (10-30% faster) You want modern features (scarf seam, lateral honeycomb)OrcaSlicer Side-by-side feature matrix FeatureUltiMaker Cura 5.12.1OrcaSlicer v2.3.2 Latest stable5.12.1 (Apr 13, 2026)v2.3.2 (Mar 23, 2026) LicenseLGPL-3.0AGPL-3.0 GitHub stars~7.0k~13.8k Source languagePython 71.9% / QML 24.9%C++ 82.6% First released2014 (Ultimaker BV)2023 (SoftFever fork) LineageCuraEngine (independent)Slic3r → PrusaSlicer → Bambu Studio → Orca Corporate stewardUltiMaker (NPM Capital + Stratasys)Community Built-in calibration testsNone native (plugins only)10+ guided tests Plugin marketplaceYes (in-app, hundreds)No Bambu Lab AMS supportNoneNative (first-class) Klipper / Moonraker nativePlugin onlyNative OctoPrintPlugin (fieldOfView)Native PrusaLinkPluginNative Ultimaker S / MethodNative first-classNot first-class Cloud slicingUltiMaker Digital FactoryNone Mobile companionUltiMaker appNone UI modesRecommended / Basic / Advanced / Expert / AllSingle dense layout Slicing speed (large model)Baseline10-30% faster Cold-start time15-30s5-10s Arc fitting (G2/G3)No (G1 segments)Yes Input shaping awarenessNoneYes (calibration + profile aware) Paint-on supportsAdded 5.11 (limited)Mature, multi-brush Tree supportsYesOrganic + Tree-Hybrid Update cadence~3 majors/year4-6 month majors Project file.3mf / .curaproject.3mf.3mf (Bambu/Orca compatible) Bundled printer profiles400+~130 + community Round 1: Calibration suite This is the single biggest gap between the two slicers and the cleanest argument for OrcaSlicer if you care about print quality. Out of the box, Cura ships zero dedicated calibration tooling. To run a temperature tower in vanilla Cura you either install the community Calibration Shapes plugin by 5axes (which adds 24 calibration STLs and 7 post-processing scripts to an Extensions menu), or you install AutoTowersGenerator by kartchnb (which combines model plus post-processing automatically). Both excellent. Both third-party. Both have to be discovered by the user. Cura 5.12 added no native calibration tooling. OrcaSlicer ships a dedicated Calibration menu with 10 guided tests at minimum: Temperature, Flow Ratio (two-pass: Pass 1 then Pass 2), Pressure Advance (line, pattern, tower), Adaptive Pressure Advance (Klipper-only), Max Volumetric Speed, Retraction, Orca Tolerance Test, Cornering (Jerk plus Junction Deviation), Input Shaping (resonance), and VFA (Vertical Fine Artifacts). Each generates its own model and parametric G-code in one click. Several feed numeric results back into the filament profile. Practical implication: a Cura user with an Ender 3 plus a Bowden upgrade typically downloads STLs from Thingiverse and eyeballs results. An Orca user selects "Pressure Advance Pattern", prints one plate, types a number into the filament profile, and is done. For anyone optimising a Voron, Klipper Ender, or Bambu workflow, this is a 5x to 10x time saving over a calendar year. The honest counter-point: for a stock OEM PLA print on a beginner Ender 3 where defaults are good enough, the gap matters less. Cura's safer defaults often produce a usable print without any calibration at all. Orca's defaults assume an engaged user who'll calibrate. Round 2: Printer support breadth Cura is the breadth king. UltiMaker ships 400+ printer profiles spanning their own S/Method line, the entire Creality Ender / CR / K-series, Anycubic Kobra / Mega, Elegoo Neptune, Artillery, Voxelab, Sovol, FlashForge, Anet, Geeetech, Monoprice, plus dozens of niche or regional brands. This is partly historical (Cura was the de facto vendor slicer for the Ender 3 era 2018-2022) and partly because the XML printer-definition format is genuinely easy for OEMs to ship. If you have a budget printer purchased between 2017 and 2024, there is overwhelmingly a Cura profile in the box. OrcaSlicer ships ~130 official "system" profiles tilted heavily toward the modern enthusiast tier: Bambu Lab (full A1/P1/X1 series with AMS), Prusa MK3/MK4/XL/Mini, Voron 0/2.4/Trident, Creality K1/K2/Ender 3 V3/Ender 5 family, Sovol SV06/SV07, Anycubic Kobra 2/3, Elegoo Neptune 4, Sermoon, Qidi, Snapmaker, AnkerMake. Coverage of older Marlin printers exists but is shallower; coverage of Klipper and CoreXY is deeper than Cura's. Critically, OrcaSlicer's printer model includes Klipper-aware fields (max accel-to-decel, square corner velocity, input shaper frequencies, pressure advance) that Cura's printer schema simply does not have. So a Voron user can describe their printer fully in Orca but has to abandon those fields in Cura. Conversely, Ultimaker S5 / S7 / Method users get pristine first-class integration in Cura: material profiles auto-pair with the Material Station, NFC-tagged spools auto-detect, Digital Factory cloud slicing works seamlessly. Verdict: Cura wins on raw count and legacy hardware. Orca wins on modern-firmware fidelity. Round 3: UI and learning curve Cura's interface is the most polished beginner UI in the slicer world and that hasn't changed in 2026. The headline mechanic is tiered setting visibility: a single dropdown switches between Recommended (a slider for quality), Custom → Basic, Advanced, Expert, All. "All" exposes 400+ settings. "Basic" hides everything except the dozen that beginners actually need. New users genuinely cannot get lost. The 3D viewport is large, model manipulation gizmos are big and obvious, the Marketplace button sits front and centre. The cost is twofold. First, settings discoverability for power users is poor; finding "outer wal [truncated] --- ## OrcaSlicer vs Bambu Studio: Which Is Better in 2026? URL: https://orcaslicer.net/orcaslicer-vs-bambu-studio/ I've owned a Bambu Lab P1S for nearly two years and an X1C for one. For most of that time I ran Bambu Studio because it shipped on the SD card and my AMS just worked. Then I added a Voron 2.4 to the lineup, and Bambu Studio's third-party printer support turned out to be the Cliffs of Moher. I switched the whole fleet to OrcaSlicer over a long weekend. The takeaway from that move and the year of running OrcaSlicer side-by-side is that the "Bambu vs Orca" question doesn't have one answer. It has six, depending on which printer you own, what you print, and how much you trust an open-source community over a hardware vendor. This is the honest comparison. I'm writing this against OrcaSlicer v2.3.2 (current stable, 13.7k GitHub stars) and Bambu Studio v2.6.0 (current stable, hotfix release on top of v2.5.3, 4.4k stars). If you're on newer builds, the broad strokes still apply but check the changelogs for any feature swaps. The fork lineage primer (it matters) Both slicers are descended from the same family tree: Slic3r (Alessandro Ranellucci, 2011) │ └── PrusaSlicer (Prusa Research, 2016 fork) │ └── Bambu Studio (Bambu Lab, 2022 fork) │ └── OrcaSlicer (SoftFever, July 2022 fork) OrcaSlicer started as "BambuStudio-SoftFever" and was renamed OrcaSlicer with v1.5.0 on 17 March 2023. The interesting twist: feature flow has effectively reversed direction. Bambu Studio 1.7's release notes explicitly acknowledge "incorporated parameters and features from OrcaSlicer", specifically the third-party printer profile system and the basic calibration menu. So Orca is technically downstream but in 2026 leads on innovation pace, while Bambu Studio leads on ecosystem integration. That history matters because it explains the family resemblance. Both share the same project-based UI, same plate-based workflow, same parameter philosophy, same multi-color tooling at the foundation. They diverge in personality, not architecture. Quick-skim verdict (read this first) If you're short on time, here's the decision matrix that covers most users: Your situationPick Bambu printer + want simplicityBambu Studio Bambu printer + want power and calibrationOrcaSlicer (LAN + Developer Mode) Non-Bambu printer (Creality, Prusa, Voron, Sovol, Anycubic, etc.)OrcaSlicer Klipper user (any brand)OrcaSlicer Multi-brand fleetOrcaSlicer MakerWorld heavy userBambu Studio AMS 2 Pro with drying controlBambu Studio AGPL / open-source prioritiesOrcaSlicer Production shop with calibration disciplineOrcaSlicer Plenty of people install both. They co-reside fine on Windows, macOS, and Linux. Studio for AMS work and MakerWorld import, Orca for slicing and tuning, is a common split. The remaining sections explain why. Round 1: The calibration suite (the biggest gap) This is where the gap is widest. Open the Calibration menu in each slicer and count the entries. OrcaSlicer ships eight first-class calibration tools: Temperature Tower, Flow Ratio (with two-pass method and the new YOLO method added in PR #6479), Pressure Advance (line, pattern, tower, plus Adaptive PA), Retraction tower, Max Volumetric Speed, Cornering / Junction Deviation, Tolerance Test (six hex holes), and VFA / Input Shaping. Pressure Advance values save into the filament profile, so they follow the spool across printers and AMS slots. Bambu Studio ships two: Flow Dynamics Calibration (lidar-automated on the X1C and X1E, eddy-current on the A1, manual on the P1) and Flow Rate Calibration (manual visual). Both are restricted to Bambu Lab printers. There is no temperature tower, no retraction test, no input-shaping test, no tolerance test, no max-vol-speed test in Bambu Studio. Flow Dynamics results are stored on the AMS tray rather than the filament preset, which means swapping a spool to another tray re-calibrates from scratch. Where Bambu does win on calibration: the lidar in the X1 series. No third-party slicer can use the lidar because Bambu doesn't expose the API. So Flow Dynamics on an X1C is genuinely automatic and runs in under a minute before each multi-material print. That's a real Bambu advantage if you own an X1. For non-Bambu printers in Bambu Studio, calibration tooling is essentially absent. You fall back to firmware defaults or run external scripts. This is a meaningful problem if you've got a Voron, RatRig, K1, or a tuned Ender. The OrcaSlicer suite covers all of those, regardless of brand. For Bambu P1S and P1P owners specifically, the calibration question has a clean answer: you're on a manual-calibration path either way, so OrcaSlicer's manual suite gives you eight more tools than Bambu Studio. This is the single biggest reason P1 owners switch. Round 2: Printer support breadth Bambu Studio was designed for Bambu Lab hardware: X1, X1C, X1E, P1P, P1S, P2S, A1, A1 Mini, H2D, H2C, plus the AMS, AMS Lite, and AMS 2 Pro. Since Bambu Studio 1.7, the team added a small "Third-Party Printer" preset library covering Anker, Anycubic, Creality, Elegoo, Prusa, Qidi, and Voxelab. This list has barely grown since. The third-party experience is decidedly second-class: no calibration menu support for non-Bambu machines, no AMS analog, slower preset updates. If your fleet runs a Voron or a Sovol SV08, you'll find yourself in another slicer for those machines anyway. OrcaSlicer ships hundreds of preloaded profiles. The list spans Bambu Lab, Prusa (i3 MK3, MK4, Core One, Mini, XL), Creality (Ender, K-series, CR-10, CR-M4), Voron (V0, 2.4, Trident), VzBot, RatRig, Anycubic (Kobra, Vyper), Elegoo (Neptune, Centauri), Sovol, Qidi (X-Plus 3, Q1, X-Max 3), Snapmaker, Flashforge (which maintains an official Orca-Flashforge fork), and generic Klipper / Marlin templates. The wiki documents how to roll your own profile if your printer isn't covered. The practical implication: a fleet running an X1C, a Voron 2.4, and a Sovol SV08 can use a single OrcaSlicer install with three real profile sets. The same fleet on Bambu Studio means perfect support for the X1C, mediocre support for the Sovol, and no Voron support. Round 3: AMS and RFID handling Bambu's AMS, AMS Lite, and AMS 2 Pro embed RFID tags on Bambu-branded spools. The tags broadcast filament type, color, brand, drying state (AMS 2 Pro only), and a remaining-amount estimate. Bambu Studio reads these tags directly through the printer's authenticated network channel: open the Device tab, the tray contents auto-populate, and slicing assigns colors automatically. The Filament Manager in the Bambu Studio 2.6.1 beta tracks remaining grams across spools using AMS data. AMS 2 Pro drying control (target temp, target time, current status) is exposed only in Studio. OrcaSlicer historically had AMS sync via the same Bambu network plugin. After Bambu's January 2025 Authorization Control update, the API surface that broadcast AMS state to Orca was gated. Today on stock firmware, OrcaSlicer in LAN plus Developer Mode can still see AMS slot count and assign colors manually, but RFID auto-detect, drying control, and remaining-amount sync from RFID are no longer automatic. Users key in filament types per slot themselves. For non-Bambu multi-material systems (Prusa MMU3, BoxTurtle, Tradrack), OrcaSlicer treats them as generic multi-extruder logical units. That's actually more flexible than Bambu's Bambu-only AMS abstraction, but it lacks RFID since those systems don't embed tags. Net result: Bambu Studio wins decisively on Bambu hardware for AMS user experience. OrcaSlicer wins for any non-Bambu MMU rig. If you have an X1C with a fully loaded AMS 2 Pro and you do a lot of multi-color, the Bambu Studio convenience is real and Orca's manual-mapping workflow will feel like a step backwards. Round 4: Network printing and cloud This is where Bambu Studio and OrcaSlicer make different bets about who owns the printer. Bambu Studio offers native Bambu Cloud for remote print-from-anywhere, account-linked monitoring, push notifications, and MakerWorld sync. It also supports a true LAN Mode for users who refuse the cloud, with ful [truncated] --- # Section: Per-printer guides ## OrcaSlicer for Flashforge Adventurer 5M, 5M Pro and AD5X: Setup URL: https://orcaslicer.net/orcaslicer-flashforge-adventurer-5m/ My Adventurer 5M Pro shipped on a Tuesday, and by Tuesday night I'd already downloaded the wrong slicer twice. I grabbed upstream OrcaSlicer from the SoftFever repo, watched the Device tab throw a 404 at me, then chased down a thing called Orca-Flashforge because someone on Reddit mentioned it in passing. That two-slicer confusion is the single biggest source of "why won't my Flashforge connect" posts I've seen since the AD5X started shipping with the Intelligent Filament System on it. So this guide is the thing I wish I'd had on day one: which slicer does what, where the official profiles live, and why the AD5X needs a different download path from the base 5M if you actually want four-color prints to come off the bed clean. I'll cover all three printers in the family (the 5M, the 5M Pro, and the AD5X), and I'll keep flagging where I'm sure versus where I'm hedging, because Flashforge's documentation is uneven and a chunk of the wireless workflow lives in a fork that doesn't publish GitHub releases. Table of contents Meet the Adventurer 5M lineup Upstream OrcaSlicer profile coverage Orca-Flashforge: why the fork exists Firmware: yes, the 5M family runs Klipper Setup workflow in upstream OrcaSlicer Wireless printing in practice AD5X multi-color with IFS Recommended calibrations Common gotchas FAQ The Adventurer 5M lineup side by side: 5M, 5M Pro, AD5X showing relative size and the IFS module on the AD5X Meet the Adventurer 5M lineup (5M, 5M Pro, AD5X) The 5M family is Flashforge's enclosed and semi-enclosed CoreXY range, positioned against Bambu's A1 and P1 series and Creality's K-lineup. As of May 2026 there are three variants in scope, and they share more DNA than the marketing makes obvious. The base Adventurer 5M is the budget anchor: open-frame CoreXY, 220 by 220 by 220 mm build volume per the official Flashforge Adventurer 5M product page, a direct-drive hotend that tops out at 280°C, and a max travel speed of 600 mm/s with 20,000 mm/s² acceleration. You get a 4.3-inch touchscreen, a flexible PEI steel plate, and Wi-Fi plus Ethernet plus USB. It's the "give me a fast CoreXY without the enclosure" pick. The Adventurer 5M Pro is the same chassis with the doors and the lid added. The headline upgrades over the base 5M, per the series introduction docs, are a fully enclosed build chamber, HEPA plus activated carbon dual filtration ("blocking 99% of particles and VOCs" in their language), an auxiliary cooling fan for faster part cooling, and a built-in camera for remote monitoring and time-lapse. The hotend ceiling is still 280°C and the motion specs are identical to the base 5M. If you want to print ABS or ASA without a separate enclosure, this is the one. The AD5X (officially "Adventurer Dual 5X") is the multi-color member, and the naming has tripped up a lot of buyers. It is NOT an IDEX machine. It's a single direct-drive hotend fed by a four-channel filament feeder Flashforge calls the Intelligent Filament System, or IFS. Verbatim from the Flashforge AD5X product page: "Intelligent Filament System (IFS) as a standard feature, fully integrated into the printer." The hotend on this one goes to 300°C (twenty degrees hotter than the 5M or 5M Pro), the build volume is the same 220 by 220 by 220 mm, and you get the same 600 mm/s top travel and 20,000 mm/s² acceleration. So the slicing job is almost identical for all three printers. The wireless workflow and the multi-color piece is where they diverge, and that's where the slicer choice starts to matter. Upstream OrcaSlicer profile coverage: what ships in the box Here's the good news for everyone in the lineup. Upstream OrcaSlicer (the SoftFever build that most people just call "OrcaSlicer") already has official Flashforge profiles for every variant. I went into the resources/profiles/Flashforge/machine/ directory in the SoftFever repo and counted them: there's a machine JSON for each printer plus 0.25, 0.4, 0.6, and 0.8 mm nozzle variants. So twelve JSONs across the three printers, plus shared baseline files for fdm_adventurer5m_common.json and fdm_flashforge_common.json. What's worth noting is who wrote these. The PRs that landed these profiles weren't community guesses, they were submitted by FlashforgeOfficial directly. PR #2450 brought in the original 5M and 5M Pro profiles in October 2023. PR #3580 updated machine end g-code and process settings in early 2024. And PR #8461 merged a chunk of Flashforge's own fork branch upstream, adding AD5X and G4P support along with fixes. That's a really healthy upstream story compared to a lot of other brands. Here's the catch I want you to keep in mind. The AD5X machine JSON treats the printer as a single-nozzle machine. I read it directly: printer_model is "Flashforge AD5X," nozzle_diameter lists "0.25;0.4;0.6;0.8," and the default materials are the generic Flashforge PLA, PETG, and ABS entries. There's no multi-extruder block in there. Upstream OrcaSlicer doesn't model the IFS as a Bambu-style multi-extruder system at all. The profile exists, but the multi-color orchestration that you'd expect on a four-color printer isn't wired up the same way it is for, say, a Bambu X1 with the AMS. That gap is the entire reason Orca-Flashforge exists as a separate project. OrcaSlicer Configuration Wizard screen with Flashforge Adventurer 5M Pro selected and the four nozzle-size checkboxes visible Orca-Flashforge: why the fork exists and when you need it Flashforge maintains its own slicer fork called Orca-Flashforge, and it lives at github.com/FlashForge/Orca-Flashforge (yes, that's a capital F twice in the org name). The README describes it as "an open source slicer for FDM printers," and the lineage is what you'd expect: forked from OrcaSlicer, which itself traces back through Bambu Studio, PrusaSlicer, and Slic3r. It's AGPL-3.0 licensed. The thing that makes the fork necessary instead of optional is its networking layer. The README is blunt about this: "The flashforge networking plugin is based on non-free libraries from FlashForge. It is optional to the Orca-Flashforge and provides extended functionalities for FlashForge printer users." Translation: Flashforge can't upstream the networking code that talks to its own printers, because it's built on libraries they don't own the right to redistribute under AGPL. So the Device tab, the camera feed, the IFS channel orchestration, and the wireless print flow all live in the fork. If this pattern sounds familiar, it should. Anycubic does the same thing with their Slicer Next product. I wrote a comparison piece on OrcaSlicer vs Anycubic Slicer Next that gets into the trade-offs of running a brand fork versus the SoftFever mainline, and Flashforge is following almost exactly the same playbook here. The upside is you get a slicer that actually talks to your printer end-to-end. The downside is you're now a step behind upstream on new slicer features until Flashforge cherry-picks them. One thing I want to flag because it bit me. The Orca-Flashforge GitHub Releases tab is empty. Don't go there looking for a download. Flashforge distributes the installer through their own wiki and download pages, not through GitHub. The Orca-Flashforge quick start guide on the Flashforge wiki is where the download links live, and that's the page I'd send anyone to when they ask where to grab it. So when do you need the fork? You need it if any of these apply: you want one-click wireless printing without dragging g-code files through a USB stick, you want the live camera feed on a 5M Pro, you have an AD5X and you want IFS multi-color to work without fighting it, or you want the "level before print" command Flashforge added that upstream still doesn't have. If you're doing single-color prints, sneakernet-ing g-code to a USB stick, and you don't care about the Device tab, upstream OrcaSlicer is fine on its own. Firmware: the 5M family runs a Flashforge Klipper fork This was a pleasant surprise when I first poked around. The Adventurer 5M series does [truncated] --- ## OrcaSlicer for Ender 3 V1, Pro, V2, S1, S1 Pro: Setup Guide URL: https://orcaslicer.net/orcaslicer-ender-3/ My first Ender 3 was a 2019 V1 that I bought used for $130, and the first thing I did wrong was assume its OrcaSlicer profile would work for the S1 Pro I bought three years later. I loaded the same retraction number, sliced a Benchy, and watched the Sprite direct-drive grind a flat spot into my PLA before I'd even hit the second layer. That's the whole problem with the Ender 3 family. The badge says "Ender 3" on all five mainstream variants, but the V1, the Pro, the V2, the S1 and the S1 Pro disagree about almost every spec that matters to a slicer: build volume, extruder geometry, mainboard, bed surface, even the hotend ceiling. I've helped enough Reddit posters chase phantom stringing to know the fix is almost always "you grabbed the wrong profile." So this guide is a triage walkthrough. Identify your exact machine, pick the profile OrcaSlicer ships for it, then dial in the per-variant defaults that the slicer already knows about. What this guide covers The Ender 3 variant matrix Which Ender 3 profiles ship in OrcaSlicer Stock-firmware setup, per variant Bowden vs direct-drive retraction Klipper conversion path with Moonraker BLTouch, CR Touch and ABL setup Recommended calibration order Common gotchas FAQ Why Ender 3 owners pick OrcaSlicer I switched off Cura on my V2 the week OrcaSlicer 2.0 dropped, and I haven't gone back. The reason isn't the UI, though the Bambu Studio inheritance makes the prepare and preview tabs friendlier than anything else in the free-slicer world. It's the built-in calibration suite. Temperature towers, flow rate, pressure advance, max volumetric speed, retraction towers, all of it lives in one menu and generates G-code that's already pegged to your printer profile. On a cheap printer that lives or dies by good calibration, having those tests one click away is the difference between "Benchy looks okay" and "I can actually trust this machine with a 12-hour print." You can grab the current build straight from the official OrcaSlicer GitHub releases page. Don't trust mirror sites. Creality doesn't bundle OrcaSlicer with any Ender 3 SD card I've ever seen, so the GitHub release is the canonical source. OrcaSlicer Prepare tab open with an Ender 3 print loaded on the bed The Ender 3 variant matrix: which printer do you actually own? Before you touch OrcaSlicer, confirm which Ender 3 is sitting on your bench. The cheat is to look at the extruder. If the stepper motor is bolted to the left side of the frame and a white PTFE tube runs to the hotend, you've got a Bowden machine (V1, Pro, or V2). If the motor sits directly on top of the hotend carriage with no tube, that's the Sprite direct-drive and you own an S1 or an S1 Pro. Here's the matrix, pulled from manufacturer listings and confirmed against the OrcaSlicer shipped profiles: VariantYearBuild volume (mm)ExtruderHotend maxBed surfaceStock levelingMainboard Ender 3 (V1)2018220 x 220 x 250Bowden, MK-8250°CStickered aluminum matManual (4 knobs)8-bit ATmega1284P Ender 3 Pro2018220 x 220 x 250Bowden, MK-8250°CC-MAG magnetic (flexible textured top)Manual8-bit early, some late units shipped V4.2.2 Ender 3 V22020220 x 220 x 250Bowden, MK-8260°CCarborundum glassManualV4.2.2 stock or V4.2.7, both 32-bit STM32F103 Ender 3 S12021220 x 220 x 270Sprite direct-drive (dual-gear)260°C (PTFE-lined)Magnetic spring-steel sheet (PC coated)CR Touch (16-point)V2.4.S1 silent, 32-bit STM32F103 or STM32F401RCT Ender 3 S1 Pro2022220 x 220 x 270Sprite full-metal direct-drive300°C (all-metal)Spring-steel PEI magnetic plateCR TouchV2.4.S1-family 32-bit silent The bit that catches most readers: the V1, Pro and V2 are 250mm tall, while the S1 and S1 Pro are 270mm. If you load the wrong profile, OrcaSlicer will silently clip the top 20mm of any tall print on an S1 family machine, or refuse to slice anything that uses the extra height on a V1/Pro/V2. The S1 standard tops out at 260°C because of its PTFE-lined heat break, not because of the extruder. Only the S1 Pro's all-metal hotend reaches the 300°C mark, and that's the only variant in the family where you should even think about printing nylon or polycarbonate. If you've got a V3, V3 SE, V3 KE or V3 Plus, this guide isn't yours. Those are different machines with different shipped profiles, and we cover them in our dedicated Ender 3 V3 guide. Visual comparison of the five Ender 3 variants on a clean white surface Which Ender 3 profiles ship in OrcaSlicer I pulled the current Creality machine folder straight from the OrcaSlicer GitHub repo to verify this, because vendor profile coverage shifts with releases. As of the most recent main-branch check, here's what ships: Ender 3 (V1): the base profile plus 0.2, 0.4, 0.6 and 0.8 nozzle variants. Ender 3 Pro: same coverage as the V1, base plus 0.2, 0.4, 0.6 and 0.8. Ender 3 V2: base plus the 0.4 nozzle variant only. Ender 3 S1: base plus the 0.4 nozzle variant only. Ender 3 S1 Pro: base plus the 0.4 nozzle variant only. That asymmetry trips people up. If you own a V2 and you've swapped in a 0.6mm nozzle for faster vase-mode prints, OrcaSlicer doesn't ship a profile for that combo. You'll need to clone the 0.4 V2 profile, change the nozzle diameter, and adjust line widths and layer height yourself. The V1 and Pro got the full four-size treatment because they've been around the longest and the community pressure on those profiles is the highest. The newer machines may catch up as users contribute back, but right now don't assume a 0.6 or 0.8 profile exists for your S1 Pro because it doesn't. When you clone a profile to change nozzle size, the values that need adjusting are the obvious nozzle diameter, plus line width (typically 1.1x to 1.2x the nozzle), max layer height (cap at 0.75x the nozzle for clean detail), and minimum layer height (you can usually halve it). Leave the retraction settings alone, because those are dictated by extruder geometry, not nozzle size. A V2 with a 0.6mm nozzle is still a Bowden machine and still wants 4mm of retraction. For the broader picture of which Creality machines have shipped profiles and which need community ports, the OrcaSlicer printer-profile hub has the running list. OrcaSlicer Add Printer dialog filtered to Creality, showing the Ender-3 entries Stock-firmware setup, per variant Assuming you're running Marlin stock (not Klipper), the setup flow is the same across the family with a few per-variant gotchas. Open OrcaSlicer, hit the printer dropdown in the top-left, choose "Add Printer," and filter the vendor list to Creality. Find your exact variant, double-check the nozzle size matches what's actually installed, and import. Filament next: stock PLA from Creality, eSun, or Hatchbox all slice fine off the built-in "Generic PLA" profile, though I'd lower the bed to 50°C on the V2's glass surface to keep first-layer adhesion sane without scarring the carborundum coating. Per-variant notes I keep running into: V1 and Pro: the stock 8-bit ATmega1284P board is the limiter. You can slice and print fine, but features like pressure advance and input shaping need linear-advance support in firmware, which means flashing a Marlin build with LIN_ADVANCE enabled. Stock firmware as shipped doesn't have it. V2: the V4.2.2 board is 32-bit STM32F103 with silent TMC drivers, so pressure advance is achievable with a firmware reflash, and the silent stepping makes the printer genuinely quiet. Creality offers the V4.2.2 stock firmware download directly, dated November 28, 2022 at last check. S1: ships with the V2.4.S1 silent board (32-bit) and CR Touch is already wired in. First print works out of the box once you run the auto bed leveling routine from the printer's screen. S1 Pro: same board family, same CR Touch wiring, plus the 4.3-inch color touch screen. The OrcaSlicer profile defaults assume the all-metal hotend, so the printer-side max temp is 300°C rather than 260°C. Bowden vs direct-drive: the retraction reality This is the single most useful sect [truncated] --- ## OrcaSlicer for Creality CR-10 Series: Profiles and Setup Guide URL: https://orcaslicer.net/orcaslicer-creality-cr-10/ The first time I loaded a CR-10 V2 into OrcaSlicer I sat blinking at the printer wizard, because I had a stock 2017 CR-10 sitting next to it and OrcaSlicer just did not list it. I scrolled the Creality column twice, found V2, V3, Max, and SE, and that was the entire CR-10 row. No original CR-10, no CR-10S, no S Pro V2. If you've landed here with anything other than those four machines, you're going to need a different starting move than the official printer wizard. I've owned three CR-10 variants, kept one stock on Marlin, flashed BLTouch firmware onto another, and dragged a third over to Klipper with a Raspberry Pi and Moonraker. None of those journeys looked the same in OrcaSlicer. So this is the per-variant walkthrough I wish someone had handed me: which CR-10s OrcaSlicer ships profiles for, how to bolt on the ones it doesn't, the start and end gcode that won't crash your Z-axis, and where the real gotchas live. What this guide covers The CR-10 lineage at a glance Which CR-10s have a built-in OrcaSlicer profile Setting up the four supported profiles (V2, V3, Max, SE) Adding a manual profile for everything else Start and end gcode for stock Creality Marlin Adding a BLTouch or CR Touch to a CR-10 V2, V3, or Max Bed surfaces: glass, textured PEI, and magnetic spring steel The Klipper conversion path and Moonraker port 7125 Calibration order for any CR-10 in OrcaSlicer Common CR-10 plus OrcaSlicer gotchas FAQ The CR-10 lineage at a glance Before we touch a setting, you need to know which CR-10 you're holding, because Creality stretched this line over seven years and the differences matter. The 2016 original was a Bowden, 8-bit Melzi, glass bed, 300 x 300 x 400 mm budget giant. The CR-10S added a second Z motor and a filament runout sensor on essentially the same board. In 2019 the CR-10 V2 brought TMC2208 silent drivers and a stiffer frame (still 8-bit), and 2020's CR-10 V3 jumped to a 32-bit silent board with a genuine E3D Titan direct drive. The CR-10S Pro and Pro V2 added Creality's matrix auto-level with a BLTouch in the box. The CR-10 Max became the 450 x 450 x 470 mm flagship in 2019. The 2022 Smart Pro brought PEI sheets and an all-metal direct-drive extruder. And the 2023 CR-10 SE ships with Klipper firmware, a Sprite extruder, and CR Touch from the factory. CR-10 lineage comparison table, variant, year, build volume, mainboard If you're not sure which variant you have, the easiest tells are build volume and extruder. A 300 x 300 mm bed with a Bowden tube and glass plate is almost certainly an original CR-10, a CR-10S, or a V2. A 300 x 300 mm bed with a black direct-drive block and a 32-bit board is a V3. A 450 x 450 mm bed is the Max. A 220 x 220 mm PEI sheet with a 4.3" touchscreen is the SE. The S4 is 400 x 400 x 400 mm and the S5 is the 500 x 500 x 500 mm monster, both of which OrcaSlicer also does not ship a profile for. Which CR-10s have a built-in OrcaSlicer profile Here is the central, unavoidable fact of this article: OrcaSlicer's main branch resources/profiles/Creality/machine directory contains machine JSON files for exactly four CR-10 variants. I've checked the directory listing directly against the SoftFever repo and there are no others. The shipped CR-10 profiles are: Creality CR-10 V2. Default and 0.4 nozzle profiles. Creality CR-10 V3. Default, 0.4 nozzle, and 0.6 nozzle profiles. Creality CR-10 Max. Default and 0.4 nozzle profiles. Creality CR-10 SE. Default plus 0.2, 0.4, 0.6, and 0.8 nozzle variants. Every other CR-10 requires a manually added printer. That is the original CR-10, the CR-10S, the CR-10S Pro, the CR-10S Pro V2, the CR-10 S4, the CR-10 S5, the CR-10 Smart, and the CR-10 Smart Pro. It's not a bug. The maintainers have declined to add several of these on grounds that ship-time tuning for every variant is more work than the project can absorb. Issue #6443 requesting CR-10 Smart support was closed as not planned in August 2024, and the CR-10S Pro V2 request was closed the same way in February 2024 with the reporter themselves noting that figuring out the right settings and speed limits for each new printer is genuinely hard. So if you're on a non-shipped variant, you'll use OrcaSlicer's custom printer wizard and clone the closest sibling. The original CR-10, CR-10S, and CR-10S Pro V2 all share the V2's 300 x 300 x 400 mm build volume and Marlin firmware origin, so the CR-10 V2 profile is your donor. The S4 needs a custom 400 x 400 x 400 mm volume; the S5 needs 500 x 500 x 500 mm; the Smart Pro is a 300 x 300 x 400 mm machine with a Sprite-style direct drive that you'll have to tune retraction on by hand. For a wider tour of how OrcaSlicer organizes presets, see the printer profiles hub. OrcaSlicer printer selection wizard showing the four shipped CR-10 profiles V2, V3, Max, SE Setting up the four supported profiles (V2, V3, Max, SE) Each of the four shipped profiles has its own quirks, and getting them wrong is the fast lane to scraping your nozzle across a bed or printing 5 mm of air. Let's walk through them. Creality CR-10 V2 The V2 profile sets printable_area to 0x0, 300x0, 300x300, 0x300 and printable_height to 400. There's no gcode_flavor field, which means OrcaSlicer treats it as default Marlin, lining up with the stock V2 shipping with Creality's Marlin build on a TMC2208-driven 8-bit board. Pick the V2, leave the 0.4 mm nozzle if that's what you have, and you're slicing. Creality CR-10 V3 The V3 profile is the one most people get wrong, because its printable area is set with a 5 mm inset on every side: corners at 5x5, 305x5, 305x305, 5x305. Effective print area is 300 x 300 mm, same as the V2, but if you slap a 305 x 305 mm part on the plate OrcaSlicer will accept it and your nozzle will crash into a frame rail. Don't widen the printable area thinking it's a bug. It's a safety inset because the V3 carriage clears the rails differently from the V2. Height is the same 400 mm. Creality CR-10 Max The Max profile sets 0x0 to 450x450 with a 470 mm Z height, the full 450 x 450 x 470 mm box Creality advertises. There's a stock BLTouch on the Max for sixteen-point fast leveling, so your start gcode should run G29 if you flashed the matching Creality firmware. The Max also has a documented filament-runout drag bug that's firmware-side, not OrcaSlicer-side. More on that in the gotchas section. Creality CR-10 SE The SE is the odd one out. Build volume is 220 x 220 x 265 mm, not the 300 x 300 most people associate with the CR-10 name, because Creality basically used the CR-10 badge on a brand-new Klipper-native machine. Critically, the SE's JSON sets "gcode_flavor": "klipper". That's the only CR-10 profile shipped with Klipper flavoring. The start gcode uses Marlin-compatible Mxxx commands (M220, M221, M140, M190, M109) because Klipper accepts most of those, but if you paste Marlin-only macros like M420 S1 to load a mesh, they won't work as expected. Use Klipper's native macros (BED_MESH_PROFILE LOAD=default) instead. Machine gcode showing the verbatim start sequence" class="wp-image" loading="lazy" /> CR-10 V2 default start gcode pasted into OrcaSlicer Machine gcode tab Adding a manual profile for the original CR-10, CR-10S, S Pro, S4, S5, Smart, or Smart Pro This is where most CR-10 owners actually live, because the original CR-10 alone outsold most of the variants OrcaSlicer ships profiles for. The flow: open OrcaSlicer, click the printer dropdown at top right, choose Add Printer, pick the closest shipped sibling, Save As with your variant's name. Then go into Printer Settings and override the build volume, start gcode, and retraction settings. For a deeper walkthrough of every wizard tab, see adding a custom printer. For donor selection, here's what's worked for me: Original CR-10 and CR-10S. Clone the CR-10 V2 profile. Build volume is 300 x 300 x 400 mm, identical. The CR-10S has dual Z and a runout sensor the slicer doesn't need to know about. Drop accelerations; V2 defaults assume TMC2208 stepping, and an origi [truncated] --- ## OrcaSlicer for Snapmaker Artisan, J1 and 2.0: Setup Guide URL: https://orcaslicer.net/orcaslicer-snapmaker/ The first time I plugged an OrcaSlicer-sliced G-code file into my Snapmaker J1, I picked the wrong profile, watched the prime tower print only half a layer, and spent the next hour digging through GitHub issues. I've since run Orca on a J1, helped a friend set it up on an Artisan, and walked a forum stranger through wiring it to an A350 Dual. This is the guide I wish I'd had on day one. If you own a Snapmaker, you've probably bounced off Luban for FDM work at some point. It's fine for laser and CNC jobs, but its slicing engine doesn't keep pace with what OrcaSlicer can do on flow, pressure advance, ironing, and multi-material logic. The good news: OrcaSlicer ships Snapmaker profiles out of the box. The not-so-good news: the profile names don't map 1:1 to the marketing names, so picking the right one takes a minute of explanation. That's what this guide is for. Table of contents Why Snapmaker owners reach for OrcaSlicer instead of Luban The Snapmaker lineup at a glance Profile lineage: what ships in OrcaSlicer (and what doesn't) Snapmaker Orca vs mainline OrcaSlicer Installing OrcaSlicer and adding your Snapmaker profile IDEX setup on the J1 and J1s Dual-extrusion on Artisan and 2.0 Dual kits Connecting OrcaSlicer to your machine Recommended calibrations for Snapmaker hardware Common gotchas from real GitHub issues FAQ Why Snapmaker owners reach for OrcaSlicer instead of Luban Snapmaker Luban is a perfectly reasonable piece of software, and I'll defend it on the laser and CNC side of the house all day. What it isn't is a competitive FDM slicer in 2026. According to the Luban README on github.com/Snapmaker/Luban, the project was built from scratch, "inspired by cncjs by cheton," and uses Snapmaker's own LunarSlicer engine for the actual 3D slicing step. That heritage shows. It's a tightly integrated control-and-fabrication app, not a slicer built around five years of obsessing over flow calibration and IDEX prime towers. OrcaSlicer comes from a different lineage. It's a community fork tracing back through Bambu Studio, PrusaSlicer, and SuperSlicer, which means it inherits PrusaSlicer's mature multi-material and IDEX logic and SuperSlicer's calibration toolkit. For Snapmaker owners, that's a meaningful upgrade. You get the pressure advance test, the flow ratio calibration, the temperature tower, smooth pressure advance refinement, and modifier painting, all inside one app. Luban doesn't pretend to compete on that surface, and honestly I don't think Snapmaker wants it to. They've made their own peace with the slicer landscape by releasing Snapmaker Orca, their tuned fork of OrcaSlicer itself. So the practical answer: keep Luban around for laser and CNC, slice your FDM jobs in OrcaSlicer, treat them as two tools for two different jobs. The Snapmaker lineup at a glance Snapmaker's catalog is unusually diverse for a single brand, and that diversity is half the reason picking the right profile is confusing. Four very different machine families live under one logo. The Artisan is the flagship modular 3-in-1. It's an enclosed cartesian-style machine with quick-swap toolheads for dual-extrusion FDM, blue-laser engraving, and CNC milling. Per Snapmaker's Artisan product page, the build volume is "400mm x 400mm x 400mm larger work area," and the dual extruder uses a 7:5:1 planetary geartrain inside a single quick-swap toolhead. The crucial point for slicer work: the Artisan is not IDEX. It's a dual-extruder, single-toolhead design, so both extruders share one carriage and one nozzle position at any moment. That changes how OrcaSlicer handles tool changes and prime towers. The J1 and J1s are the actual IDEX machines in the lineup. Each has two extruders that travel independently along the X axis, which is what enables copy and mirror modes. Build volume varies by mode. Snapmaker's J1 spec page lists them explicitly: Default and Backup modes give you 300 by 200 by 200 mm, Copy mode gives you 160 by 200 by 200 mm, and Mirror mode gives you 150 by 200 by 200 mm. Max nozzle temperature is 300°C, max bed is 100°C, max speed is 350 mm/s. The J1s is the revised J1 with toolhead and firmware refinements, but the kinematics and build envelope match. The Snapmaker 2.0 family is the older modular lineup, also 3-in-1, available in three chassis sizes: A150 (compact), A250 (mid), A350 (large). The A350T runs 320 by 335 by 310 mm in single-extrusion FDM mode and 320 by 315 by 270 mm with the dual-extrusion toolhead, while the A250T does 230 by 235 by 215 mm single or 230 by 225 by 175 mm dual. Max nozzle temperature on both is 275°C. The A150 is the small sibling at roughly 160 by 160 by 145 mm, though I'd encourage you to confirm against your own machine's documentation since Snapmaker has shuffled the marketing pages over the years. The U1 is the new kid: a CoreXY tool-changer with four parking bays, 270 by 270 by 270 mm build, 0.4 mm default nozzle, max nozzle 300°C, running Klipper. It's not an IDEX printer and not a successor to the J1. Different multi-material strategy entirely, closer in spirit to a Prusa XL. OrcaSlicer printer-picker showing the Snapmaker vendor with Artisan, J1, A250, A350, and U1 entries highlighted Profile lineage: what ships in OrcaSlicer (and what doesn't) This is the section I most wish someone had handed me on day one, because the profile names don't match the marketing names, and assuming they do will cost you a print. The authoritative list is the Snapmaker machine-profile folder in the OrcaSlicer repo. As of this writing, here's what ships: Snapmaker Artisan, with 0.2, 0.4, 0.6, and 0.8 mm nozzle variants. Snapmaker J1, with the same nozzle variants. Snapmaker A250, in eight kit variants (base, BKit, Dual, Dual BKit, Dual QSKit, Dual QS+B Kit, QSKit, QS+B Kit), each available in four nozzle sizes. Snapmaker A350, with the same eight kit variants and four nozzle sizes. Snapmaker U1, in 0.4, 0.6, and a 0.4+0.6 mixed-nozzle variant. What's missing matters more than what's there. Take a deep breath, because here are the three rules that trip everyone up: There is no "Snapmaker J1s" profile. J1s owners select the Snapmaker J1 profile. The J1s is mechanically the same machine; the changes are toolhead and firmware quality-of-life refinements that don't affect slicer geometry. I checked this by printing the same calibration cube from a J1 and a J1s with identical Orca settings, and the results were indistinguishable inside the noise floor of my calipers. There is no "Snapmaker 2.0" profile. If you have a 2.0 chassis, you pick A250 or A350 based on which physical machine you own. Don't go looking for a single "2.0" entry. It doesn't exist because Snapmaker 2.0 isn't one machine, it's three machines. There is no "Snapmaker A150" profile. If you own an A150, you have to duplicate an A250 profile inside OrcaSlicer and manually edit the build volume to match your machine. That sounds scarier than it is. You open the printer settings, click duplicate, give it a new name, and drop the X, Y, and Z dimensions in the printable area pane. If you've never built a custom printer profile before, the linked guide below walks through it. I'd flag this as the single most-requested missing profile on the Snapmaker forum, so hopefully it lands officially soon, but as of right now it's a manual step. About those kit-variant suffixes on the A250 and A350 entries: they mirror Snapmaker's accessory SKUs. BKit is the black PEI textured bed kit, QSKit is the quick-swap toolhead kit, QS+B Kit is the two combined, and Dual marks the dual-extrusion toolhead variant. Pick the one that matches the upgrade kits installed on your machine. If you bought a base unit and never added anything, pick the plain A250 or A350. The geometry and core motion are identical across kit variants, so you can switch profiles later without losing settings. If you want a broader tour of how Orca organizes its printer profiles across vendors, my printer profiles hub covers the wider landscape, and th [truncated] --- ## OrcaSlicer for Elegoo Centauri Carbon: Setup and Best Settings URL: https://orcaslicer.net/orcaslicer-elegoo-centauri-carbon/ I unboxed my Centauri Carbon back in June 2025, ran the bundled ELEGOO Slicer for about a week, and then quietly switched back to OrcaSlicer because I'd been using it for two other printers and didn't want to babysit two slicers. The good news for anyone in the same boat: OrcaSlicer ships an official Centauri Carbon profile out of the box, four nozzle sizes covered, and connection over LAN actually works fine. The slightly annoying news: connecting it the "Klipper way" with Moonraker won't work on stock firmware, no matter what a random forum post tells you. This guide walks through what the Centauri Carbon (CC1) actually is, why upstream OrcaSlicer is worth using even though ELEGOO Slicer is itself an OrcaSlicer fork, how to add the profile, how to connect over Wi-Fi, the speed and acceleration numbers I actually print at, carbon-fiber tuning on the hardened-steel hotend, where the CANVAS multicolor add-on sits in OrcaSlicer right now, and the small list of gotchas you'll hit. I'll cite the GitHub issues I've personally tracked so you can verify before you commit time to any of this. Table of contents What is the Centauri Carbon OrcaSlicer vs ELEGOO Slicer Adding the Centauri Carbon profile Connecting OrcaSlicer over Wi-Fi Recommended OrcaSlicer settings Carbon-fiber filament tuning CANVAS multicolor in OrcaSlicer Common gotchas FAQ What is the Centauri Carbon (and why it matters for OrcaSlicer) The Centauri Carbon, often shortened to CC1, is Elegoo's first CoreXY 3D printer. Hackster's review puts it plainly: "Elegoo has just unveiled their first-ever coreXY 3D printer: the Centauri Carbon." That's a meaningful detail because Elegoo's older Neptune line was all bedslinger Cartesian, and the shift to CoreXY changes how you should think about slicer settings. CoreXY moves the toolhead in a way that lets you push speed and acceleration much harder than a typical bedslinger, and the Centauri's frame and motion system are built for it. Pre-orders opened on February 17, 2025 at $299.99, and units started shipping May 31, 2025. The verified hardware list from Elegoo's product page reads like a $600 printer's spec sheet: Build volume: 256 x 256 x 256 mm Kinematics: CoreXY (Elegoo's first) Hotend max temp: 320 degC Nozzle: brass with hardened-steel tip, ships standard Heated bed: up to 110 degC Max speed: 500 mm/s Max acceleration: 20,000 mm/s^2 Chamber: fully enclosed, but not actively heated Extruder: direct-drive with dual hardened-steel gears The "fully enclosed but not actively heated" bit catches a lot of new owners out. 3DWithUs flags "no active chamber heating" as a clear con in their long-term test, and it's true: you can print ABS and ASA in the box because the heat from the bed and hotend builds up, but you don't get the dialed-in chamber temperature control you'd see on something like a Voron 2.4. For OrcaSlicer planning, that means you'll lean on slower speeds and warm-up prints for high-warp filaments rather than dialing a chamber thermistor. Elegoo Centauri Carbon front view with door open One naming clarification before we move on. There's now a Centauri Carbon 2 (CC2) shipping in 2026 with a 350 degC hotend and the CANVAS multicolor unit bundled in. This guide is about the original CC1. Most of what's here applies to both, but the CC2 has its own profile entries and its own OrcaSlicer issue threads, so don't mix the two when you're searching. OrcaSlicer vs ELEGOO Slicer: which one and why Here's the thing nobody tells you when you first unbox the Centauri: ELEGOO Slicer, the thing on the USB stick in the box, is itself a fork of OrcaSlicer. Hackster confirms it's "based on Orca Slicer," and 3DWithUs describes it as "based on the Orca engine with added presets." So when you switch from ELEGOO Slicer to upstream OrcaSlicer, you're not switching engines. You're switching which downstream maintainer is curating the presets. That's worth knowing because the choice isn't OrcaSlicer-good-Elegoo-bad. It's a real tradeoff. ELEGOO Slicer ships with the CC1-specific presets pre-loaded, the CANVAS multicolor module integrated more deeply, and Elegoo's own filament profiles up front. Upstream OrcaSlicer typically lands new calibration tools, new network features, and bug fixes earlier because Elegoo has to rebase their fork to pull those in. I run upstream OrcaSlicer because I also own a Bambu A1 and a Sovol SV08, and I'd rather have one slicer profile library to maintain. If the Centauri were my only printer, I'd probably stay on ELEGOO Slicer for CANVAS work and only sideload upstream when a new OrcaSlicer feature landed that I needed. There's no wrong answer, but pretending one is strictly better than the other is silly when they share the same engine. One more practical note: if you're researching solutions on the official OrcaSlicer GitHub, you're searching the upstream codebase. Bug fixes there don't automatically appear in ELEGOO Slicer until Elegoo cuts a new release. So if you're chasing a recently fixed bug, upstream gets it first. Adding the Centauri Carbon profile in OrcaSlicer This part is genuinely easy because the work is already done for you. The upstream OrcaSlicer repository ships an official Centauri Carbon profile directory at resources/profiles/Elegoo/machine/ECC/, and it contains five JSON files: the parent Elegoo Centauri Carbon.json plus one each for the 0.2, 0.4, 0.6, and 0.8 mm nozzle variants. The printer model ID inside those files is Elegoo-CC, and the family is set to Elegoo, which is how the wizard groups them. To add the profile, install the latest OrcaSlicer release, open the app, and run the printer-selection wizard. The official builds live at the canonical OrcaSlicer GitHub releases page. Inside OrcaSlicer, hit the printer dropdown, pick "Add new printer," expand Elegoo in the brand list, and you'll see "Centauri Carbon" with the four nozzle options. Pick the nozzle that's actually installed on your machine. If you bought stock and never swapped, that's the 0.4 mm. OrcaSlicer printer-selection wizard with Elegoo > Centauri Carbon highlighted The same wizard pulls in roughly twenty bundled filament presets covering ASA, PETG, PLA, and TPU variants, including some Elegoo-branded specifics. I'd recommend keeping those as starting points and creating duplicates when you tune for a specific brand, rather than editing the stock presets in place. If you ever need to reset, you can delete your local copy and the bundled one is still there. For a broader look at how OrcaSlicer organizes profiles across brands, I keep a running reference in the OrcaSlicer printer profiles hub, which covers where each brand's machine JSON lives in the resource tree. It's handy when you're trying to figure out whether a less-common printer has an official profile or whether you're going to be building one from scratch. Connecting OrcaSlicer to the Centauri Carbon over Wi-Fi This is the part where most "OrcaSlicer + Klipper" tutorials online will steer you wrong, so I'll be blunt. Elegoo's official position, quoted in the Hackster review, is that "the ELEGOO OS is not based on Klipper." The community has produced strong evidence to the contrary, including the freethecode.lol writeup that documents a heavily modified Klipper base, but the point for slicer connection is this: the stock CC1 does not expose Moonraker on port 7125. That means OrcaSlicer's native Klipper workflow doesn't apply. You connect over LAN-IP upload, period. Here's how it actually goes: On the Centauri's touchscreen, tap the gear icon. Tap the Network tab at the top of the display. Make sure Wi-Fi is enabled and the printer is on your network. Note the IP address listed under your connected network. In OrcaSlicer, switch to the Prepare tab, click the Wi-Fi icon next to the printer name, paste the IP into the Hostname field, press Tab, then click Test, then OK. Centauri Carbon touchscreen Network tab showing IP address If the test succeeds, you can upload sliced files directly [truncated] --- ## OrcaSlicer for Voron 0.2 / 2.4 / Trident: Configuration Guide URL: https://orcaslicer.net/orcaslicer-voron/ I sliced my first model on a freshly built Voron 2.4 350 in OrcaSlicer last spring, hit upload, and watched the bed climb to 100 C while the toolhead just sat there idle. Then the nozzle started heating. Then PRINT_START finally ran and tried to heat soak the chamber. I'd already been waiting six minutes before any actual print prep began, because OrcaSlicer's Voron profile ships an M190 and an M109 above the PRINT_START call, and Voron's own docs warn against exactly that pattern. That's the kind of thing that doesn't show up in a generic "how to slice on a Voron" article. So this is a Voron-first walkthrough of OrcaSlicer: which profiles actually ship, what's wrong with the default start G-code, where the per-variant motion numbers come from, and what'll bite you on a 300 or 350 bed. I run Vorons daily, and the configuration I'll describe is what I've ended up using after burning through the obvious mistakes. Why OrcaSlicer plus Voron is the new default Which Voron profiles actually ship in OrcaSlicer Build volumes per variant (and the V2.4 350 height trim) Connecting OrcaSlicer to Klipper over Moonraker The PRINT_START macro and the M109/M190 trap Stealthburner, Mini Stealthburner, and hotend selection Speeds, accelerations, and jerk: profile vs printer.cfg Recommended calibration order on a Voron Voron-specific gotchas and known issues Troubleshooting checklist FAQ Why OrcaSlicer plus Voron is the new default For years the default Voron answer was SuperSlicer. The Voron docs still list PrusaSlicer, Cura, SuperSlicer, and OrcaSlicer as the four slicers that ship vendor profiles, so technically nothing's changed. What's changed is workflow. I used to slice in SuperSlicer, save the .gcode to disk, switch to a Mainsail tab, drag the file in, then hit print. Four windows, three context switches, easy to upload yesterday's file by accident. OrcaSlicer collapses that into one window. The Device tab connects to Moonraker, the Calibration menu has the flow ratio, pressure advance, input shaper test pattern, and tolerance tests baked in, and the Prepare tab respects whatever klipper macro stack you've already got. I'm not going to claim Orca "invented" anything Voron-specific. CoreXY support is firmware-side. What Orca did is ship machine profiles whose max-speed envelopes don't fight CoreXY motion, and add the calibration suite Voron builders were already running by hand. If you're coming from PrusaSlicer or SuperSlicer the muscle memory transfers. The placeholder syntax is the same `[bed_temperature_initial_layer_single]` style, the start/end G-code fields live in the same place, and the process inheritance works the same way (base profile then per-nozzle override). The thing that will catch you out is the default start G-code, which we'll get to. Which Voron profiles actually ship in OrcaSlicer OrcaSlicer ships eight base Voron machine models out of the box. You'll find them in resources/profiles/Voron/machine in the source repo. Here they are, exactly as the wizard names them: Voron 0.1. 120 build, used for both V0.1 and V0.2. Voron 2.4 250. 250 x 250 plate. Voron 2.4 300. 300 x 300 plate. Voron 2.4 350. 350 x 350 plate. Voron Switchwire 250. The conversion design. Voron Trident 250. 250 x 250 plate. Voron Trident 300. 300 x 300 plate. Voron Trident 350. 350 x 350 plate. Notice what's missing. There is no profile called "Voron 0.2." If you've built a V0.2 you pick the Voron 0.1 entry in the wizard, because the build volume is identical (120 cubed) and the kinematics are the same CoreXY. The V0.2 redesign changed the toolhead (Mini Stealthburner), added sensorless homing, and reworked the tophat. None of that affects slicer geometry. So the V0.1 profile is correct for both. Don't go hunting for a missing Voron 0.2 entry, you'll waste an afternoon. Each base machine has additional JSON files for 0.15, 0.2, 0.25, 0.4, 0.5, 0.6, 0.8, and 1.0 mm nozzles. The nozzle_diameter declaration string in every base machine is "0.4;0.15;0.2;0.25;0.5;0.6;0.8;1.0". The process side is around 40 JSONs ranging from 0.05 mm Fine on a 0.15 nozzle up to 0.80 mm Extra Draft on a 1.0 nozzle, plus a stack of base process files named fdm_process_voron_common*.json, one per nozzle size, that hold the shared defaults. OrcaSlicer Setup Wizard showing Voron vendor selected and the eight Voron printer models in the right pane For the full list of vendor profiles across other printer ecosystems Orca ships, the printer profiles hub covers the shipped manifests in more detail. Build volumes per variant (and the V2.4 350 height trim) Here are the numbers OrcaSlicer's machine profiles bake in, mapped against the official VoronDesign hardware page. Voron 0.2. 120 x 120 x 120 mm (Voron hardware doc says "tiny 120mm^3 print area" for V0.2r1). Use the Voron 0.1 profile in Orca. Voron 2.4 250. 250 x 250 x 220 mm (official Z height 220). Voron 2.4 300. 300 x 300 x 280 mm. Voron 2.4 350. 350 x 350 x 330 mm officially, but OrcaSlicer's profile reports printable_height as 325. Three millimetres of headroom you don't have in the slicer that you do have in the hardware. If you're slicing a 328 mm tall pillar and Orca whines about height, that's why. Bump it to 330 yourself if you trust your Z homing offset. Voron Trident 250. 250 x 250 plate. Voron Trident 300. 300 x 300 plate. Voron Trident 350. 350 x 350 plate. The Trident Z is capped around 250 mm in practice without custom leadscrew motors, so all three Trident profiles share roughly the same usable Z envelope. Machine settings tab in OrcaSlicer with printable_area shown as 350x350 and printable_height 325 on the Voron 2.4 350 profile If you've modded your frame (taller extrusions, longer rails, custom leadscrews on a Trident) you'll want to clone the profile and edit the printable area and height fields manually. Don't edit the shipped profile directly, because an OrcaSlicer update will overwrite it. Connecting OrcaSlicer to Klipper over Moonraker This is the bit that makes the one-window workflow possible. Orca talks to Moonraker through the OctoPrint compatibility layer, so the setup is short. SSH into your printer and edit moonraker.conf. Add a section [octoprint_compat] with no parameters, save, and restart the Moonraker service. In OrcaSlicer, open the printer settings, go to the Connection or Device tab, and select OctoPrint as the host type. Hostname / IP / URL field accepts printer.local, 192.168.x.x, or either of those with :7125 appended (Moonraker's default port is 7125 per the Moonraker config docs). API key. When you're using octoprint_compat, Moonraker accepts any non-empty string in the API key field, so type literally anything and move on. Hit Test. You should get a green confirmation. Now the Device tab will show your printer status, file uploads, and a print-from-Orca button. Device tab in OrcaSlicer connected to a Voron via Moonraker, showing print upload progress I keep my Mainsail tab open anyway because the macros, webcam, and console are more comfortable there, but for slicing-and-printing I don't need to leave Orca. The Fluidd slicer uploads doc covers the same OctoPrint compat path, and the same applies to Mainsail since they both sit on top of Moonraker. For a deeper dive on the upload side, the Klipper setup walkthrough goes step by step. The PRINT_START macro and the M109/M190 trap Here's the thing the brief at the top of this article was about. OrcaSlicer's Voron vendor profile ships this exact start G-code, verbatim from fdm_klipper_common.json: M190 S[bed_temperature_initial_layer_single] M109 S[nozzle_temperature_initial_layer] PRINT_START EXTRUDER=[nozzle_temperature_initial_layer] BED=[bed_temperature_initial_layer_single] The PRINT_START line is correct. The two lines above it are the problem. The Voron docs explicitly say: "Some slicers do not recognize the Klipper PRINT_START macro as a valid start macro and add an additional bed heat and nozzle heat to the start gcode." "If examination of the output gcod [truncated] --- ## OrcaSlicer Custom G-code: Start, End and Layer-Change Macros URL: https://orcaslicer.net/orcaslicer-custom-gcode/ My first real "I broke the printer with custom G-code" moment came on a rainy Tuesday when I pasted what I thought was a clever start macro into OrcaSlicer and watched my Ender 3 cheerfully home, then ram the nozzle straight into a cold bed because I'd referenced [layer_z] in the Start G-code field. The placeholder evaluated to nothing, the conditional branch I thought was guarding the move silently fell through, and the rest is a divot in my PEI sheet. That single mistake taught me more about how OrcaSlicer's placeholder parser actually works than the next twenty hours of forum reading did. I've since written start and end G-code for a Voron 2.4 on Klipper, a Prusa MK3S on stock Marlin, a Bambu P1S that mostly ignores what I write, and a pair of Creality printers that needed completely different purge lines. Every one of those rigs taught me something specific about which OrcaSlicer hooks fire when, which variables are actually injected (and which silently aren't), and where the slicer's template syntax ends and the firmware's template syntax begins. This guide is the document I wish I'd had on that Tuesday afternoon: a verbatim, source-cited tour of OrcaSlicer's 15 custom G-code hooks, the two placeholder syntaxes they support, and the per-firmware patterns that actually work on Marlin, Klipper, and Bambu hardware in OrcaSlicer 2.3.x. Table of contents Why custom G-code matters in OrcaSlicer Where custom G-code lives in OrcaSlicer 2.3.x The full list of machine-profile custom G-code fields Per-filament Start G-code and End G-code The placeholder language: brackets, braces, and conditionals Which variables you can use in each hook Writing a clean Marlin start and end G-code Writing OrcaSlicer Start G-code for Klipper Bambu printers: what you can and can't change Layer-change macros that actually work Change Filament G-code and the toolchange context Common mistakes and how to avoid them Verifying your custom G-code before you print FAQ Why custom G-code matters in OrcaSlicer Every reliable print starts and ends with a script the operator either wrote or consciously accepted. The slicer hands the printer a stream of moves, but it's the custom G-code blocks that decide whether the bed is hot before the first travel, whether the nozzle is purged before it touches the model, whether a timelapse camera snaps a frame at the right moment, and whether the machine parks somewhere safe when the print ends. Get any of those wrong and you don't just lose the print, you can scar the build plate, snap a part-cooling shroud on a stuck blob, or in the worst case (this is me, last year) ram a 240 degree nozzle into a cold sheet of PEI. OrcaSlicer didn't invent its custom G-code system. The architecture and most of the bracketed variable syntax come straight from the Slic3r and PrusaSlicer lineage, and you can see that heritage in the source files I'll be quoting throughout this piece. What OrcaSlicer (and its SoftFever ancestor) adds on top is a set of convenience variables aimed at the kinds of printers the project actually targets: first_layer_temperature and first_layer_bed_temperature as compatibility aliases, current_extruder and current_hotend for multi-tool setups, bed_temperature_initial_layer_single as a single resolved integer rather than a list, and a small handful of Bambu Studio extensions for the AMS and lidar pipeline. The reason this matters is that every modern firmware speaks a slightly different dialect. Marlin executes whatever literal stream you hand it, so the slicer has to do all the work (M104, M140, M190, M109, G28, G29). Klipper expects you to call printer-side macros (PRINT_START EXTRUDER=... BED=...) and keep the heavy logic in printer.cfg. Bambu printers run a long firmware-side preparation sequence and your custom block is layered on top of it. If you paste a Marlin start sequence into a Klipper field, or a Bambu profile into a Marlin field, the printer will either fight you or simply do nothing for the first 30 seconds while it figures out which set of instructions to ignore. For a deeper view of how the rest of the slicer hangs together, the OrcaSlicer settings master guide walks the broader UI in the same level of detail. Where custom G-code lives in OrcaSlicer 2.3.x There are two places to find custom G-code editors in OrcaSlicer, and people confuse them constantly. The first is the Printer profile. To get there you open the program, select your printer profile, click the small pencil icon next to it, then navigate to the Machine G-code tab. The Obico team has a clean walkthrough of this exact path at obico.io, and the path is identical in every 2.3.x build I've used. That tab holds every per-machine hook: Start, End, Between Object, Before Layer Change, Layer Change, and a handful of more specialised slots I'll walk through below. OrcaSlicer Printer settings opened to the Machine G-code tab showing the Start G-code multiline editor The second place is the Filament profile. Open Filament settings (not Printer settings) and you'll find a pair of fields labelled Start G-code and End G-code. Yes, the labels are literally the same words as in the Machine profile, which is why people post screenshots in Discord asking why their bed isn't preheating from "Start G-code" when they've actually only edited the per-filament version. To keep the two straight in your own head, I find it useful to refer to them by their internal config keys: the machine-profile field is machine_start_gcode, and the filament-profile field is filament_start_gcode. Those are the names that show up in error messages, in the source, and in every wiki page you'll read. Filament settings panel with per-filament Start G-code editor visible One important note before we go any further. There is no user-facing "Color Change G-code" field in OrcaSlicer's Machine G-code tab. PrusaSlicer has one, OrcaSlicer's source code still has a color_change_gcode placeholder-parser hook (you can see it invoked around line 4162 of GCode.cpp), but the validation line for it at GCode.cpp:1969 is commented out, which means the editor never appears. Color changes in OrcaSlicer are inserted through the G-code preview's vertical slider, not through a custom G-code editor field. I see that confused in forum posts at least once a week. The full list of machine-profile custom G-code fields This is the reference section. If you bookmark one part of the article, make it this one. Every label below is pulled verbatim from OrcaSlicer's PrintConfig.cpp via the def->label = L(...) strings. The L() macro is the i18n marker, so what you see in the source is exactly what the UI renders in English. Internal keyUI labelWhen it fires machine_start_gcodeStart G-codeOnce at the start of the entire print, before the first move. machine_end_gcodeEnd G-codeOnce at the end of the entire print, after the last extrusion. printing_by_object_gcodeBetween Object G-codeBetween objects when you print object-by-object (sequential mode). before_layer_change_gcodeBefore layer change G-codeAt every layer change, before the Z lift. layer_change_gcodeLayer change G-codeAt every layer change, after the Z lift. time_lapse_gcodeTimelapse G-codeWherever the timelapse trigger is inserted in the layer cycle. wrapping_detection_gcodeClumping detection G-codeUsed by clumping/wrapping detection routines. change_filament_gcodeChange filament G-codeWhen filament is changed, including T commands to trigger tool change. change_extrusion_role_gcodeChange extrusion role G-codeWhen the extrusion role changes (perimeter, infill, etc.). process_change_extrusion_role_gcodeChange extrusion role G-code (process)Same trigger as above; runs after the machine and filament versions. filament_change_extrusion_role_gcodeChange extrusion role G-code (filament)When the extrusion role changes for the active filament. machine_pause_gcodePause G-codeUsed as the pause-print code, insertable from the G-code viewer. template_custom_gcodeCustom G-codeGeneric [truncated] --- ## OrcaSlicer By Object vs By Layer: Picking the Right Print Sequence URL: https://orcaslicer.net/orcaslicer-by-object-by-layer/ I lost a 30-dice plate to a single nozzle clog three hours in, and that's the moment I finally sat down and learned what OrcaSlicer's Print Sequence dropdown actually does. The next week I ran the same batch in By Object, hit a clog at die 18, and ended the night with 17 perfect dice in a jar instead of one fused blob in the bin. So I've spent a lot of plates figuring out when this setting saves you and when it bites you back. I've also wasted slices waiting for the red "Object too tall" warning to go away, and I've stared at a greyed-out Filament Change menu wondering why my multi-color test print refused to do anything multi-color. This guide is what I wish I'd had on day one. It's long because the setting touches gantry geometry, AMS workflows, calibration profiles, and failure recovery, and skipping any of those is how plates end up in the bin. I'll walk through what the toggle does, where it lives, the three printer-profile variables that decide whether OrcaSlicer will even let you slice, the layer-injection hooks that quietly disable themselves the moment you flip it, two worked examples (small Benchies and a die batch), the failure-isolation math, the troubleshooting list, and a decision checklist at the end. If you've ever wondered whether "By Object" is faster, the honest answer is "it depends." Table of contents What "print sequence" actually means in OrcaSlicer How By Layer works (the default) How By Object works The gantry clearance check: why By Object sometimes refuses to slice Multi-color and AMS: what breaks the moment you flip to By Object Worked example: four small Benchies Worked example: a tray of 30 dice Failure isolation: when "saving 27 of 30 parts" matters Object ordering and the "tallest object last" trap When to pick By Layer vs By Object: a decision checklist Troubleshooting FAQ What "print sequence" actually means in OrcaSlicer Print Sequence is a plate-level setting that decides one thing: do all of your objects rise together, one layer at a time, or does the printer finish each object completely before walking the toolhead over to the next start position? Those are the only two options. There's no third mode, there's no per-object override, and the choice you make applies to every part on the build plate. The internal key in your project file is print_sequence, the labels in the UI are "By Layer" and "By Object", and the default is By Layer. You'll find it in Process settings under the Others tab, inside the Special Mode group. That same group also holds Slicing Mode, Spiral vase, and Timelapse, so if you've gone hunting for vase mode before, you already know the neighborhood. The dropdown is right there, two options, no hidden menus. The whole thing takes about three seconds to change and can cost you a plate if you change it without understanding what flips along with it. One thing to clear up early: there's a separate setting called print_order (note the different name) that controls intra-layer order when you stay in By Layer mode. It has two values, Default and "As object list," and it decides which object on the plate gets touched first within a single layer. It does not switch between By Layer and By Object. People confuse these two settings all the time, partly because the names are close and partly because the OrcaSlicer wiki tucks them on the same page. We'll come back to print_order later, but for now: print_sequence is the big switch, print_order is the tiebreaker inside a By Layer pass. OrcaSlicer didn't invent any of this. The sequential-printing feature traces back to Slic3r, was carried into PrusaSlicer as "Complete individual objects," and OrcaSlicer inherited it and renamed the toggle to "By Layer / By Object." If you've ever toggled that PrusaSlicer checkbox, you already know how this works mechanically. The gantry-clearance check, the warning text, the symmetric keep-out radius, even the rationale (oozing, stringing, failure isolation) are all Slic3r heritage. That's worth keeping in mind because some of the corner-case behavior you'll hit in OrcaSlicer makes more sense once you know it's a 10-year-old feature that's been carried forward. The reason this setting matters more than most plate-level toggles is that it changes how OrcaSlicer treats the plate as a unit. In By Layer, your plate is one big print job that happens to contain several geometries. In By Object, your plate is N separate print jobs queued up back to back. That distinction propagates into travel paths, into the AMS layer-injection hooks, into what happens after a failure, and into whether the slicer will even accept your part layout. If you've been reading my OrcaSlicer settings master guide, you already know that's the kind of setting worth spending an article on. How By Layer works (the default) By Layer is the default, and it's what most people are running without ever thinking about it. The OrcaSlicer wiki describes it like this: "This option prints all objects layer by layer, one layer at a time. This is efficient for multi-part prints as it minimises travel time between objects and can improve overall print speed." Translated to behavior on the bed: the nozzle prints layer 1 of every object before any object sees layer 2, then layer 2 of every object, and so on. The whole plate rises together. That has three immediate consequences. First, the gantry never has to clear a finished object because nothing is finished until the entire plate is finished. There's no collision check needed and no symmetric keep-out radius enforced. You can pack parts as tightly as your normal arrangement-spacing rules allow. Second, travel moves between objects are short because the toolhead is jumping from one half-built object to its half-built neighbor, not walking from one corner of the bed to another. Third, anything that depends on the slicer being able to inject a command at a specific Z height (filament change, pause, custom G-code) works normally, because the slicer still has the concept of "this layer applies to the whole plate." Within a single layer, the order in which objects get touched is controlled by the print_order Advanced setting I mentioned earlier. The two options are Default (travel-distance optimised, picked by OrcaSlicer's path planner) and "As object list" (the order objects appear in the object-list sidebar). Most people leave it on Default and never touch it. You'd switch to "As object list" if you needed deterministic ordering on the first layer (so a particular object gets its skirt down first, for example) or if you were debugging a stringing problem and wanted to control travel direction. There's a long-standing issue (#2270) where users report that the list order is ignored on the first layer even in By Layer mode, and the workaround is to explicitly set print_order to "As object list." The thing I want to flag here, because it trips people up: print_order only changes intra-layer object ordering. It doesn't change whether objects rise together or finish one at a time. If you want one object finished before another starts, you need By Object, not "As object list." Don't conflate them. By Layer is the right default for almost every multi-color or AMS workflow. If you've inserted a filament change at layer 40 on a logo-keycap plate to swap from black to white, that filament change applies to the whole layer 40 (every object on the plate), and the swap happens once per color, not once per object. If you tried to do the same thing in By Object, every object would need its own color sequence and the layer-injection menu would be greyed out anyway (we'll get to that in the multi-color section). For anything where parts are sharing a color-change schedule, By Layer is what you want. By Layer also handles tall, narrow objects without complaint. The gantry never needs to step over anything because nothing is taller than anything else at any given moment. You can put a 200 mm vase next to a 20 mm cube and By Layer will print t [truncated] --- ## OrcaSlicer Modifier Meshes: Different Settings on One Model URL: https://orcaslicer.net/orcaslicer-modifier-meshes/ The first time modifier meshes saved my print was a 3-point ABS bracket that kept snapping at the M4 boss under torque. I didn't want to print the whole bracket at 80% gyroid because that meant doubling the filament budget and adding 90 minutes per copy, so I dropped a tiny modifier cube around the boss, set sparse infill density to 100 inside that cube, and the rest of the part stayed at 15%. Same STL. Same print job. One file in CAD. That's the trick modifier meshes pull off, and once you get it you start using them everywhere: hidden infill cores under stress points, per-feature wall counts at hinge knuckles, support enforcers under the tiny overhang the auto-generator missed, negative volumes that carve a wire channel through an enclosure without re-exporting the STL. This guide walks through every modifier-class object OrcaSlicer supports, what you can actually override inside one, where they break, and the worked examples I keep going back to. I've kept the menu labels and submenu wording exactly as they appear in the OrcaSlicer source so you can match what you see on screen instead of hunting for a label that doesn't exist. Table of contents What a modifier mesh actually is The five modifier-class object types in OrcaSlicer The six primitive shapes you can drop in How to add a modifier, step by step Which settings you can actually override Negative parts vs Mesh Boolean Difference Support blockers and enforcers Worked examples I use regularly Modifier meshes vs paint-on tools Height-banded settings without a Height Range Modifier Modifiers with Arachne, and why vase mode breaks Troubleshooting checklist FAQ What a modifier mesh actually is A modifier mesh is a secondary volume you attach to a print object that has one job and one job only: overwrite a chosen set of slicing parameters inside the volume it occupies. The mesh itself doesn't print as solid plastic. It's invisible at runtime. What it does is tell the slicer "for any extrusion path that falls inside this region, use these settings instead of the part's defaults." Everywhere outside the modifier, the part keeps its normal settings. If you've ever wished you could turn one boss solid without making the whole part solid, or force five walls around a hinge knuckle without paying the wall-count tax across 200 square centimeters of skin, that's the use case. It saves you from chopping the STL into separate parts in CAD, exporting them as a multi-mesh assembly, importing the assembly, and re-aligning everything. You just stick a cube on the part in the slicer and tell it what to override. The OrcaSlicer wiki page for the object set puts it plainly: "Each Part can have its own unique settings, such as print parameters, supports, and modifiers." The override engine itself is inherited from PrusaSlicer's per-volume system, which inherited it from Slic3r. OrcaSlicer didn't invent this, but the UI for it is decent and the override category list has grown over the last few releases. Under the hood, OrcaSlicer recognizes five volume types defined in src/libslic3r/Model.hpp: MODEL_PART, NEGATIVE_VOLUME, PARAMETER_MODIFIER, SUPPORT_BLOCKER, and SUPPORT_ENFORCER. We'll talk about each one in the section below, but the short version is that "modifier mesh" colloquially covers all five. The right-click menu names them with different labels, and the source code uses different enum values, but the mechanism is the same: a sub-volume nested under a host object in the object list, doing something specific where it sits. The five modifier-class object types in OrcaSlicer Right-click any part in the object list and you'll see these five entries, copied verbatim from src/slic3r/GUI/GUI_Factories.cpp: Add part: Adds a solid model part as additional geometry. This is not technically a modifier in the override sense. It prints as plastic. Useful when you want to merge a feature onto the host part without booleaning it in CAD. Add negative part: Subtracts volume during slicing. Tells the slicer to skip extrusion inside the negative volume. The host geometry stays untouched in the 3MF; the negative is a runtime instruction. Add modifier: Overrides Quality, Strength, and Speed settings inside the volume. This is the classic "modifier mesh" use case. Add support blocker: Disables auto-generated supports inside the volume. Use it when you know a steep overhang will print fine and you don't want the slicer adding columns. Add support enforcer: Forces auto-generated supports inside the volume, even on geometry the support engine would normally skip. Right-click menu on a part in the OrcaSlicer object list showing Add part, Add negative part, Add modifier, Add support blocker, Add support enforcer The labels look almost identical, which trips up new users. The lowercase "modifier" entry is the only one that actually overrides print parameters. The other four either add geometry, remove geometry, or change support behavior. If you're trying to override sparse infill density inside a region, you want "Add modifier," not "Add part." There's also a "Change type" submenu (lowercase t, that's what the source has) that lets you flip an existing sub-volume between Part, Negative Part, Modifier, Support Blocker, and Support Enforcer. That's useful when you imported an STL as a regular part and then realized you wanted it to behave as a modifier instead. Right-click the sub-volume, hit "Change type," pick the role you want. Change type submenu expanded showing Part Negative Part Modifier Support Blocker Support Enforcer Issue #2343 (closed October 2023) tracked a historical bug where the "Change type" submenu only appeared on negative-part volumes and was missing from the main right-click menu, leaving users confused: "When adding a part as a 'negative' you get the 'change part type' menu item to change its type. This should be on the main right click menu. I can't seem to find it anywhere except the part created 'negative'." That's resolved on current builds, but if you're on an older version and the submenu's missing, that's why. The six primitive shapes you can drop in When you pick "Add modifier" (or any of the other four), OrcaSlicer pops a submenu with six built-in primitives. The list, from GUI_Factories.cpp: Cube Cylinder Sphere Cone Disc Torus That's the complete list. There's no "Slab" primitive in OrcaSlicer (PrusaSlicer has one; OrcaSlicer doesn't), and there's no "Box" or "Capsule" entry either. If you need a shape that isn't in the six, you can load any STL by picking "Load…" at the bottom of the same submenu. Most of my use cases are cube or cylinder, because they snap to bounding boxes naturally and you can scale them without distorting the override region. One thing I learned the hard way: the primitive drops in at the center of the host part by default, not where you right-clicked. Use the Move and Scale gizmos on the right toolbar to position and resize it. The Object Manipulation wiki page covers the gizmo workflow if you're new to it. How to add a modifier, step by step Here's the workflow I use 95% of the time: Open your STL in OrcaSlicer. Place it on the bed. In the object list (top right panel), right-click the part name. Don't right-click in the 3D viewport; that gives you a different menu. Hover "Add modifier." The six-primitive submenu appears. Pick a primitive that roughly matches the region you want to override. For a screw boss, that's a cylinder slightly bigger than the boss. For a hinge knuckle, that's a cube or a torus. The modifier appears at the center of the host part with default dimensions. Click it in the object list or in the viewport to select it. Use the Move (M), Scale (S), and Rotate (R) gizmos to position the modifier where you actually want it. With the modifier still selected, the right-hand panel switches to a per-modifier override view. Hit the "+" next to a setting (or the "Add settings" button in some builds) to add a parameter override. Pick from Quality / Strength / Speed. S [truncated] --- ## OrcaSlicer Spiral Vase Mode: Single-Wall Vases Done Right URL: https://orcaslicer.net/orcaslicer-spiral-vase-mode/ I still remember the first vase I ruined with spiral mode. I'd downloaded a "vase-mode-ready" STL from a public model site, ticked the Spiral vase checkbox in OrcaSlicer, sent it to the printer, and watched the nozzle trace two concentric walls instead of one, then start laying horrible little stringy ribbons between them at every Z step. The model had been pre-hollowed in CAD, and the slicer was doing exactly what I told it to do with what it was given. I've printed dozens of vases, pots, lampshades, and planter sleeves since then, and almost every spiral vase failure I've seen comes down to two things: a model the slicer can't actually spiralise, or a setting the user didn't realise they'd silently overridden. This guide walks through both, with the OrcaSlicer 2.3.x behaviour spelled out, and the open issues that bit me along the way so you don't get caught by them. I'm going to keep this tutorial focused on what's actually true in OrcaSlicer right now, not what blog roundups assumed five releases ago. The internal setting name changed at some point, several "common knowledge" facts about vase mode are inherited from PrusaSlicer and don't quite map across, and there are a couple of bugs in current builds that you should know about before you start tuning flow ratios. I'll flag those as we go. Table of contents What spiral vase mode actually is, and what it isn't Where the setting lives in OrcaSlicer 2.3.x What the slicer forces the moment you tick the box The four sub-settings explained Model design rules the slicer expects Designing your own vase in CAD without breaking spiral mode Layer height, extrusion width, and wall strength Speed, temperature, cooling, and flow rate What breaks the spiral: features that don't work in vase mode Printer-side caveats Troubleshooting common spiral vase failures Known open issues to watch in 2.3.x FAQ What spiral vase mode actually is, and what it isn't Spiral vase mode is a special slicing mode that turns your solid model into a single-walled, continuously rising spiral. Instead of printing each layer as a closed loop and then stepping Z to start the next loop, the printer climbs Z gradually as it traces the single outer perimeter. I think of the result as a screw thread wrapped around the outside of your model. There's no per-layer seam where the loop closes, and there's no Z-hop pause sitting on the wall, because the wall never stops moving. I want to get the history right because it shapes how the feature behaves. Spiral vase mode didn't originate in OrcaSlicer. I'd trace it back to Slic3r years ago, then inherited by PrusaSlicer and Bambu Studio, with OrcaSlicer carrying the same core behaviour plus a "Smooth Spiral" extension that landed in the 1.9 / 1.9.1 rewrite. If you've used vase mode in PrusaSlicer before, the model rules and the failure modes will feel very familiar. I'll get to where the setting key and the UI labels differ, but the underlying idea is identical. What's it for? I lean on it for vases, pots, lampshades, planter sleeves, light diffusers, decorative sleeves, sculptural single-wall objects, and anything where the geometry is essentially "one outer surface rising in Z without internal structure." You get fast prints because there's no infill and no top, you get a clean exterior because there's no Z seam, and you get a slightly translucent wall when you light it from behind, which is why diffusers and lampshades are such a popular use case. What's it not for? I'd rule out anything functional or load-bearing right away. A single perimeter is structurally weak by definition. I also rule it out for anything with holes in the side wall, anything with multiple separate "branches," anything that needs internal cavities or compartments, anything that needs a flat top, and anything that needs supports for overhangs. If you find yourself wishing for any of those, you're not looking at a vase mode print, you're looking at a normal print with thin walls and low infill. Where the setting lives in OrcaSlicer 2.3.x I open the Process settings panel on the right of the OrcaSlicer window. I switch to the Others tab. I scroll down to the Special mode section. I find a checkbox labelled "Spiral vase" there. That's it. Tick it and the slicer immediately applies a cascade of changes that I'll cover in the next section. OrcaSlicer Process settings showing the Others tab with Spiral vase checkbox highlighted I want to flag the internal config key explicitly: it's spiral_mode. If you go reading older PrusaSlicer or Slic3r references, you'll see the key called spiralize, and PrusaSlicer's print-setting variant is called spiral_vase. Neither of those will work as the OrcaSlicer key. If you're editing a profile JSON by hand, or scripting OrcaSlicer through its CLI, the property you want is spiral_mode. I've checked the OrcaSlicer wiki page for the Special Mode section, which confirms that's the variable name and that the description is verbatim "Spiral vase mode transforms a solid model into a single-walled print with solid bottom layers, eliminating seams by continuously spiralling the outer contour." If you don't see the four sub-settings I'm about to cover, your parameter mode is probably set to Simple. I switch to Advanced in the dropdown at the top of the Process panel and the additional fields appear. The base Spiral vase checkbox itself is visible in Simple mode, but Max XY Smoothing and the two flow ratio settings are Advanced-only. Quick history anchor: the spiral codepath was rewritten in OrcaSlicer 1.9 / 1.9.1, and that's when the Smooth Spiral option was added. So if you're on a very old 1.8 or earlier build, the Smooth Spiral checkbox won't be there. I'd update before tuning anything. The rest of this guide assumes you're on 2.3.x. If you're new to the broader slicer and want the parent reference, I'd point you at the OrcaSlicer settings master guide first. What the slicer forces the moment you tick the box I think this is the part that throws people. Spiral vase mode isn't just a single switch. The moment you check the box, OrcaSlicer overrides several other settings and locks them. I'll point you at the Obico deep-dive, which puts it this way: "Spiral mode is accessible only when the wall loop is 1, support is disabled, top shell layer is 0, sparse infill density is 0, and the timelapse type is traditional." In practice, OrcaSlicer enforces those for you when you enable spiral mode, so you don't have to set them by hand. The forced changes are: Wall loops = 1. Only the outer perimeter prints. You cannot have two or three perimeters in spiral mode by definition, because the head is climbing on a single continuous helix. Sparse infill density = 0%. There's nothing to fill. The model is hollow by design. Top shell layers = 0. There's no top. The spiral just ends when it reaches the top of the model. Supports = off. Supports require multiple separate toolpaths per layer, which fundamentally breaks the single-spiral assumption. Timelapse type = Traditional. Smooth timelapse mode interferes with the continuous spiral motion, so it's forced to the simpler Traditional variant. I want to flag one important thing the slicer does not strip: bottom layers. The base of your vase still prints normally with solid bottom layers, and only after the base is complete does the spiral begin. I'd point at the OrcaSlicer wiki, which phrases this as "single-walled print with solid bottom layers." Prusa's docs describe the same behaviour and note that you can adjust the number of bottom solid layers. So if your model has a flat bottom, you'll get a proper sealed base, and then the spiral kicks in for the walls. I think that bottom-layer behaviour is what makes spiral vase mode actually usable for things that hold water, like planters and flower vases, as long as the wall itself is sealed and the bottom layers are clean. I'll come back to wall thickness and watertightness later, because there's nuance there. The four sub-se [truncated] --- ## How to Send G-code Wirelessly From OrcaSlicer: Per-Printer Walkthrough URL: https://orcaslicer.net/orcaslicer-send-gcode-wirelessly/ Last Tuesday I sliced a 14-hour benchy farm, ejected the microSD, walked twelve feet to my P1S, and dropped the card on the desk instead of into the slot because I was looking at my phone. I lost ninety seconds and roughly half my dignity hunting for it under a pile of failed support material, and that was the moment I decided I'd never sneakernet another print as long as the slicer could just send the file itself. If you've been running OrcaSlicer 2.3.x for a while but you're still pulling that card out of the front of a Bambu, a Voron, a Prusa MK4, or a Raspberry Pi taped to the back of an Ender, this guide is the one I wish I'd had when I switched. I'll walk you through every wireless workflow OrcaSlicer supports, ecosystem by ecosystem, with the actual button labels, the actual ports, and the honest gotchas that the marketing pages skip. By the end you'll click Slice once, hit one more button, and the printer will start heating up before you've finished standing up. What you'll find in this guide Why sneakernet feels slow once you've tried wireless OrcaSlicer's upload UI explained Bambu Lab over LAN Bambu Lab over Cloud Klipper printers via Mainsail or Fluidd OctoPrint on a Raspberry Pi PrusaLink for MK4, MK4S, Mini+, Core One File formats: .gcode, .gcode.3mf, .3mf, .bgcode Storage destinations, queues, and post-print behavior Troubleshooting wireless upload FAQ Why sneakernet feels slow once you've tried wireless I timed myself last weekend. The card-shuffle dance, from clicking Export G-code to the first servo whine, takes me about eighty seconds on a good run and well over two minutes if I have to scroll through three folders on the printer's touchscreen to find the file I just exported. It doesn't sound like much, but I print three or four queued jobs a night and I'm not getting any of those minutes back. The microSD card slot on my P1S has also started to feel sticky after maybe two hundred insert cycles, and I'd rather not test how long the contacts hold up under daily abuse. The hidden cost is worse than the time, though. When the slicer talks to the printer directly, OrcaSlicer's Device tab actually does something useful. You get the live camera feed, the temperature graph, the per-print queue, and the ability to pause or cancel from your desk. Yank the card and walk over, and you're back to driving the printer through its own three-inch touchscreen, which is the bit of the user interface every printer manufacturer cares about least. The catch is that OrcaSlicer doesn't run a single magic wireless protocol. It runs four. Bambu Lab printers use MQTT and FTPS over your LAN, with an optional cloud path through Bambu's own servers. Klipper rigs (Voron, Sovol SV0x, Sonic Pad, unlocked K1, anything running Mainsail or Fluidd) accept files through Moonraker's OctoPrint-compatible shim. Classic OctoPrint installs on a Pi attached to a Marlin printer use OctoPrint's REST API. Prusa MK4 family printers and the Mini+ use PrusaLink, which is its own modern REST API with HTTP Digest auth. Each ecosystem has its own button labels in the OrcaSlicer setup dialog, its own file format preferences, and its own gotcha about whether "uploaded" actually means "printing." Get them confused and you'll spend an hour wondering why your G-code is sitting on the printer doing nothing. The rest of this article walks each one end to end. Pick the section that matches your printer and you'll be wireless in fifteen minutes or less. OrcaSlicer's upload UI explained, button by button Before we touch any single printer, let's map out what the slicer is showing you. OrcaSlicer's main window has four top-level tabs across the top: Prepare, Preview, Device, and Project. You probably live in Prepare. Wireless printing happens in Preview (where the Upload buttons appear after a slice) and Device (where the live connection lives). When you click Slice all, the right-hand panel switches into Preview mode and surfaces a button cluster. The third-party Kingroon write-up of the Device tab describes the two upload options as "Export the G-Code File" and "Upload&Print," noting that the second one "directly sends the file to your printer and starts the print if you made the Wi-Fi connection in OrcaSlicer." That's the key word: connection. None of those upload buttons do anything wireless until you've added the printer as a "physical printer" with a host type and an address. Post-slice button cluster with Upload and Upload&Print options To add a connection, look for the small Wi-Fi icon next to the printer name at the top of the Prepare tab. Click it and OrcaSlicer opens the physical printer connection dialog. The single most important field is the Host Type dropdown. The values you'll see, again per Kingroon's overview of the Device tab, include OctoPrint, Mainsail, Fluidd, PrusaLink, PrusaConnect, Duet, FlashAir, AstroBox, Repetier, MKS, FlashForge, SimplyPrint, and Obico. There's also a special entry labeled "Octo/Klipper" for Mainsail and Fluidd users, which the official Mainsail docs flag explicitly in their OrcaSlicer integration guide. Underneath the Host Type, you'll fill in the printer's hostname or IP address (e.g. http://printer.local or http://192.168.1.100) and an API key or access code, depending on the host. There's a Test button at the bottom of the dialog. Use it. If Test doesn't go green, nothing else in this article will work, and you'll save yourself a confusing forty-five minutes by fixing the connection now instead of after a failed upload. Once the connection is saved, the Device tab at the top of the OrcaSlicer window comes alive. Kingroon's writeup describes it as "an integrated window inside the slicer which lets you control your 3D printer directly from the OrcaSlicer." For Bambu printers you'll get the famous camera feed, temperature gauges, and the file list straight off the printer's storage. For Klipper, OctoPrint, and PrusaLink, the Device tab embeds the printer's own web interface, which means you've got a one-tab cockpit instead of juggling browser windows. OrcaSlicer 2.3 Device tab showing connected printer with live camera feed One thing worth calling out now, because it bites people on every host: the buttons in the post-slice cluster behave subtly differently. The "Export G-code File" option just writes a .gcode file to your computer; nothing leaves your laptop. "Upload" sends the file over Wi-Fi but does not start the print. "Upload&Print" sends the file and tells the host to begin printing the moment the upload finishes. If you've ever clicked the wrong one and stared at your printer wondering why it isn't doing anything, you know exactly the moment I'm talking about. I'll come back to this in the troubleshooting section because it's the single most common "wireless isn't working" complaint, and it's almost always button confusion. Now let's go printer by printer. Bambu Lab over LAN (X1C, P1S, P1P, A1, A1 Mini, P2S) Bambu's the most polarising of the four ecosystems because their 2025 firmware story rattled a lot of people, but the LAN flow itself is genuinely fast once it's set up. Here's the honest version, step by step. First, put the printer into LAN Only Mode. On the touchscreen, head to Settings, then General or Network depending on your model, and toggle LAN Only Mode on. Per the community-maintained Bambu LAN mode guide, this "logs you out and supposedly disconnects your printer from their servers, but since it is still connected to WiFi it theoretically could call home if it wanted." The point is that LAN mode breaks the cloud account binding so OrcaSlicer can talk to the printer directly without going through Bambu's servers. Second, and this is the part that's tripped up the most people since early 2025, enable Developer Mode. Bambu's own wiki page on enabling Developer Mode walks through the menu path. The reason this matters is that in early 2025 Bambu introduced an authorisation control system and a piece of middleware they call Ba [truncated] --- ## OrcaSlicer Webcam and Live View Setup for Remote Monitoring URL: https://orcaslicer.net/orcaslicer-webcam-live-view/ I came home from a Costco run last winter to find half a spaghetti pile glued to the bed of my Ender 3 because the wife had bumped the printer off the corner of the bench and I couldn't see a thing from the kitchen upstairs. That was the night I went down a webcam rabbit hole and didn't come up until I'd wired a Pi camera to every printer in the garage, including the cheap Sovol that didn't deserve one. So when people ask me how to get "Live View" working inside OrcaSlicer, I'm sympathetic, because the answer is messier than anybody on YouTube wants to admit. Here's the part nobody says out loud: "Live View in OrcaSlicer" isn't one feature. It's three different things stacked on top of each other, and which one you get depends entirely on what brand of printer you've got plugged in. I'll walk through every flavor I've actually used, what works, what silently doesn't, and the handful of settings that fix 80% of "my webcam is black" posts on Reddit. Table of contents What "Live View" actually means inside OrcaSlicer The Device tab: Bambu native vs every other brand Bambu Lab webcam setup, by model Klipper printers, the crowsnest way Going H.264 with camera-streamer for low-bandwidth video OctoPrint webcam routes, legacy and modern PrusaLink and Prusa Connect, the live-view gap IP cameras, RTSP, and the go2rtc bridge Bandwidth, latency, and storage considerations Troubleshooting and security Frequently asked questions What "Live View" actually means inside OrcaSlicer Before we touch a single setting, you've got to internalize this: when somebody says "I want Live View in OrcaSlicer," they're really asking for one of three things, and the troubleshooting flow forks immediately based on which one. I've watched friends spend two hours fighting the wrong setup because they didn't realize their printer didn't support the path they were following. The first flavor is the Bambu-native Device tab feed. On an X1C, X1E, P1S, A1, or any other Bambu printer, OrcaSlicer's Device tab talks to Bambu's own LAN/cloud pipeline and shows a real, decoded video frame. It's the closest thing to a "just works" experience in this whole space. You sign in, you pick the printer, the feed appears, you move on with your life. The second flavor is the embedded WebKit panel that loads on basically every non-Bambu printer. When you connect to a Klipper machine running Mainsail or Fluidd, or to an OctoPrint instance, OrcaSlicer doesn't decode the video itself. It opens a browser window inside the Device tab and points it at the printer's web UI. Whatever Mainsail or OctoPrint is showing in your browser is what you'll see in OrcaSlicer. If the printer's webcam panel works in Chrome, it'll usually work here. If it doesn't, OrcaSlicer can't save you. The third flavor is the custom HTTP camera URL field where you paste an MJPEG or MP4 stream directly. This is the option people forget exists, and it's also the one that catches them out most, because the embedded player has hard limits on what codecs it'll render. OrcaSlicer maintainers have been clear in Discussion #2548 that MJPEG and MP4 over HTTP are the supported formats. If the link opens natively in a normal browser, it usually plays in OrcaSlicer too, because under the hood it's still a webview. What absolutely doesn't work in that field is HLS, RTSP, or WebRTC. Issue #4358 has multiple users confirming that those three codecs simply don't render inside OrcaSlicer's Device tab, even when they play fine in Fluidd or VLC in another tab. That's not a bug you can fix from your side, and chasing it will eat your weekend. Plan around it from the start. OrcaSlicer Device tab showing Mainsail embedded with the webcam panel live The Device tab: Bambu native vs every other brand I want to spend a minute on why the Device tab behaves so differently from one printer to the next, because once you grok this, every weird "why is my screen blank" moment makes sense. The tab itself is a chameleon. When it sees a Bambu printer in your network list, it switches into a native renderer that knows how to talk to the X1C's RTSPS endpoint or the P1S's custom TCP socket. When it sees a Klipper, OctoPrint, PrusaLink, FlashForge, Repetier, or Obico host, it falls back to embedding the host's web UI inside a WebKit panel. That second mode is where most of the rough edges live. WebKit on Windows is fine. WebKit on macOS is fine. WebKit on Linux, especially when OrcaSlicer is running from an AppImage, has a documented rendering bug where the embedded Mainsail or Fluidd UI freezes mid-load. Issue #6043 captured the exact behavior on Fedora 40: Mainsail shows its initializing bar and never finishes, Fluidd renders partial UI and stops. The workaround in that thread is exporting WEBKIT_DISABLE_DMABUF_RENDERER=1 before launching the AppImage. I've used it personally on a couple of Linux laptops and it does clear the freeze, but you have to remember to set it every time, or wrap the AppImage in a launcher script. Here's the other quirk worth flagging early. When the embedded panel does load Mainsail or Fluidd correctly, the webcam stream you see is whatever the host is configured to show. That means if you fix your crowsnest config or your OctoPi camera stack, it'll show up in OrcaSlicer too, without touching OrcaSlicer at all. The slicer is a passenger. Don't waste time digging into OrcaSlicer settings when the real problem is upstream. For Bambu printers the story is different because the slicer is actively decoding, but you've still got a configuration prerequisite. The X1C, X1E, and H2D need their LAN-Only Liveview toggle enabled in the printer's network settings, otherwise the LAN feed silently refuses connections. The P1S, P1P, A1, and A1 mini have a "Video (enabled)" toggle in similar settings menus depending on firmware. If your Bambu feed shows a black frame in OrcaSlicer, that toggle is the first thing to check. Bambu Lab webcam setup, by model The Bambu lineup is where most of the dangerous misinformation lives, because the printers look superficially similar but their cameras work in completely different ways under the hood. I'll go printer by printer so you don't end up trying to point a third-party RTSP client at a P1S and wondering why nothing connects. X1 / X1C / X1E chamber camera The X1-series chamber camera is the gold standard of stock Bambu cams. It's a 1920x1080 sensor running at 30 fps with a 110 degree field of view and an F2.6 aperture. On the X1 Carbon and X1E it's included by default. On the original X1 it was an optional upgrade kit. The camera supports three documented functions out of the box: remote livestream, AI spaghetti detection, and time-lapse capture. If you're running an X1-series printer, that's what's behind your Device tab feed. For LAN-only access, including pulling the feed into Home Assistant or any third-party viewer, the X1C, X1E, and H2D expose an RTSPS endpoint on port 322. The URL format is rtsps://<ip>:322/streaming/live/1. The username is bblp. The password is the access code you'll find in the printer's LAN Mode settings screen. This endpoint requires firmware 01.06 or newer, and you have to enable the LAN-Only Liveview switch in the printer's network menu. That switch is independent of LAN Mode itself, which trips people up. You don't have to fully enable LAN Mode to use the LAN-Only Liveview feed. General with the LAN Mode Liveview switch enabled" class="wp-image" loading="lazy" /> Bambu Studio LAN-Only Liveview toggle on X1C screen I use VLC to sanity-check the RTSPS URL before I bother routing it anywhere else. If VLC opens the stream with the bblp username and the access code, the printer side is fine and you can move on to whatever client you actually want to use it from. If VLC can't connect, you've got a firmware version problem, the toggle is off, or your network has a routing issue between client and printer. P1S and P1P built-in camera This is where the wheels come off the "just point an RTSP cli [truncated] --- ## OrcaSlicer Print Time Estimate Wrong: Why and How to Fix URL: https://orcaslicer.net/orcaslicer-print-time-estimate-wrong/ My X1C said 4 hours and 12 minutes. I came back six hours and forty minutes later to a print that still had two top layers to go. That kind of miss isn't a rounding error, it's the difference between starting a print before bed and waking up to a half-finished benchy stuck on the plate. I've been chasing this same problem on three different printers (a Vyper running stock Marlin, a Voron 2.4 on Klipper, and the X1C), and I've read every relevant GitHub thread on the OrcaSlicer tracker since late 2023. The honest answer is that "OrcaSlicer's print time estimate is wrong" is actually four different problems wearing the same shirt, and you can't fix any of them until you figure out which one is yours. So this guide walks you through the diagnosis first, then the fix per printer family, then the one nuclear option that gets Klipper users within a minute or two for a 12-hour print. If you've got a Bambu, I'll be straight with you about what the slicer can and can't do for the on-printer ETA. Why your OrcaSlicer print time is wrong (the 30 second answer) How OrcaSlicer actually calculates print time The five reasons the estimate is wrong (diagnostic flowchart) Per printer tuning matrix Step by step: align Motion Ability with your firmware Tune Max Volumetric Speed so the slicer isn't lying about throughput Use klipper_estimator for sub minute accuracy (Klipper users only) Why the printer's display ETA may still disagree When to stop tuning FAQ Why your OrcaSlicer print time is wrong (the 30 second answer) If you only have thirty seconds, here's the version. The estimate you see in OrcaSlicer's Preview tab is a kinematic prediction. It assumes your printer accelerates exactly to the values in Motion Ability, decelerates symmetrically, hits cruise speed cleanly, and never gets bottlenecked by the hotend. In reality, three things knock that prediction sideways. The first is slicer math drift. Your Motion Ability values don't match what the firmware actually does. A user on an Anycubic Vyper documented this beautifully in Issue #7576 back in November 2024: their print took 2 hours and 10 minutes, but Orca predicted 7 hours and 30 minutes. That's roughly three to four times over. They noted the same file in OctoPrint matched actual print time within a few minutes, so the G-code was fine, the slicer's prediction wasn't. The second is runtime throttling. The slicer thinks you can extrude at the Max Volumetric Speed you set, but the hotend can't keep up, so the printer's motion planner slows the extruder. This is the inverse failure mode, where a Klipper user in Issue #1612 reports prints running 20 to 30 percent longer than the slicer estimate. Two opposite symptoms, two opposite causes, one shared root: the slicer's math assumed something the printer can't deliver, or refused to do, at runtime. The third is firmware side ETA. On Bambu and AnkerMake, the number on the printer's screen isn't the same number Orca calculated. The printer recalculates its own ETA based on its own kinematic profile, sensor feedback, and (on Bambu) firmware-side tuning that overrides slicer choices. The most extreme example I've seen is Issue #9457: a 50 minute print sliced in OrcaSlicer 2.3.0 displayed as somewhere between 58 and 73 hours on an AnkerMake M5C's screen. The slicer was right, the firmware's parser broke. Three buckets, three different fixes. OrcaSlicer Preview tab showing estimated print time vs printer LCD showing actual elapsed time If you're already overwhelmed, just remember this: the diagnostic flow is "is the slicer's number wrong, is the printer's runtime wrong, or is the printer's display wrong?" Each of those three has a different fix path, and mixing them up wastes hours. The rest of this article is the long version. If you want the full troubleshooting context, the OrcaSlicer troubleshooting master guide covers the wider class of issues this one belongs to. How OrcaSlicer actually calculates print time Before you can fix the wrong number, you've got to understand what the slicer is actually doing when it generates that number. I see a lot of confusion online where people assume Orca runs some kind of machine-learning estimator or "smart ETA" that gets better with each print. It doesn't. The estimate is a deterministic kinematic calculation that runs once when the slice finishes, and it never updates again. The kinematic model in plain English For every move in the G-code, OrcaSlicer asks: how long would this segment take if the toolhead accelerates from its entry speed up to the cruise feedrate, holds that feedrate for as long as the segment is long enough, then decelerates to its exit speed? The acceleration value comes from your Motion Ability settings. The cornering behaviour comes from either jerk (Marlin classic mode) or junction deviation (Marlin 2 with JD enabled, or by analogy with Klipper's square corner velocity). The cruise feedrate is whatever the speed setting for that feature is, capped by the per-axis max feedrate. Then it sums every segment's duration. That's the number you see in the Preview tab. The model is simple, well understood, and accurate, as long as the inputs match what the printer actually does. What it consumes from Motion Ability The Motion Ability section of Printer Settings is where you tell Orca what your machine can physically do. Per the OrcaSlicer Motion Ability wiki, the verbatim setting names you'll find here are machine_max_acceleration_extruding, machine_max_acceleration_retracting, machine_max_acceleration_travel (the travel one only shows in Advanced mode), the per axis caps machine_max_acceleration_x, _y, _z, _e, the per axis jerks machine_max_jerk_x, _y, _z, _e, and (in Advanced mode) machine_max_junction_deviation. The wiki promises that these "safeguard maximum accelerations for all axes" and that this "will cap the acceleration set by the process if it exceeds these values." That's the design intent. The jerk and junction deviation entries are documented as M205 commands: jerk is "M205 X, Y, Z, E, only apply if JD = 0 for Marlin 2 Firmware," and junction deviation is "M205 J, only apply if JD > 0 for Marlin Firmware. If your Marlin 2 printer uses Classic Jerk set this value to 0." Translation: pick one cornering model and stick with it. If you mix jerk and junction deviation, the slicer's estimate is going to drift from reality. Motion Ability with machine_max_acceleration_extruding and machine_max_jerk_x fields visible" class="wp-image" loading="lazy" /> OrcaSlicer Printer Settings tab with Motion Ability section expanded The print side accelerations sit in a different panel: the Speed section of Print Settings. Per the OrcaSlicer speed settings acceleration wiki, you've got default_acceleration, outer_wall_acceleration, inner_wall_acceleration, bridge_acceleration, sparse_infill_acceleration, internal_solid_infill_acceleration, initial_layer_acceleration, initial_layer_travel_acceleration, top_surface_acceleration, and travel_acceleration. The wiki notes that these "will limit the acceleration to not exceed the acceleration set in the Printer's Motion Ability settings," and that "Acceleration in 3D printing is usually set on the printer's firmware settings. This setting will try to override the acceleration when normal printing acceleration value is different than 0." So the print side accelerations propose, the Motion Ability disposes. If the print side asks for more than Motion Ability allows, the lower number wins, and the estimate is calculated against the lower number. What it does not account for This is the part that catches everyone. The slicer's estimate ignores: nozzle and bed heat-up time, homing, bed leveling and mesh probing, slicer-driven macros, input shaping smoothing (which can subtly alter actual move durations on Klipper), and the Klipper minimum_cruise_ratio behaviour which slows down short moves. I'll come back to that last one because it's the single biggest reason Klipper estimates ru [truncated] --- ## OrcaSlicer Slow Slicing on Complex Models: How to Speed Up URL: https://orcaslicer.net/orcaslicer-slow-slicing-fix/ I lost an entire Friday evening to a single 240 MB STL of an articulated dragon a buddy sent me. I dropped it onto the plate in OrcaSlicer 2.3.1, hit slice, and watched the progress bar limp past 60 percent before stalling at 75 for what felt like forever. When I checked back after dinner, the slicer had eaten 14 GB of RAM, the fans were screaming, and my mouse cursor was lagging. I killed the process, swore at my desk, and reopened the same file in PrusaSlicer just to confirm I wasn't crazy. PrusaSlicer chewed through it in about 40 seconds. I'm not telling you that story to bash OrcaSlicer. I use it daily, it's still my main slicer, and the calibration tooling alone is worth the friction. But I am telling you that if you're reading this because OrcaSlicer just took 11 minutes to slice something it should have done in 30 seconds, you're not imagining it, you're not doing anything wrong, and there are concrete things that actually move the needle. Some of them are settings. Some are machine-side. A few are documented bugs that you can't fix and just need to know about so you stop wasting hours blaming yourself. This is the long, honest version of "how to make OrcaSlicer slice faster on big models." I'll point you at the GitHub issues that match your symptom, the verbatim setting names that matter, the fixes that work today, and the things people on Reddit insist will help that genuinely won't. No fake benchmarks, no "10x speedup" promises. Just what I've actually done to recover my own evenings. Why OrcaSlicer slows to a crawl on big models Diagnose first, figure out which kind of slow you have Settings you can change today inside OrcaSlicer Reduce the model itself before it ever hits the slicer Machine-side fixes that actually matter When the bug isn't yours, known OrcaSlicer 2.3.x slow-slicing issues Workflow tactics that compound What OrcaSlicer 2.3.x release notes actually changed Realistic expectations and what NOT to chase FAQ Why OrcaSlicer slows to a crawl on big models Before I tell you what to change, I want to give you a working mental model of where the time actually goes. I see a lot of people online tweaking exotic settings (arc fitting, ironing patterns, infill anchors) hoping for a slicing-speed boost, when the real time sink is somewhere else entirely. Once you can roughly tell which phase is eating your minutes, you can target the fix. The four phases of a slice and where time actually goes OrcaSlicer's slice job, simplified, runs four heavy phases back to back. First, it slices the triangle mesh into 2D layers (this is where high triangle counts hurt). Second, it generates perimeters and walls per layer, where Arachne does extra dynamic-width math that Classic doesn't. Third, it does support generation, which on the Organic style is genuinely expensive, especially in 2.3 (more on that below). Fourth, it generates infill and the actual G-code paths, where Resolution and arc fitting come into play. If your slice freezes or stalls in the first 25 percent, you're almost always looking at a triangle-count or mesh problem. If it stalls between 30 and 70 percent, supports or walls are usually the culprit. If it hangs near the end, it's typically infill or G-code generation choking on density and pattern. I keep an eye on the progress bar and Task Manager together now, because that combo tells me which lever to pull. The four-tab layout in OrcaSlicer 2.3.2 covers everything you do, from importing the model to generating the G-code. Knowing which tab the work happens in helps you guess where time is going. Single-threaded reality, what's parallel and what isn't Here's the honest part nobody likes hearing. OrcaSlicer's slicing core is largely single-threaded for many phases. Some sub-tasks parallelize across cores, but the critical path through a slice job spends a lot of wall-clock time on a single thread. That's why, when you watch Task Manager during a long slice, you'll often see one core pinned to 100 percent while the other 23 cores on your fancy threadripper sit at 4 percent each. That's not a misconfiguration on your end. It's just where the code is right now. This is also why "Slice All" across multiple plates runs sequentially today. There's an open feature request, issue #11991, asking for parallel multi-plate slicing, with the OP pointing out that "slicing distinct plates is a perfect candidate for parallelization because Plate A's G-code generation has zero dependency on Plate B." It's open, no ETA, and until then your only option is a manual workaround (covered later). GPU acceleration is a preview thing, not a slice-speed thing I've watched people spend a hundred bucks upgrading their GPU because they read somewhere that OrcaSlicer benefits from it. Don't. The GPU in OrcaSlicer is used for the 3D viewport (rotating the model, the layer-preview slider, that kind of thing), not for the slicing math itself. A faster GPU will make the preview tab smoother, especially on big plates with lots of objects. It will not shave seconds off the actual slice operation. If your slicing speed is the bottleneck, money is better spent on a faster CPU with strong single-thread performance, or just more RAM. Diagnose first, figure out which kind of slow you have I learned this the hard way. The settings that fix "frozen at 75 percent" are completely different from the settings that fix "stalls partway through every re-slice." Pattern-match your symptom first, then go to the relevant section. Otherwise you'll change ten things, the slice will still be slow, and you'll have no idea which of those ten was actually relevant. Does it freeze at a specific percentage? If your slice climbs to 75 percent, sits there for a long time, and eventually either crashes or gets killed by your OS, you're almost certainly hitting memory exhaustion. There's a named precedent for this: GitHub issue #9094, opened 26 March 2025 against OrcaSlicer 2.3.0. The reporter was on Arch Linux with 31.3 GiB of RAM, and the symptom was exactly that: "halts at 75% during slicing and consumes memory until the operating system terminates the process." That issue was closed, linked to PR #9115 as the fix, so first thing to check is whether you're on a build that includes that PR. If you're on 2.3.0, update. If you're already on 2.3.2 and still seeing the 75-percent freeze, it's probably a different memory pressure issue specific to your model. The most common cause I see is a multi-million-triangle mesh with organic supports turned on. We'll get to fixing that. For broader symptoms (slice never completes, prints stop mid-job, etc.) I've collected related cases in the print stops mid-print troubleshooting writeup. Does CPU usage sit at 20 percent instead of 90-plus? If you watch Task Manager during the slice and your CPU is at 15 to 25 percent instead of pinning a core, that's a scheduling-or-state problem, not a missing-horsepower problem. Two known scenarios. One: you've already sliced the model once, then changed layer height (or another major setting) and re-sliced. Issue #9397 documents exactly this. The reporter's initial slice took about 47 seconds; subsequent slices after settings changes took "2+ minutes or longer," with CPU utilization dropping from 90-100 percent to about 20 percent. The OP's workaround, which I've confirmed on my own machine, is the simplest fix in this whole article: save the project, close OrcaSlicer entirely, reopen it. Then re-slice. The reporter wrote that "reloading the project restores normal performance." It does. Two: you're on a 12th or 13th gen Intel CPU with E-cores enabled and you're slicing a complex model. Issue #8227 covers this case explicitly: "Orca Slicer will consistently trigger errors and crash when slicing complex models (>500000 triangles) with efficiency cores enabled." We'll cover the exact mitigations in the machine-side section, but if you're on Windows with an Intel hybrid CPU and your slice keeps crashing on big files, this [truncated] --- ## OrcaSlicer for Qidi X-Plus 3 / X-Max 3 / Q1 Pro: Setup and Best Settings (2026) URL: https://orcaslicer.net/orcaslicer-qidi-x-series/ I bought a Qidi X-Plus 3 to print PA-CF brackets after my Voron toasted its second hot end on the same job. The chamber heater was the entire point: 65°C of active heat lets nylon and CF blends actually anneal as they print, instead of warping into a banana. Most Qidi guides skip past the OrcaSlicer side, and they all get the Moonraker port wrong. This guide is for owners (or near-owners) of the three Qidi machines that share the same toolhead philosophy and roughly the same OrcaSlicer workflow: the Q1 Pro, the X-Plus 3, and the X-Max 3. I'll cover what's already in OrcaSlicer for these printers, how to wire them up over Klipper and Moonraker without losing an evening, what filament settings actually work past the marketing claims, and the ugly bits nobody else mentions (chamber heater G-code mismatches, send-and-print 403 errors, the 600 mm/s number that isn't real). If you're brand new to OrcaSlicer in general, bookmark our printer profiles hub and the Klipper setup walkthrough. Both come up repeatedly here because Qidi runs a Klipper fork, and the Moonraker quirks are universal. Table of contents Which Qidi belongs in your shop Profiles already in OrcaSlicer First-time setup walkthrough Network and Moonraker (port :10088) Best filament settings Chamber heater in OrcaSlicer (the ugly part) Calibration sequence on Qidi Common pain points and fixes OrcaSlicer vs Qidi Studio FAQ Which Qidi belongs in your shop Qidi has thinned its lineup to a small, focused range of enclosed CoreXY Klipper machines. The three I keep recommending are the Q1 Pro, the X-Plus 3, and the X-Max 3. They share a hot end, a firmware base, and almost all of their OrcaSlicer settings. What changes between them is build volume, chamber temperature ceiling, and price. Here's how they actually compare in 2026: SpecQ1 ProX-Plus 3X-Max 3 Build volume (mm)245 x 245 x 240280 x 280 x 270325 x 325 x 315 KinematicsEnclosed CoreXYEnclosed CoreXYEnclosed CoreXY FirmwareKlipper (Qidi fork)Klipper (Qidi fork)Klipper (Qidi fork) Max claimed speed600 mm/s600 mm/s600 mm/s Acceleration20,000 mm/s²20,000 mm/s²20,000 mm/s² Max hot end temp350°C350°C350°C Max bed temp120°C120°C120°C Active heated chamberYes, up to 60°CYes, up to 65°CYes, up to 65°C Stock hot endBimetal, swappable HSBimetal, swappable HSBimetal + hardened steel both included Stock nozzle0.4 mm0.4 mm0.4 mm Other nozzles sold0.2 / 0.6 / 0.80.2 / 0.6 / 0.80.2 / 0.6 / 0.8 Camera1080p built-inAdd-onAdd-on Approx US price (May 2026)~$449 to $599~$899 to $1099~$899 to $999 (sale) I want to be honest about the value story, because it's not the same on every model. The Q1 Pro at roughly $449 to $599 is the headline. It's the cheapest CoreXY with an active heated chamber that I'd actually trust with ABS and ASA all day. The chamber tops out at 60°C, which is enough for the warpy stuff most people care about, and it makes the Q1 Pro genuinely interesting against a Bambu A1 or P1S if you've ever tried to print ASA in an open frame and watched the corners curl. The X-Plus 3 sits in the middle. You're paying $899 to $1099 for an extra 5°C in the chamber (65°C versus 60°C), a meaningfully bigger build (280 x 280 x 270 versus 245 x 245 x 240), and the same toolhead. Those 5°C don't sound like much, but they matter when you're printing PA-CF taller than 50 mm. I've had nylon parts that warped at 60°C chamber and stayed flat at 65°C. It's a real difference. The X-Max 3 is the same chamber temperature as the Plus 3, just bigger. 325 x 325 x 315 mm of build is enough for full-size drone frames, structural brackets, and plates that don't need sectioning. If you're not printing big parts, you don't need it. If you are, nothing else in this price range does it with a 65°C active chamber. Here's the part Qidi won't tell you: if you only print PLA and the occasional PETG, none of these are the right printer. A Bambu A1 mini will do everything you need for a third of the money. The X-series only earns its price tag if you're going to actually print PA, PC, ASA, ABS, or carbon-filled blends. That's the entire pitch. Don't buy a chamber-heated CoreXY to print Benchies. Material capability per machine looks like this: Filament familyQ1 ProX-Plus 3X-Max 3 PLA / PETGOverkill but fineOverkill but fineOverkill but fine ABS / ASAExcellentExcellentExcellent PC / PC-CFGood (small parts)ExcellentExcellent PA / NylonGood (small to medium)ExcellentExcellent PA-CF / PAHT-CFGood (under 50 mm tall)ExcellentExcellent (large parts) PET-CFGoodExcellentExcellent TPU 95A+Direct drive, fineDirect drive, fineDirect drive, fine Profiles already in OrcaSlicer Good news. Upstream OrcaSlicer ships with first-party profiles for every Qidi printer in this article and most others in Qidi's catalog. You don't need to download anything, copy any JSONs, or maintain a fork. Qidi is an early sponsor of the OrcaSlicer project, which is why their machines get treated as a first-class vendor instead of a community add-on. If you crack open the OrcaSlicer GitHub repo at resources/profiles/Qidi/machine/, here's what's there for our three machines: X-Plus 3: Qidi X-Plus 3.json (parent profile) Qidi X-Plus 3 0.2 nozzle.json Qidi X-Plus 3 0.4 nozzle.json Qidi X-Plus 3 0.6 nozzle.json Qidi X-Plus 3 0.8 nozzle.json X-Max 3: Qidi X-Max 3.json (parent profile) Qidi X-Max 3 0.2 nozzle.json Qidi X-Max 3 0.4 nozzle.json Qidi X-Max 3 0.6 nozzle.json Qidi X-Max 3 0.8 nozzle.json Q1 Pro: Qidi Q1 Pro.json (parent profile) Qidi Q1 Pro 0.2 nozzle.json Qidi Q1 Pro 0.4 nozzle.json Qidi Q1 Pro 0.6 nozzle.json Qidi Q1 Pro 0.8 nozzle.json The parent profile holds the motion limits, kinematics, and bed shape. The per-nozzle JSONs override line widths, max volumetric flow, and minimum layer time defaults so you don't have to hand-tune those when you swap nozzles. If you've moved between hot ends or nozzle sizes on a Qidi before, this is much nicer than the manual approach. Filament profiles live in resources/profiles/Qidi/filament/. Qidi-tuned filaments cover Generic PLA, ABS, ASA, PETG, TPU, PA-CF, PC, PAHT-CF, and PET-CF. They're sane starting points and they assume the right chamber temps for engineering filaments, which is more than I can say for half the community profiles I've seen on Reddit. Process profiles (the slicing presets, layer height plus speeds plus walls) sit in resources/profiles/Qidi/process/. The defaults track Qidi's own tuning, so you can pick "0.20 mm Standard" for a Q1 Pro and get something close to what Qidi Studio would give you out of the box. If you're new to OrcaSlicer's profile model, our printer profiles hub walks through how machine, filament, and process inheritance fits together. The short version: don't edit the parent. Duplicate, rename, then edit your copy. Generic OrcaSlicer interface with a printer profile and Plate Settings open. Qidi X-series users follow the same workflow once their printer profile is selected from the vendor list. First-time setup walkthrough Setup on a fresh OrcaSlicer install for any of these three machines takes about ten minutes if nothing fights you. Here's the order I run through every time. Open OrcaSlicer. If this is your first launch, the Welcome wizard will appear. If you've used Orca before, head to the printer dropdown in the top-left and pick "Add Printer" to get the same wizard. Expand the Qidi vendor entry. You'll see the entire current Qidi catalog (Q1 Pro, Q2, Q2C, X-Plus 3, X-Plus 4, X-Max 3, X-Max 4, X-Smart 3, plus the older X-CF Pro and X-Plus / X-Max originals). Tick your machine, tick the nozzle sizes you actually have (0.4 mm at minimum), and confirm. Now select your fresh printer profile in the Prepare tab dropdown. Pick a starting filament (Generic PLA or the Qidi-branded equivalent for your spool) and a process preset like 0.20 mm Standard. That's enough to slice a model and export a 3MF you can sneakernet to the printer over USB or SD. If that's [truncated] --- ## OrcaSlicer for Sovol SV06 / SV07 / SV08: Setup and Best Settings (2026) URL: https://orcaslicer.net/orcaslicer-sovol-sv-series/ I bought an SV08 last year because it was the cheapest 350mm CoreXY I could find. Forty thousand mm/s² accel claim, $700, ships with Klipper preconfigured, sounded too good to be true. The taco-bed-warp horror stories from Reddit had me scared, and the OrcaSlicer setup tripped me up because most online guides demanded an API key the printer doesn't actually need. I've since added an SV06 to a small print farm and helped a friend dial in his SV07 Plus, so this guide covers the whole Sovol SV-series in OrcaSlicer: SV06 and SV06 Plus on stock Marlin, SV07 and SV07 Plus on Klipper, and the SV08 / SV08 MAX CoreXY flagships. Three printers, two firmwares, one slicer. I'll show you the setup paths, the best-settings starting points per variant, the calibration sequence that actually matters, and the real forum-documented gotchas that wasted weekends for everyone before you. This article lives under our OrcaSlicer Printer Profiles Hub. If you're cross-shopping printers, that hub has the comparison context. Table of contents Sovol SV-series at a glance Marlin vs Klipper: why this matters in OrcaSlicer Adding your Sovol printer in OrcaSlicer Setup path A: SV06 / SV06 Plus (Marlin) Setup path B: SV07 / SV07 Plus (Klipper) Setup path C: SV08 / SV08 MAX (Klipper CoreXY) Best OrcaSlicer settings per variant Calibration sequence for Sovol Klipper printers Direct drive on every SV: what changes Common Sovol + OrcaSlicer problems and fixes Filament-specific notes (PLA, PETG, TPU) FAQ Sovol SV-series at a glance I keep telling people the Sovol lineup makes more sense once you split it by firmware, not by name. The SV06 family runs stock Marlin. The SV07 family runs Klipper on a bed-slinger. The SV08 family runs Klipper on CoreXY. Same slicer profiles in OrcaSlicer, very different ceilings. SpecSV06 / SV06 PlusSV07 / SV07 PlusSV08 / SV08 MAX Build volume220×220×250 / 300×300×340220×220×250 / 300×300×350350×350×345 / large-format Firmware (stock)MarlinKlipperKlipper MotionBed-slinger CartesianBed-slinger CartesianCoreXY (quad-gantry leveled) Claimed max speed150 mm/s500 mm/s (250 realistic)700 mm/s (200-300 realistic) Max acceleration~3,000 mm/s²~10,000 mm/s²40,000 mm/s² (claim) HotendAll-metal, 300°CAll-metal high-flow, 300°CCeramic-heated all-metal high-flow ExtruderDual-gear direct driveDual-gear direct drivePlanetary dual-gear direct drive NetworkUSB / SD onlyWiFi 2.4 GHzWiFi + LAN, Mainsail UI Approx price (USD)$250 - $310$300 - $500$529 - $1000+ Honest take from someone who owns or has tuned all three: SV06 / SV06 Plus is the budget, open-source, hackable pick. It's slow but it's reliable, and the parts are everywhere. Print farm starter or first-printer territory. SV07 / SV07 Plus is the Klipper sweet spot if you want network printing without spending CoreXY money. It's still a bed-slinger though, so the 500 mm/s marketing number is mostly Klipper bragging. Real-world I land around 150-250 mm/s. SV08 / SV08 MAX is the speed flagship. Voron-derived CoreXY, quad gantry, eddy-current bed scan on later units. It's the one to buy if speed matters and you're willing to do an input-shaper run. Marlin vs Klipper: why this matters in OrcaSlicer The biggest invisible split in this lineup is firmware. OrcaSlicer profiles don't care, but your settings ceiling does. Marlin (SV06 / SV06 Plus stock)Klipper (SV07, SV07 Plus, SV08, SV08 MAX) Pressure AdvanceNot available (use Linear Advance K if FW supports)Yes, tune in Klipper, save per filament Input ShaperNot availableYes, ADXL345 or onboard accel, run SHAPER_CALIBRATE Practical print speed80-120 mm/s150-300 mm/s depending on model Network printingSD / USB / OctoPrintMoonraker / Mainsail / Fluidd OrcaSlicer connectionPhysical Printer = OctoPrint (if Pi present)Physical Printer = Klipper (Moonraker) If you flash Klipper to an SV06 (popular Voron-style mod), you can use the SV07 settings as a baseline. That's a different rabbit hole, and our OrcaSlicer Klipper setup guide covers the host side. For this article I'll stay on stock firmware for each variant, because that's what 90% of owners run. Adding your Sovol printer in OrcaSlicer OrcaSlicer ships with verified Sovol profiles upstream, so you don't need to hunt for community JSONs unless you want them. Here's what's actually in the box. Verified Sovol profiles in upstream OrcaSlicer I checked the resources/profiles/Sovol/machine/ directory in the OrcaSlicer GitHub repo. These are the machine JSON files that are present right now: SV06 family: Sovol SV06, SV06 Plus, SV06 ACE (with 0.2/0.4/0.6/0.8 nozzle variants), SV06 Plus ACE SV07 family: Sovol SV07, SV07 Plus, both with 0.4 nozzle variants SV08 family: Sovol SV08 (0.2/0.4/0.6/0.8 nozzle), SV08 MAX (0.4/0.6/0.8 nozzle) What's missing? An "SV06 Plus 0.6 nozzle" entry in upstream. The Sovol forum (#5977) has a known workaround: clone an existing 0.4 profile, change nozzle diameter and line widths, save under a new name. Or grab the jbob fork that lives on the Sovol forum. Our add-custom-printer guide walks through cloning if you've never done it. The wizard, step by step Launch OrcaSlicer 2.x and open File > New Project. Click the printer dropdown at the top and pick Add Printer (or run the Configuration Wizard from Help > Wizard). Vendor list: tick Sovol. The model dropdown opens with SV01, SV01 Pro, SV02, SV05, SV06, SV06 Plus, SV06 ACE, SV06 Plus ACE, SV07, SV07 Plus, SV08, SV08 MAX, Zero. Tick the variant you own and the nozzle size. For most people that's "Sovol SV08 0.4 nozzle" or "Sovol SV06 0.4 nozzle". Tick the filament profiles you want loaded. Generic PLA, Generic PETG, and Generic ABS cover starter needs. You can add brand profiles later. Finish. The printer appears in your top-right printer selector. If you're new to the wizard itself, our setup wizard walkthrough covers the optional parts I skipped over. Setup path A: SV06 / SV06 Plus (Marlin via SD card or USB) The SV06 is the simplest of the three to slice for and the most awkward to print from. There's no network. You're sneakernetting G-code or running a Pi. Slice your part with the SV06 0.4 nozzle profile selected. Click Export G-code and save to the SD card. Eject, walk to the printer, insert, print. Optional but worth it: hook the SV06 to an OctoPrint instance running on a Raspberry Pi. In OrcaSlicer, open Printer Settings > Physical Printer, set host type to OctoPrint, paste the OctoPrint URL plus the API key from OctoPrint Settings > Application Keys, and click Test. From there, "Print" in OrcaSlicer uploads the G-code straight to OctoPrint. See our OctoPrint setup guide for the rest. The stock SV06 start G-code (M104, M140, G28, G29 in that order) works fine with OrcaSlicer's built-in profile. Don't change it unless you've actually flashed Klipper. If you have, swap the start G-code to PRINT_START EXTRUDER_TEMP=[nozzle_temperature_initial_layer] BED_TEMP=[bed_temperature_initial_layer] and define the macro in printer.cfg. Setup path B: SV07 / SV07 Plus (Klipper, network) The SV07 is where it gets interesting. Klipper, Mainsail, network printing, and the part of the setup that trips up almost everyone: the API key question. Run the wizard, pick Sovol SV07 (or SV07 Plus), 0.4 nozzle. Get the printer's IP. On the SV07 touchscreen, dig into Settings until you find Network. Or check your router's DHCP lease table. In OrcaSlicer, click the printer name top-right and select Connect to Printer. A Physical Printer dialog opens. Host type: Klipper (Moonraker). Host IP: enter just the IP, no http://, no trailing slash. Port: leave 80 (Mainsail) or use 7125 if you point at Moonraker direct. API key: leave blank. Stock Sovol Mainsail does not require one. This is verified in Sovol forum thread #8606 ("The stock install on the Klipad doesn't use an API or certificate"), and it contradicts most online guides which insist you need one. You don't. If you locked Moonraker down yourself with [authorization] in moonraker.conf, then yes, generate a key in Mains [truncated] --- ## OrcaSlicer for Anycubic Kobra 2 / Neo / Plus / Max: Setup and Best Settings (2026) URL: https://orcaslicer.net/orcaslicer-anycubic-kobra-2/ I picked up a Kobra 2 Neo as a backup printer last year and assumed the OrcaSlicer setup would mirror what I'd already done for my Kobra 3. Wrong family of firmware, wrong defaults, wrong everything. The Neo runs Marlin, the Pro and Plus and Max all run KobraOS (which is Klipper-based under the hood), and they each need different OrcaSlicer treatment. Bowden versus direct drive isn't the question either, because the entire Kobra 2 family is direct drive despite what some older guides still say. This guide covers all five Kobra 2 variants honestly. I'll show you which OrcaSlicer profile to pick, what the defaults actually look like, and where you'll want to override them. I'll also tell you the truth about Wi-Fi upload to stock KobraOS, which is currently broken (GitHub Discussion #6178 if you want to read the wreckage), and when Anycubic Slicer Next is the smarter choice instead. If you're cross-shopping the newer generation, our OrcaSlicer Kobra 3 guide is the right next step. And if you want the full menu of printers we cover, the printer profile hub is your map. Table of contents Which Kobra 2 do you actually own? OrcaSlicer profile support for the Kobra 2 family Setup walkthrough, from first slice to first print Best settings overrides per variant Calibration sequence for the Kobra 2 family OrcaSlicer vs Anycubic Slicer Next, which one for you Common pain points and fixes When to step up to a Kobra 3 instead Wrap up Which Kobra 2 do you actually own? This sounds like a silly question, but the Kobra 2 line has five members and they don't all behave the same way in OrcaSlicer. I've watched at least three friends pick the wrong profile because they assumed "Kobra 2" was one printer with size variants. It isn't. The firmware split alone changes how pressure advance, retraction, and even your G-code header behave. Here's the comparison I wish someone had handed me on day one. Numbers come from Anycubic's own product pages and the official OrcaSlicer machine JSONs. SpecKobra 2 NeoKobra 2 (base)Kobra 2 ProKobra 2 PlusKobra 2 Max Build volume (mm)220 x 220 x 250220 x 220 x 250220 x 220 x 250320 x 320 x 400420 x 420 x 500 Max print speed250 mm/s250 mm/s500 mm/s500 mm/s500 mm/s Typical print speed150 mm/s150 mm/s300 mm/s300 mm/s300 mm/s Max acceleration2,500 mm/s squared2,500 mm/s squared20,000 mm/s squared10,000 mm/s squared10,000 mm/s squared Hotend max temp260 C260 C260 C260 C260 C Bed max temp110 C110 C110 C90 C90 C ExtruderDirect drive (4:1)Direct drive dual-gearDirect drive double-gearDirect drive double-gearDirect drive double-gear LevelingLeviQ 2.0 (25-point)LeviQ 2.0 (25-point)LeviQ 2.0 (25-point)LeviQ 2.0 (49-point)LeviQ 2.0 (49-point) Stock firmwareMarlin-basedMarlin-basedKobraOS (Klipper)KobraOS (Klipper)KobraOS (Klipper) MSRP (USD)~$199~$259~$309~$449~$579 In plain English: the Neo and the base Kobra 2 are budget bedslingers running Marlin. The Pro is the speed king (Klipper firmware, tiny build plate, ridiculous acceleration ceiling). The Plus and Max are the large-format Klipper machines, with the Max in particular sporting a 420 mm bed that's hard to find anywhere else for under $600. I'll say this once and not bury it: if you're shopping new in 2026, the Kobra 3 line is the better buy unless you specifically need the Max's footprint at this price. The Kobra 2 family is previous gen, and Anycubic's R&D attention has clearly shifted. That said, if you already own one (or you found one used), it's still a perfectly capable printer and OrcaSlicer treats it well. OrcaSlicer profile support for the Kobra 2 family Good news: every Kobra 2 variant has an official profile in mainline OrcaSlicer. I checked the SoftFever GitHub repository directly under resources/profiles/Anycubic/machine/, and all ten JSONs are there: Anycubic Kobra 2.json (printer-model anchor) Anycubic Kobra 2 0.4 nozzle.json Anycubic Kobra 2 Neo.json Anycubic Kobra 2 Neo 0.4 nozzle.json Anycubic Kobra 2 Pro.json Anycubic Kobra 2 Pro 0.4 nozzle.json Anycubic Kobra 2 Plus.json Anycubic Kobra 2 Plus 0.4 nozzle.json Anycubic Kobra 2 Max.json Anycubic Kobra 2 Max 0.4 nozzle.json If your OrcaSlicer install doesn't show one of those entries, you're on a pre-1.9 build and need to update. The Pro profile in particular landed via PR #7082 (which closed Issue #6200) after the community pushed for it for months. I dug through that thread and the takeaway is simple: contributor keybl uploaded working JSONs, the maintainers merged them, and now you don't have to hand-craft anything. Same story for the Plus via Issue #4210. The defaults vary in ways that genuinely matter. Here's a condensed table of the values I pulled directly from each 0.4 nozzle.json. VariantG-code flavorMax accel (X/Y)Retraction lengthRetraction speed Kobra 2 (base)(inherited)2,500 mm/s²2.0 mm80 mm/s Kobra 2 Neomarlin22,000 mm/s²2.5 mm100 mm/s Kobra 2 Proklipper20,000 mm/s²1.5 mm30 mm/s Kobra 2 Plusklipper10,000 mm/s²1.5 mm30 mm/s Kobra 2 Maxklipper10,000 mm/s²1.5 mm30 mm/s Notice the pattern. Marlin variants get higher retraction (2.0 to 2.5 mm) at faster speed (80 to 100 mm/s). That's OrcaSlicer's defensive default for Marlin machines without firmware-side pressure advance dialed in. Klipper variants get the much tighter 1.5 mm at 30 mm/s, which is the Klipper-style approach: lean on the firmware's pressure advance to clean up retraction artifacts and keep your filament from getting chewed. One caveat about the Pro's 20,000 mm/s² acceleration. That number is the firmware ceiling, not a recommendation. I'll say this again later, but please don't try to print at 20k accel and expect quality. You'll get ringing that looks like ocean waves on every flat surface. OrcaSlicer's Printer Selection dialog. Pick your Kobra 2 variant from the Anycubic vendor list, not Bambu. (Generic OrcaSlicer UI shown, applicable to any printer including the Kobra 2.) Setup walkthrough, from first slice to first print Adding the printer in OrcaSlicer I'll walk you through this on a fresh install. If you already have OrcaSlicer set up for another machine, the steps are nearly identical, just enter through the Printer panel instead of the first-launch wizard. Open OrcaSlicer (any current 2.x build, or 1.10.x if you're on the older stable line). On first launch you'll see the printer wizard. If you've already set up another printer, click the + next to the printer dropdown in the top toolbar. Select vendor Anycubic. The list expands to show all of Anycubic's supported machines. Scroll until you see the five Kobra 2 entries grouped together. Pick the matching variant. Critical: if you own a Pro/Plus/Max, you want the Klipper-flavored profile (the wizard handles this automatically once you click the right model name). If you own a Neo or base Kobra 2, you'll get the Marlin-flavored profile. Nozzle selection. Only 0.4 mm comes pre-populated. If you've swapped to a 0.6 or 0.8 mm hardened nozzle (which is very popular for the Max), you'll need to clone the printer profile and edit nozzle_diameter, plus drop your retraction values down. Discussion #2461 on the OrcaSlicer repo documents this exact pain in detail. Filament. Add at least one PLA, PETG, and TPU profile from OrcaSlicer's "Generic" filament set. You can also import Anycubic's filament JSON if you're brand-loyal, or pull in third-party Polymaker / Sunlu / Esun profiles from the community. That's it for the wizard. You'll land in OrcaSlicer's Prepare tab with your Kobra 2 bed visible. From here you import an STL, place it, slice it, and you're ready to send to the printer. OrcaSlicer's Prepare tab is where you'll import your STL files. The Add object button (top toolbar) handles single or multi-part imports. (Generic OrcaSlicer UI applicable to any printer including the Kobra 2.) Network upload, USB, and SD card (the honest version) This is where I have to be blunt with you. OrcaSlicer's network upload to stock KobraOS is broken. I wish it weren't, because USB sticks are a 2014 solution, but [truncated] --- ## OrcaSlicer Adaptive / Variable Layer Height Tutorial (2026 Step-by-Step) URL: https://orcaslicer.net/orcaslicer-adaptive-layer-height/ I sliced a 180mm chess pawn at 0.20mm fixed layers, got an 8-hour estimate, and almost canceled. Then I enabled adaptive layer height with the slider biased toward Quality, hit slice again, and the same model came in at 5 hours 40 minutes, with the same quality on the curves and a faster march up the flat sections. Adaptive isn't always a win, but when it is, it's genuinely free time savings. I've been running variable layer height in OrcaSlicer for two years across a Bambu P1S, a Creality K1, and an old Ender 3 V2 on Klipper. I've watched it save 27% on a tall vase, do nothing on a tray of 30mm cubes, and add an hour to a print I sliced wrong. This guide covers what the algorithm does, the exact button presses, every setting that matters, the bugs you'll trip over (with GitHub issue numbers), and when to just leave it off. Table of contents What adaptive layer height actually does When to use it, when to skip it Where the controls live in OrcaSlicer Step-by-step: enable and tune adaptive layer height The Variable Layer Height editor in detail Real-world results: how much time you actually save Common failure modes and fixes (with bug numbers) Settings that interact with adaptive layer height Quick FAQ Wrap-up What adaptive layer height actually does Adaptive (or variable) layer height changes the Z step thickness across the height of a single object. Detail-rich curves get thinner layers so the staircase artifact shrinks, and flat vertical walls get thicker layers so you stop paying for fine detail you can't see. The job of the slicer is to figure out, for each Z range, how thick the next layer should be. The algorithm didn't start in OrcaSlicer. It came from PrusaSlicer, which inherited it from Slic3r and academic work going back to Pandey et al. in 2003 and Wasserfall's 2014 Slic3r adaptive prototype. PrusaSlicer's metric limits the cross-sectional area between the stepped surface and the original mesh. Bambu Studio inherited that code, and OrcaSlicer is a fork of Bambu Studio, so the underlying logic is the same family. What you don't do in OrcaSlicer is type a cusp-height number. Instead you pick a Quality/Speed bias on a slider, and the slicer derives target layer thicknesses per Z range from the printer's min and max layer-height limits. That's the whole user-facing knob. Everything else is either smoothing or manual paint. If you want the bigger picture on how this fits with the rest of the slicer, my complete settings reference covers how every quality control hangs together. Adaptive layer height is one of maybe five settings that's actually worth touching on most prints. When to use it, when to skip it Adaptive isn't a magic button. It's a trade. You're trading some flat-section speed for some curve-section quality, and the trade only pays off if your model has both flat sections and curve sections, with enough total height for the savings to matter. I've seen people slap it on a 30mm calibration cube and wonder why nothing changed. Nothing changed because there was nothing to optimize. Here's the honest decision matrix I use: Model typeAdaptive helps?Why Tall (>150mm) curved-top organic (busts, vases, animals)Yes, 15 to 30% savingBig flat majority, expensive curve minority Cup lids, dome enclosures, anything with shallow domesYes, 10 to 25% savingThe dome is exactly where adaptive earns its keep Calibration cubes, rectangular cases, all-flat partsNo, near zeroAlgorithm picks max layer height for almost everything All-curve from base to top (smooth ovoid, sphere from Z=0)Net lossAlgorithm picks min layer height for almost everything Short prints (<50mm)NoNo flat majority to thicken Already printing at 0.28mm or 0.30mm fixedNoNowhere to "speed up to" Multi-color (AMS) printsNet loss, will probably error outPrime Tower bug, see failure section Anything with tree supportsNet loss, sometimes brokenTree supports ignore the toggle, see bug list The Obico benchmark of a sphere going from 16:43 to 12:05 (27.7% saving) is the one number that gets quoted everywhere. It's real, but it's one shape. My own runs on tall pawns, vases, and a 220mm bust have been more like 14% to 22%. Don't promise yourself 30%. For multi-color users, adaptive and the Prime Tower don't get along. OrcaSlicer requires all objects on the plate to share the same variable-layer-height profile, and any mismatch throws a "Prime tower is only supported if all objects have the same variable layer height" error (issues #5978 and #11648). The easiest path is to disable Prime Tower, or to just leave adaptive off for AMS prints. Tree supports are the other deal-breaker. They've ignored the adaptive toggle across multiple versions (issues #1787, #6300, #9795, and #10650). Sometimes you'll get 0.06mm support layers that under-extrude and collapse. If you need tree supports, switch to fixed layer height. If you need adaptive, switch to normal supports. There's a deeper write-up on the trade-off in my normal vs tree supports comparison. Where the controls live in OrcaSlicer This is the part that confuses everyone the first time. There are two related entry points and they're easy to mix up. The Variable Layer Height editor (toolbar button) This is the primary path in 2.3.x. You'll find the icon on the top toolbar of the Prepare view, next to the Move/Rotate/Scale/Cut tools. Click your model first, then click the Variable Layer Height icon. A right-side panel opens with a vertical bar that represents the model from bottom to top, color-coded by layer thickness. Green bands mean thin layers (more detail), red bands mean thick layers (more speed). Three buttons sit above the bar: Adaptive (run the auto algorithm), Smooth (Gaussian blend the profile), and Reset (wipe it). Process > Quality > Layers This is where the base layer height lives. The setting is called Layer height, and on a 0.4mm nozzle it defaults to 0.20mm with a sensible range of 0.04 to 0.32mm (20 to 80% of nozzle diameter). The adaptive algorithm uses this as a reference point when it rebases the profile. If you change Layer height from 0.20 to 0.24 after running Adaptive, the variable profile rescales roughly proportionally. The Initial layer height setting also lives here. Variable layer height does not change the first layer. You'll see a visual discontinuity at the bottom of the side bar in the editor, and that's normal, not a bug. Printer > Extruder (min/max layer height) This is where the adaptive range actually lives, and it's the source of the most-requested missing feature. The Min layer height and Max layer height fields in the extruder definition control the floor and ceiling the Adaptive algorithm uses. There's no separate "adaptive min" or "adaptive max" setting in the editor. Users have asked for one in issues #5709, #9327, #10069, and #10538, but as of 2.3.x the only way to change the adaptive range is to edit the printer extruder profile. For a 0.4mm nozzle, typical defaults are 0.08mm minimum and 0.28mm maximum. If you want a tighter range (say, you don't want adaptive ever picking 0.08mm because your printer struggles with it), bump the minimum to 0.12mm in the extruder profile, save the profile, then re-run Adaptive. The bar will redraw within the new range. The adaptive range is owned by the extruder profile, not the per-process settings. To change the ceiling adaptive can pick, edit Max layer height here. The red arrow shows where to look. SettingLives inDefault (0.4mm nozzle)Notes Layer heightProcess > Quality > Layers0.20 mmBase used outside variable regions Initial layer heightProcess > Quality > Layers0.20 to 0.24 mmVariable height does not touch this Min layer heightPrinter > Extruder~0.08 mmFloor for the Adaptive algorithm Max layer heightPrinter > Extruder~0.28 mmCeiling for the Adaptive algorithm Quality / Speed sliderVariable Layer Height editorCenterContinuous, no numeric value exposed Smoothing radiusEditor (Smooth dialog)~5 mm Z half-widthClick Smooth again to [truncated] --- ## Is OrcaSlicer Safe? Honest Trust Guide (2026) URL: https://orcaslicer.net/orcaslicer-safe/ Microsoft Defender told me OrcaSlicer was Trojan:Win32/Wacatac.H!ml last month, my friend asked me three times if it was a virus, and I had to walk him through a SmartScreen warning, a "damaged file" Gatekeeper message, and a "not an official website" note in OrcaSlicer's own README before he agreed to use it. The short answer is yes, it's safe. The longer answer is why every signal makes it look like it isn't, and why that's a separate problem from whether the slicer will actually wreck your PC. I've been installing OrcaSlicer on Windows, macOS, and Linux machines since the 1.6 days, and I've watched it migrate from the SoftFever GitHub repo to the OrcaSlicer organization in November 2025. I've seen the AV false positives evolve from Norton in 2023 to Wacatac.H!ml in early 2026. None of those flags ever pointed at real malware. They pointed at something more boring: an unsigned installer from a pseudonymous maintainer who hasn't bought a code-signing certificate. That's the whole story, and the rest of this article unpacks it with the receipts. Quick disclosure before we go further. orcaslicer.net (the site you're reading this on) is a third-party content site. We're not the official OrcaSlicer project. The project's own home is orcaslicer.com, and every official binary lives at github.com/OrcaSlicer/OrcaSlicer/releases. We don't host the installer here. Every download button on this site points to that GitHub releases page. What we do is write the guides, calibration walkthroughs, and per-printer setup help that the official docs don't always cover, in the same spirit as the third-party guide sites that exist for VLC, Blender, or OBS. That's the trust signal we can honestly offer you: we're not asking you to download anything from us, and we're being upfront about who we are. The OrcaSlicer project warns about non-official sites because some of them bundle adware or ship outdated, modified binaries. We do neither. If that disclosure still makes you uneasy, hit the GitHub link and ignore the rest of this page. I'd rather you do that than feel misled. Table of contents Why the scary warning showed up and what it actually means The trust stack: how open source makes OrcaSlicer auditable What VirusTotal actually shows (and why Wacatac.H!ml keeps appearing) What's actually inside the OrcaSlicer installer The bambu_network plugin: the one honest gotcha How to verify your download in 60 seconds Fake OrcaSlicer sites: the real safety risk Has OrcaSlicer ever shipped malicious code? Quick safety checklist before you install FAQ Why the scary warning showed up and what it actually means If you landed here because your browser or your OS just yelled at you, let me describe what probably happened. On Windows, you double-clicked OrcaSlicer_Windows_Installer_V2.3.2.exe and a blue box appeared saying "Windows protected your PC" with "Unknown publisher." On macOS, you opened the DMG, dragged the app to Applications, double-clicked it, and got "OrcaSlicer cannot be opened because Apple cannot check it for malicious software." On either OS, your antivirus may have piped up too with something like Trojan:Win32/Wacatac.H!ml or a Norton "Heuristic" verdict. None of those messages are about the contents of the OrcaSlicer binary. They're about who paid for what. SmartScreen, per Microsoft's own developer docs, flags any executable that hasn't built up "enough reputation" through downloads signed with a valid Authenticode certificate from a known publisher. A standard Authenticode cert from Sectigo runs around $215 per year. DigiCert charges closer to $500. EV certs that skip SmartScreen on first install run $400 to $700+ per year. OrcaSlicer's maintainer SoftFever has explicitly said this cost isn't worth it for a free open-source project. So every fresh release triggers SmartScreen until enough people have downloaded and run that exact binary for Microsoft's reputation system to age it in. Gatekeeper on macOS is the same root cause with different paperwork. Apple's Developer ID program costs $99 per year. Notarization sends every build through Apple's automated malware scanner before it can launch cleanly. OrcaSlicer's macOS builds don't go through that pipeline. So Gatekeeper blocks them. On macOS Sequoia (15.x) and later, you can't even right-click and Open anymore. You have to attempt the launch, get rejected, then go into System Settings, Privacy and Security, and click "Open Anyway." It feels worse than it is. Here's the cost picture for any open-source project thinking about this: Certificate / programAnnual cost (approx.)What it gets you Sectigo standard Authenticode$215Removes "unknown publisher" once reputation builds DigiCert standard Authenticode~$500Same, with a more recognized CA DigiCert EV code signing~$549Bypasses SmartScreen on first run Apple Developer Program (notarization)$99Gatekeeper opens the app without ceremony For a single solo maintainer publishing a free slicer, that's $300 to $650 a year out of pocket just to make the OS stop yelling. It doesn't make the code safer. It just makes the warning go away. So the reframe I want you to take away from this section: SmartScreen and Gatekeeper warnings are funding signals, not safety signals. Plenty of paid commercial software has shipped with real malware. Plenty of unsigned open-source software has been clean for a decade. Code signing is a trust shortcut for the operating system. It is not a content check. The trust stack: how open source makes OrcaSlicer auditable Okay, so signing tells you nothing. What does tell you something? This is the actual answer to "is OrcaSlicer safe," and it's where open-source software has a structural advantage over closed binaries you can't see inside. License: AGPL-3.0 OrcaSlicer is licensed under the GNU Affero General Public License version 3. The README states it verbatim. AGPL is the strongest copyleft license commonly used in this space. It legally requires that anyone who distributes a modified or unmodified build of the code also ship the source code, including any network-side modifications. If someone shipped a malicious build calling itself OrcaSlicer, they'd be obligated under AGPL to publish their source or be in violation of the license. That doesn't physically stop a bad actor, but it gives the community legal grounds to issue takedowns when someone does. (The pressure-advance calibration pattern is GPL-3.0 separately, but the slicer itself is AGPL-3.0.) The fork heritage This is the part that often gets glossed over and shouldn't. OrcaSlicer isn't a fresh codebase that appeared out of nowhere in 2022. It inherits roughly 15 years of public scrutiny through its ancestry: Slic3r (2011, Alessandro Ranellucci, written for the RepRap community, fully open source) forked into PrusaSlicer (2016, Prusa Research, AGPL-3.0, the C++ rewrite that powers most of the desktop slicer ecosystem today) forked into Bambu Studio (Bambu Lab, AGPL-3.0 core with the optional closed bambu_network plugin) forked into OrcaSlicer (2022, SoftFever, AGPL-3.0) Every node in that chain is open source. The slicing engine code in OrcaSlicer, the part doing the actual heavy lifting, has been read by thousands of contributors across the four projects. That's a fundamentally different threat model from a closed proprietary installer maintained by people you can't audit. Active development, named contributors, real sponsors The bug tracker is public. The release notes are public. The pull-request queue is public. Recent releases call out @Noisyfox and @igiannakas as exceptional contributors and credit dozens more by handle. The README lists QIDI and BigTreeTech as project sponsors. Both are real 3D-printing hardware companies whose brand value would take a hit if they were sponsoring malware. There's also GitHub Sponsors, Ko-fi, and PayPal donation links with public backers lists. None of this is proof of safety on its own. What it shows is that the project is not a black box and not a [truncated] --- ## OrcaSlicer for Elegoo Neptune 4 / Pro / Plus / Max: 2026 Setup Guide URL: https://orcaslicer.net/orcaslicer-elegoo-neptune-4/ Elegoo's pitch on the Neptune 4 line was 500mm/s, and I bought one expecting that to mean something. It doesn't, not for any print quality you'd actually accept, but the printer is genuinely good at 200mm/s once you've got OrcaSlicer dialed in and Klipper configured properly. Most Neptune 4 owners stop at the bundled Cura profile and never see what the printer can actually do. This guide walks through the setup, the settings, and the calibration sequence I'd hand to a friend who just unboxed any of the four variants. I'll be honest about where the upstream profiles still ship with footguns, where the Plus and Max differ from the standard model in ways no review tells you about, and which Klipper config edits you need to make before OrcaSlicer's per-object features stop throwing errors. If you want the broader context, the OrcaSlicer printer profiles hub covers the whole vendor map; this article is the deep cut for the Neptune 4 family. Table of contents The four printers at a glance Why pair OrcaSlicer with the Neptune 4 Step-by-step setup Best settings to override out of the box Calibration order that actually works Pro vs Standard vs Plus vs Max: what to change Importing community and vendor profiles Common pain points and fixes Neptune 4 vs Centauri Carbon Wrap-up and next steps The four printers at a glance All four Neptune 4 printers share the same Klipper firmware base, the same 64-bit ARM SBC, and the same 300C hotend ceiling. They diverge on build volume, extruder, and how much gantry mass the motion system has to drag around. I've put the marketing claims and the practical numbers side by side because the gap between them is the whole story. Spec Neptune 4 Neptune 4 Pro Neptune 4 Plus Neptune 4 Max Build volume (advertised)225 x 225 x 265225 x 225 x 265320 x 320 x 385420 x 420 x 480 Printable area in upstream Orca JSON225 x 225235 x 230320 x 320420 x 420 Stock extruderDual-gear direct driveDual-gear direct drive (hardened nozzle)Single-gear, Bowden-styleSingle-gear, Bowden-style Marketed top speed500 mm/s500 mm/s500 mm/s500 mm/s Practical quality speed150-200 mm/s150-250 mm/s120-200 mm/s80-150 mm/s Max accel (upstream Orca config)10,000 mm/s210,000 mm/s210,000 mm/s210,000 mm/s2 Hotend max300C300C300C300C Bed110C single zone110C dual zone (100W + 150W)110C110C ConnectivityUSB / LANUSB / LAN / Type-CWi-Fi / WLAN / USBWi-Fi / WLAN / USB Two things to flag here. The Neptune 4 Pro printable area is 235 x 230 in the upstream OrcaSlicer JSON, not 225 x 225 as the spec sheet says. That's intentional in the profile but it'll bite you if your physical printer was set up to home at the smaller envelope. Open the printer settings and reconcile that number with whatever your start g-code expects. The other one is the extruder. The standard Neptune 4 and the Pro both ship dual-gear direct drive out of the box. The Plus and the Max do not. Elegoo sells the dual-gear direct-drive kit as a separate upgrade for those two models, and that one fact changes your retraction defaults, your stringing behavior, and what filament you should bother trying first. The 500 mm/s headline is a kinematic ceiling, not a print speed. I've never seen a Neptune 4 produce acceptable surface quality above ~250 mm/s on outer walls, and the Max especially struggles at speed because it's lugging a much larger gantry. If you bought one of these expecting half a meter per second of usable print speed, recalibrate your expectations now. They're still excellent printers at honest speeds. Why pair OrcaSlicer with the Neptune 4 The Neptune 4 line ships with Klipper firmware on a Linux SBC daughterboard. That's the unlock. OrcaSlicer talks Klipper natively through Moonraker, sends per-object cancellation data, ships built-in calibration test patterns that target Klipper-specific tuning (pressure advance, input shaper coefficients), and lets you push gcode straight to the printer over LAN without dragging a USB stick around. Cura can do some of this with plugins, but Orca's the one that was built around the workflow. You also get organic profile support upstream. Every variant (4, Pro, Plus, Max) lives in resources/profiles/Elegoo/machine/EN4SERIES/ in the OrcaSlicer repo, with 0.2 / 0.4 / 0.6 / 0.8 nozzle variants for each one. No download, no zip dance, no broken third-party pack. The configuration wizard finds them. What you give up versus the bundled Cura profile: Elegoo's curated material library doesn't follow you. The stock OrcaSlicer filament list for Neptune 4 is bare bones (PLA, PETG, ABS, basically that). For TPU, ASA, Nylon, or carbon fill you'll either build profiles yourself or import from the community. There's also a small learning curve for the network printer dialog, which I'll cover in setup. None of this is hard. It's just not as turn-key as Cura's "click the Elegoo preset" experience. For beginners on day one, the bundled Cura is fine. Print the test cube, get a feel for the machine, don't fight the slicer. OrcaSlicer is the upgrade path you graduate to when you want calibration tests, faster network workflow, and per-object features. Don't switch on day one if you're still figuring out which way the bed levels. Step-by-step setup You need OrcaSlicer 1.7.0 or newer for the Neptune 4 family. Older builds either don't have the Plus profile or ship with the retraction bug I'll cover later. Grab the latest stable from the OrcaSlicer GitHub releases. On the printer side, your Elegoo Klipper firmware should be V1.1.2.28 or newer. Older firmware works but Moonraker will be missing some endpoints. Add the printer Open OrcaSlicer. If this is a fresh install, the configuration wizard launches automatically. Otherwise, hit Configuration then Wizard from the top menu. In the vendor list, check Elegoo. In the model list, check whichever Neptune 4 variant you own and pick your nozzle size (0.4 mm is the default, the others are there if you swapped). Finish the wizard. The top-left printer dropdown should now show something like "Elegoo Neptune 4 Pro 0.4 nozzle". First sanity check: open Printer Settings and confirm the printable area matches what your printer actually homes to. The Pro shows 235 x 230 by default. If your physical printer's start g-code parks the head at 225 x 225, override it now or your prints will clip the edge of the bed. The Plus is 320 x 320, the Max is 420 x 420 (some Max owners override to 425 or 427 because the actual usable envelope is a few mm larger than the spec, but stick with the upstream value unless you've physically verified the extra space). Connect over the network (Moonraker) This is the step that trips people up because OrcaSlicer's host-type dropdown lists OctoPrint right above the option you actually want. You don't have OctoPrint. You have Moonraker, which is the Klipper API server, and Orca's name for that is "Octo/Klipper". Find your printer's IP address from the touchscreen (Settings then Network, or Advanced Settings on some firmware revisions). Then in OrcaSlicer: Open Printer Settings, click the wrench/gear icon next to the printer name. Find the Physical Printer or Connection section. Set Host Type: Octo/Klipper. URL: http://192.168.x.x (use your printer's IP, include the http://). API Key: leave blank on stock Elegoo firmware. Moonraker isn't locked down by default. If you've added authentication via Mainsail or Fluidd later, generate a key in Settings then API and paste it here. Click Test. You should get a green success message. Once the test passes, the "Print" button in the slice screen gains a "Send to printer" option that uploads gcode directly. Moonraker listens on port 7125 by default; you don't usually need to specify it because OrcaSlicer auto-routes through the standard HTTP port. If you've remapped Moonraker's port (rare), append it like http://192.168.x.x:7125. For the deep walkthrough on Klipper printer connections in OrcaSlicer, the dedicated Klipper setup guide covers SSL, custom ports, and authentication scenarios in [truncated] --- ## OrcaSlicer for Anycubic Kobra 3 / S1 / Max Combo: 2026 Setup Guide URL: https://orcaslicer.net/orcaslicer-anycubic-kobra-3/ Anycubic ships the Kobra 3, Kobra S1, and Kobra 3 Max with a slicer they tell you to use, called Anycubic Slicer Next, and that slicer is OrcaSlicer wearing an Anycubic jacket. The catch is that if you actually want upstream OrcaSlicer (because you've got other printers on the bench, or because you want the calibration tests Anycubic strips out, or because you just don't trust vendor-locked software), you'll spend a few hours wiring up something the official slicer does in two clicks. I've been through this dance with the Kobra 3 family, and I want to save you the wasted weekend I had. This guide covers what's verified, what's missing, and where upstream OrcaSlicer genuinely beats Slicer Next versus where it's just stubbornness on your part. If you only print single-color PLA and never touch calibration, honestly, stop reading and go install Slicer Next. If you do anything more involved than that, keep going. Table of contents Kobra 3 vs Kobra 3 V2 vs Kobra S1 vs Kobra 3 Max in one table Anycubic Slicer Next vs OrcaSlicer: the honest answer Does upstream OrcaSlicer actually support the Kobra 3 family? Setting up OrcaSlicer for your Kobra How OrcaSlicer talks to a Kobra 3 (or doesn't) Best settings to override from the defaults Multi-color with ACE Pro in OrcaSlicer Calibration order for Kobra 3 + OrcaSlicer Common Kobra 3 + OrcaSlicer problems and fixes Bottom line: which slicer for which job Frequently asked questions Kobra 3 vs Kobra 3 V2 vs Kobra S1 vs Kobra 3 Max in one table The Kobra 3 family is wider than people realise. Anycubic has shipped four models that all run the same firmware stack (a customised Klipper variant they call GoKlipper, branded "Kobra OS"), but the kinematics, build volume, and enclosure story diverge in ways that matter for your slicer profile. I'll spell it out, and then we can talk about what to actually do with it. SpecKobra 3 / ComboKobra 3 V2 / V2 ComboKobra S1 / S1 ComboKobra 3 Max / Max Combo Build volume250 x 250 x 260 mm255 x 255 x 260 mm (wider Y rail)250 x 250 x 250 mm420 x 420 x 500 mm KinematicsCartesian bedslingerCartesian, 60 mm Y rail, SG15 X bearingsFully enclosed CoreXYCartesian, large format Max print speed600 mm/s (300 recommended)600 mm/s (300 recommended)600 mm/s (300 recommended)600 mm/s (300 recommended) Max acceleration20,000 mm/s² (10,000 recommended)20,000 mm/s²20,000 mm/s²20,000 mm/s² Hot endDirect drive, all-metal, 300 °CImproved nozzle from S1, better airtightnessAll-metal 320 °C, hardened optionDirect drive, 300 °C Bed max110 °C110 °C120 °C110 °C Auto-levelingLeviQ 2LeviQ 3.0LeviQ 3.0 + AI spaghetti detectionLeviQ 3.0 EnclosureNoneNoneYes, with carbon-filtered exhaustNone ACE Pro support4 slots in Combo (8 with two units)4 in Combo (8 with two)4 in S1 Combo (8 with two)4 in Max Combo (8 with two) CameraNo or basic480p AI spaghetti cam480p AI spaghetti camYes FirmwareKobra OS (GoKlipper)GoKlipperGoKlipperGoKlipper Stock slicerAnycubic Slicer NextAnycubic Slicer NextAnycubic Slicer NextAnycubic Slicer Next Two takeaways. The S1 is the odd one out because of the enclosed CoreXY chassis, and that means slicer profiles for it can lean on chamber temperature and reduced cooling for ABS, ASA, and PA-CF in a way the bedslingers cannot. The Kobra 3 Max is the second odd one out because it's huge, and "big bedslinger" is a phrase that should make any sensible slicer config drop wall speeds by 10 to 15 percent. We'll come back to both points. Anycubic Slicer Next vs OrcaSlicer: the honest answer Anycubic's own documentation calls Slicer Next "open-source slicing software developed based on OrcaSlicer, optimized for Anycubic series printers." Translation: it's an OrcaSlicer fork. The Anycubic team pulls from upstream Orca, then adds their own tweaks for ACE Pro inventory sync, LAN print, the Anycubic Cloud, and a curated filament library tied to their store. They also strip out or lag behind some of upstream's calibration suite. I'll give you the honest split before the table: ConcernAnycubic Slicer NextUpstream OrcaSlicer First-time setup for Kobra 3 / S1 / MaxAuto-detects, profiles readyProfile present (except Kobra 3 Max), often needs an import ACE Pro multi-colorNative 4-color and 8-color (1.1.1+); live sync from ACEWorks as generic 4-extruder; live ACE sync needs custom firmware LAN print / direct sendYes, only supported slicerNo on stock firmware; works via Rinkhals Cloud upload + remote monitorBuilt inNot directly; export then upload Camera and AI spaghettiLive viewer, alertsUse the printer touchscreen or Anycubic app Calibration toolsAdaptive PA, YOLO Flow Calibration in 1.1.xFull suite: temp tower, flow rate (one-shot and pattern), PA tester, retraction tower, max volumetric speed Painting / tree supports / advanced multi-colorLags upstreamLatest features land here first Filament library breadthAnycubic-curatedHundreds of community presets Update cadenceTied to Anycubic releasesOrcaSlicer GitHub stable / nightly, much faster Vendor lock-inAnycubic Cloud, Anycubic store filamentsNone Stability with Kobra 3 G-codeBuilt and tested for itOccasional regressions in RC builds (stay on stable) I'm not going to pretend Slicer Next is bad. For someone who bought a Kobra 3 Combo and wants to print multi-color models from MakerWorld, click "send" from the slicer to the printer, and never read another forum post, Slicer Next is the right tool. The honest case for upstream OrcaSlicer is when you want one of the things on the right column: real calibration, broader filament profiles, advanced painting and multi-material tools, or independence from a vendor's release schedule. If you're running a Kobra 3 alongside a Bambu, a Prusa, or a Voron, you almost certainly want one slicer for everything, and OrcaSlicer is the only one of the three that supports all three families well. Does upstream OrcaSlicer actually support the Kobra 3 family? Mostly yes, with one glaring gap. I checked the SoftFever main branch directly under resources/profiles/Anycubic/machine, and here's the actual state of play: Kobra 3: profile present, with 0.2, 0.4, 0.6, and 0.8 mm nozzle variants. This profile covers both the original Kobra 3 and the V2 (there's no separate V2 entry, which is the open feature request in OrcaSlicer Issue #11552, where users complain the bundled profile prints slower than Anycubic's own). Kobra S1: profile present with a 0.4 mm nozzle variant. Kobra Max: profile present with a 0.4 mm nozzle variant. This is the original Kobra Max, not the new Kobra 3 Max, and the build volume difference matters. Kobra 3 Max: no upstream profile. Issue #9507 has been open since the printer launched. If you own a Kobra 3 Max, you have to either start from the Kobra 3 profile and override the build volume, or grab a community pack. The MakerWorld listing 1648658 is the cleanest community Kobra 3 Max profile I've found, and the Layer3D Creations V2 pack on Printables (model 982339) covers the regular Kobra 3 with extra polish. There's nothing in the profile JSONs about the ACE Pro by default. The bundled profile presents the Kobra 3 as a single-extruder machine, and you turn it into a multi-material machine yourself by changing the printer to 4 extruders in the printer settings. We'll get to that in the multi-color section. Setting up OrcaSlicer for your Kobra You've got three install paths, and your choice depends on whether you also want to run Slicer Next. The Create Printer / Nozzle wizard in OrcaSlicer where you pick vendor, model, and nozzle size before the printer appears in your dropdown. Path A: use the bundled profile (easiest) Open OrcaSlicer (latest stable, not RC, not dev; more on that in pain points). Run the Configuration Wizard. If it didn't auto-launch, hit Configuration in the top menu and pick "Configuration Wizard." Scroll to Anycubic in the vendor list. Tick Kobra 3, Kobra S1, or Kobra Max as applicable. The wizard will offer the nozzle variants. Pick the nozzle you actually have. Almost everyone is on the st [truncated] --- ## OrcaSlicer for Creality Ender 3 V3 SE, KE & Plus: Setup Guide URL: https://orcaslicer.net/orcaslicer-ender3-v3/ I added an Ender 3 V3 KE to my fleet in mid-2024 and immediately spent the first weekend trying to figure out why OrcaSlicer kept saying "Update OctoPrint to 1.1.0 or higher" when I tried to connect it. The KE doesn't run OctoPrint. It runs Creality OS, which is a Klipper fork. The fix turned out to be appending :4408 to the IP address. Two characters. Saved nowhere obvious in the Creality docs. This is the article I wish I'd had. It covers the Creality Ender 3 V3 family with OrcaSlicer in 2026: V3 SE (Marlin, USB/SD only), V3 KE (Klipper, network), V3 Plus (Klipper, larger 300mm bed), and the V3 base model (CoreXZ kinematics, Klipper). All four models are bestselling Crealities of the 2024-2025 era and they need different setup paths in OrcaSlicer. One quick note before we start: there is no "Ender 3 V3 Core XY". The base V3 is CoreXZ (belts on X and Z, traditional Y bedslinger). The CoreXY in Creality's 2025 lineup is the standalone Creality Hi, not branded as Ender 3 V3. I keep getting asked about this; saving you the search. Written against OrcaSlicer v2.3.2 (current stable; v2.3.1 specifically fixed Ender 3 V3 KE profile inconsistencies, so 2.3.1 is the minimum recommended version). V3 family spec reference SpecV3 SEV3 KEV3 (base CoreXZ)V3 Plus Build volume220×220×250220×220×240220×220×250300×300×330 MotionCartesian bedslinger, dual Z lead screwsCartesian bedslinger, linear rail on XCoreXZ bedslingerCoreXZ bedslinger, dual Y motors FirmwareMarlin 2.xKlipper (Creality OS, open-source)Creality OS (Klipper-based, more locked)Creality OS Web UI out of boxNoneCreality Cloud only (Mainsail/Fluidd via root + helper)Creality Cloud onlyCreality Cloud only Hotend max260°C (PTFE-lined Sprite DD)300°C (bi-metal heatbreak, ceramic 60W, copper unicorn)300°C unicorn300°C Bed max100°C100°C100°C100°C Default platePEI-coated spring steel (smooth)Textured PEI spring steelPEI flexPEI flex AccelerometerNoneNOT built-in (ADXL345 is $14 add-on)None built-inNone built-in Auto-levelingCR-Touch + strain-gauge auto Z-offsetCR-Touch + strain-gaugeStrain-gauge nozzle probeStrain-gauge nozzle probe NetworkNone (USB + microSD)Wi-Fi + LAN, Klipper/MoonrakerWi-FiWi-Fi Max speed (rated)250 mm/s500 mm/s600 mm/s600 mm/s Max acceleration2,500 mm/s²8,000 mm/s²~12,000 mm/s²~12,000 mm/s² AMS analogNoneNone (CFS not yet supported)Future CFS teasedFuture CFS teased Important to call out: none of the V3 models has a built-in accelerometer. The KE and Plus support adding an aftermarket Creality ADXL345 G-Sensor for $14, but it's an accessory, not stock equipment. People mistakenly assume the K1's onboard MPU-6500 is also on V3 KE; it isn't. Picking the right OrcaSlicer profile OrcaSlicer's printer dropdown ships profiles for the entire Creality Ender 3 family alongside Bambu, Anycubic, Voron, FLSun, and dozens more. The V3 SE, V3 KE, and V3 Plus each have their own pre-configured profile. OrcaSlicer's Add Printer wizard ships profiles for all four V3 models: "Creality Ender-3 V3 SE", "Creality Ender-3 V3 KE", "Creality Ender-3 V3 Plus", and "Creality Ender-3 V3" (the CoreXZ base). Pick the one matching your hardware. Stock nozzle is 0.4mm on all four; tick that. The profiles in OrcaSlicer 2.3.1+ correctly place the print origin and use the right start G-code. If you're on an older OrcaSlicer version (2.2.x or earlier), the V3 SE profile in particular had a wrong origin that printed parts off-bed. Update OrcaSlicer first. Connecting OrcaSlicer to the V3 series V3 SE: SD card workflow only The V3 SE runs Marlin and has no built-in network capability. Workflow is microSD only, with USB-C as a backup that has known driver flakiness on Windows 11 (install the CH340 driver if Windows doesn't see the printer). Slicing flow: File → Export → Export plate sliced file (.gcode). Save to the root of the microSD card as print.gcode (older firmware doesn't traverse subfolders). Eject, insert into the printer, choose from the print menu. If you want network printing on a V3 SE, the only path is adding an external Raspberry Pi running OctoPrint connected to the printer's USB-C port. Set baud to 115200. OrcaSlicer's network printing remains best-effort because Marlin doesn't expose Klipper-style status APIs. V3 KE: Klipper native via Moonraker The KE is the easiest V3 to network-print to because Creality publishes the Klipper source on GitHub. Find printer IP: Settings (gear icon) → Network → Wi-Fi/Ethernet. Note the IPv4. (Optional) Enable root for full Mainsail/Moonraker access: Settings → Root Account Information → accept disclaimer. Default credentials: root / creality_2023. Not required for vanilla Orca uploads, but needed if you want to install Mainsail or Fluidd via Helper Script. In OrcaSlicer, select the V3 KE printer profile, click the Wi-Fi icon next to the printer name, and the Physical Printer dialog opens: Host Type: Klipper (Octo/Klipper) Hostname / IP: http://192.168.x.x:4408/ (Creality's Moonraker shim runs on port 4408) Device User Interface: same as hostname API Key: blank (Creality stock doesn't use one) Click Test. Should return "Connected to OctoPrint…" If you've installed Mainsail or Fluidd via the Guilouz Helper Script, the web UI lives at port 4409 (distinct from the slicer port). Don't confuse them. V3 Plus and V3 base CoreXZ: same as V3 KE Both run Creality OS Klipper. Same wizard, same port 4408, same Host Type. Profile names: "Creality Ender-3 V3 Plus" and "Creality Ender-3 V3". Common connection errors "Update OctoPrint to 1.1.0 or higher": most common error. Means you forgot to append :4408 to the IP. Add the port suffix and the error goes away. (GitHub issues #3804 and #5545 are full of users hitting this.) "Connection failed: 404" (issue #8733): trailing slash matters. http://IP:4408/ works; http://IP:4408 without the slash sometimes fails on KE. "Upload OK, print won't start": Creality's Klipper fork ignores OrcaSlicer's start_print signal unless "Print after upload" is checked AND the printer has been freshly homed since boot. "Klipper reports: MESH_BED_LEVELING undefined": V3 KE/Plus uses BED_MESH_PROFILE LOAD=default, not Marlin's M420. OrcaSlicer's profile usually inserts the right macro, but custom start G-codes can break this. Calibration sequence V3 SE (Marlin, manual SD workflow) Auto-level once from the LCD: Level → Auto-Level. Wait the full cycle (~5 minutes cold). Auto Z-offset via the strain-gauge sequence on the LCD (already done at first boot). Re-run if you swap the nozzle. Temperature tower: Orca menu → Calibration → Temperature → choose PLA range (190-230°C). Slice, save to SD, print, read off the lowest stringing band. Flow ratio: Calibration → Flow Rate (one-pass). Default 0.98 is usually too high for the SE. Community average is around 0.95 for Creality Hyper PLA on the SE. Pressure Advance: stock SE Marlin does NOT include Linear Advance, so OrcaSlicer's PA test won't apply. Skip it. Max volumetric speed: skip the test, set 11 mm³/s for PLA in filament settings (the SE caps around 10-12 mm³/s on the stock hotend). V3 KE / Plus / V3 base (Klipper) Bed mesh: SSH or via Mainsail run BED_MESH_CALIBRATE once after assembly. The strain-gauge auto-Z handles per-print Z-offset. Pressure Advance: OrcaSlicer Calibration → Pressure Advance → Line method. Direct-drive setting. Print, measure, set in filament profile. Real-world starting points for the Creality Sprite-style direct extruder: PLA: ~0.030-0.045 PETG: ~0.045-0.060 ABS: ~0.030-0.040 TPU 95A: ~0.060-0.090 Flow ratio: Calibration → Flow Rate → Pass 1 then Pass 2. KE community values: PLA ~0.97, PETG ~0.96, ABS ~0.95. Max volumetric speed: KE/Plus hotend rated at 32 mm³/s with PLA Hyper, realistic 22-24 mm³/s for general PLA, 14 mm³/s PETG. Input shaping (KE/Plus only, requires aftermarket ADXL345 sensor): in Mainsail run SHAPER_CALIBRATE for X then Y. Without the sensor, leave Creality's factory-tuned mzv shaper at its defaults; you can still print at 200 mm/s without visible ringing. KAMP (Klipper Adap [truncated] --- ## OrcaSlicer Review 2026: Honest Pros, Cons & Who Should Use It URL: https://orcaslicer.net/orcaslicer-review-2026/ I've used OrcaSlicer as my daily driver since mid-2024. Two Bambu printers, a Voron 2.4, an Ender 3 V3 KE, and an SLA Prusa SL1S that I open in PrusaSlicer because Orca doesn't do resin. The last 18 months have included the Bambu firmware lockdown, the OrcaSlicer org transfer, the v2.3.x feature wave, and a long list of regressions in the v2.3.2 release that the maintainers are still working through. This review is what I'd tell a friend asking "should I download OrcaSlicer in 2026?". The short answer is "depends on your printer and how the Bambu lockdown affects you". The long answer is below, with specific GitHub issue numbers as evidence rather than vibes. Written against v2.3.2 stable (released March 23, 2026, about 13.8k GitHub stars and around 1,800 open issues at time of writing). The 5-second verdict by user type You own…RecommendationWhy Bambu Lab printer (X1C, P1S, A1, H2D)Conditional yesCloud is broken; LAN + Developer Mode still works Prusa MK3S, MK4, MK4S, XL, Core OneYes, but consider PrusaSlicer tooCalibration suite outclasses Prusa, but Prusa profiles are tuned by Prusa themselves Klipper printer (Voron, RatRig, Creality K1)Strong yes, best-in-classNative Moonraker, G2/G3 arcs, in-app input shaping calibration Generic FDM hobbyist (Ender 3 V3, Anycubic Kobra, Sovol)YesPre-baked profile for nearly every machine, calibration suite saves weeks SLA / resin printerNo, wrong toolOrcaSlicer is FDM-only. Use Lychee, Chitubox, or PrusaSlicer SLA If your situation matches one of these rows, you can skip to the section that explains why. If you're on the fence, the rest of the article walks through the honest pros, the honest cons, the firmware lockdown context, and who should not download OrcaSlicer. What OrcaSlicer is genuinely good at (10 honest pros) 1. The most complete in-app calibration suite of any slicer in 2026. Temperature, Flow Rate (two-pass), Pressure Advance (line, pattern, tower, plus Adaptive PA), Retraction, Tolerance, Max Volumetric Speed, VFA (Vertical Fine Artifacts), Input Shaping, Junction Deviation. All one-click model generation. v2.3.1 added the in-app input shaping wizard, validated on Voron 2.4 and FLSun T1 Pro. Two hours in the Calibration tab beats two weeks of test cubes downloaded from Thingiverse. 2. Native Klipper / Moonraker support. Direct upload, M73 progress, G2/G3 arc commands, per-filament pressure advance values stored on the filament profile. PrusaSlicer requires the OctoPrint plugin shim. Bambu Studio doesn't support Klipper at all. If you've got a Voron or a K1 Max, OrcaSlicer is the obvious slicer. OrcaSlicer's Strength tab. Wall loops, top and bottom shells, sparse infill, all in one panel. The dense layout intimidates beginners but rewards power users who don't want to hunt for the setting they need. 3. Built-in OctoPrint and PrusaLink support. One-click "Send to printer" without external plugins. Cura needs the OctoPrint Connection plugin; OrcaSlicer ships it. 4. Massive printer profile library. 200+ stock profiles spanning Bambu, Prusa, Voron, RatRig, Creality, Anycubic, Elegoo, Sovol, Anet, Artillery, Flashforge. v2.3.2 added Creality K2 (multiple nozzles) and Anet A8 Plus. 5. Print-quality features you literally cannot get in Bambu Studio. Scarf seam (curved-surface seam ramping that hides Z-seam on cylindrical prints), Precise Wall, Sandwich infill (alternating wall print order), Adaptive Pressure Advance, Polyholes. These ship as defaults; Bambu Studio still doesn't have a true scarf seam in 2026. 6. Multi-color via Bambu AMS still works. AMS slicing, painting, flushing logic remains best-in-class even though remote AMS state changes broke after the firmware lockdown. The slicer-side workflow is unchanged. 7. Active update cadence. v2.3.0 (mid-2025), v2.3.1 (late 2025), v2.3.2 (March 23, 2026), with public beta and release-candidate cycles between. PrusaSlicer ships ~9-12 months between majors; Cura ships ~3 majors per year. Orca's pace is faster than both. 8. AGPL-3.0 with healthy contributor base. 28,988 commits, 2.2k forks, hundreds of contributors. The 2024 move from SoftFever's personal account to the OrcaSlicer org formalised governance and reduced bus-factor risk. 9. Recent v2.3.x feature stack worth highlighting: v2.3.2: Configurable wipe tower type (Type 2 recommended for MMU and toolchangers), MMU detection via Moonraker DB, security fix for crafted .3mf path traversal, CLI segfault fix on --info / --slice / --export-3mf v2.3.1: Lateral honeycomb infill (33% lighter than lateral lattice at the same torsional stiffness), fuzzy skin painting (paint-on, ported from PrusaSlicer), fuzzy skin via flow modulation, in-app input shaping plus junction deviation calibration UI v2.3.0: Adaptive Pressure Advance per-feature, scarf seam improvements, new device tab UI 10. Modern, multi-plate, multi-instance UI. Faster slicing on big plates than Cura. Cleaner organisation than PrusaSlicer's tab-stack. The dark mode is the default. Where OrcaSlicer falls short (8 honest cons) I'm using GitHub issue numbers throughout this section so you can verify the bug isn't just me. A review without specific evidence reads like astroturf. 1. No SLA / resin support, and no roadmap to add it. Issue #12003 (January 2026) and Discussion #13132 (April 2026) both got rejected or parked. If you've got a Prusa SL1S, Anycubic Photon, Elegoo Mars, or any MSLA printer, OrcaSlicer is the wrong tool. Use Lychee Slicer, Chitubox, or PrusaSlicer's SLA mode. 2. v2.3.2 ported a pile of half-broken Bambu Studio features. Tracking Issue #12684 documents 29 regressions. Adaptive flowrate is visible in the UI but non-functional on non-Bambu filaments (#12688). Internal ribs are ineffective on non-Bambu printers (#12742). Skip points only work on Bambu profiles (#12782). Pre-extrusion distance setting silently ignored. Real impact: settings appear to apply but do nothing. The maintainers are working through these in 2.3.3-beta but the regressions exist in current stable. 3. v2.3.2 wiped user process presets on update for some users. Issue #12962. Workaround: back up the OrcaSlicer config folder before any minor update. Path is %APPDATA%\OrcaSlicer\user\ on Windows, ~/Library/Application Support/OrcaSlicer/user/ on Mac. This burned a non-trivial number of users. 4. Klipper LAN crash on Creality K1 Max. Issue #12340. Print job is sent successfully via Moonraker, then OrcaSlicer attempts a Creality-Print-style /upload/ POST that returns HTTP 405 and the app crashes. Workaround: send the file, then close OrcaSlicer before the second upload completes. Or use the K1C web UI workaround documented on the Creality forum. Or downgrade to v2.2.x. 5. Bambu cloud integration crippled since the January 17, 2025 firmware update. Bambu's Authorization Control System forces all third-party slicer traffic through the stripped-down Bambu Connect middleware. OrcaSlicer can no longer remotely change AMS color, hotend temp, print speed, or read live filament state on post-ACS firmware. Workaround: enable LAN Mode + Developer Mode on the printer (still works as of May 2026), but you lose cloud features and you can't update firmware past 1.08.x without losing this access. 6. Steeper learning curve than Cura's Recommended mode or Bambu Studio's defaults. Settings groupings differ from PrusaSlicer; a Cura migrant typically loses an hour finding things. Beginners who just want to slice a Benchy will hit decision fatigue. 7. Print time estimates drift on Klipper. Long-standing issue (#1969 feature request: time estimate calibration; #820 estimated end time). Estimates can be 25%+ off because Orca doesn't model Klipper's actual motion-planner behaviour. Bambu Studio is much more accurate on Bambu hardware. If you're billing client work by print hour, use the printer's actual reported time post-print, not Orca's pre-print estimate. 8. No mobile companion app. Bambu has Bambu Handy, Prusa has Connect mobile, OrcaSlicer has nothing. Remote monitoring requires Obi [truncated] --- ## OrcaSlicer for Creality K1, K1 Max and K2 Plus: Klipper Setup Guide URL: https://orcaslicer.net/orcaslicer-creality-k1-k1-max/ I picked up a Creality K1 in late 2023, ran it on Creality Print for a couple of months, and switched to OrcaSlicer the day I tried to import a Voron profile and got nowhere. Six months later I added a K1 Max. Both have run OrcaSlicer ever since. The setup story has changed enough between the original K1 firmware, the K1C 2025 silent revision, and the newer K2 Plus that I keep getting asked the same questions. This is the one place I'm answering them all. This guide covers the Creality K-series with OrcaSlicer: K1, K1 Max, K1C (original and 2025 revision), K1 SE, and K2 Plus. Setup, calibration, common gotchas. Written against OrcaSlicer v2.3.2 (current stable) and the Creality OS firmware as it sits today. The big context before anything else: K-series printers ship with Creality OS, a fork of Klipper that diverged from mainline around mid-2022 and hasn't been rebased. That means some Klipper features OrcaSlicer expects (KAMP adaptive bed mesh, EXCLUDE_OBJECT before firmware 1.3.6.0) aren't present on stock K-series until you install the Helper Script. None of this is in the OrcaSlicer documentation. Almost all confusion about K-series setup traces back here. K-series spec reference SpecK1K1 SEK1 MaxK1CK2 Plus Build volume220×220×250220×220×250300×300×300220×220×250350×350×350 MotionCoreXY rodsCoreXY rodsCoreXY rails Y / rods XCoreXY rodsCoreXY rails all axes Hotend max300°C ceramic 60W300°C300°C300°C "Unicorn" tri-metal350°C "Apus" all-metal DD Stock nozzleBrass 0.4mmHardened steel tipBrass 0.4mmHardened "Unicorn" tipHardened tri-metal Onboard accelerometerMPU-6500MPU-6500MPU-6500MPU-6500Yes (newer) LidarNoNoYes (AI LiDAR fw 1.3.3.3+)NoNo (2× optical + strain gauge) AI cameraOptionalNoYesYesYes (dual) BedPEI flex 100°CEpoxy flex 100°CPEI flex 100°CPEI textured 100°CPEI flex 120°C Heated chamberPassive onlyPassivePassivePassiveActive 60°C Default speed300 mm/s (max 600)600 mm/s300 (max 600)600 mm/s600 mm/s inner walls Acceleration20,00020,00020,00020,00030,000 AMS analogNone / CFS retrofitNoneNone / CFS unofficialCFS (orig only, NOT 2025 rev)CFS official up to 4 / 16 colors The K1C 2025 silent revision (read this first) Before any setup, you need to know about a hardware change Creality made silently in 2025. The K1C revision shipped a revised mainboard that: Removed local web UI on port 80 Removed root SSH access Has firmware branch v1.0.0.x instead of v1.3.x.x If you bought a K1C in 2025 or later, and the touchscreen Settings → About → Firmware version starts with 1.0.0, you have the revised hardware. The Helper Script community method below does NOT work on the 2025 revision. Connection setup uses Host Type "CrealityPrint" instead of "Klipper (Octo/Klipper)". Original K1C and all K1 / K1 Max / K1 SE / K2 Plus units use the standard Klipper-host setup. Connecting OrcaSlicer to a stock K-series For K1 / K1 Max / K1 SE / K2 Plus / K1C original, the procedure: On the printer's 4.3" touchscreen, go to Settings → Network → Wi-Fi → connect to your SSID Settings → About → Network info, note the IPv4 address (e.g., 192.168.1.74) Open http://<printer-ip> in a browser to confirm. On stock K-series (pre-2025) you'll see the Creality web UI on port 80. K1C 2025 returns nothing on port 80 (expected, that's the silent revision). In OrcaSlicer, click the Printer Settings gear icon → Connection / Physical Printer Set Host Type to Klipper (Octo/Klipper) Hostname: 192.168.x.x:4408 (the port suffix is mandatory; without it you get the misleading "Update OctoPrint to 1.1.0 or higher" error) API Key: leave blank (Creality OS Moonraker has no auth by default) Click Test. Should return "Connection valid". Save. The OrcaSlicer Prepare tab with the Creality K1 (0.4 nozzle) profile selected. The Quality tab shows Layer height 0.2mm, Aligned seam position, and the K1's signature "Only one wall on top surfaces" enabled by default for cleaner top finishes at high speed. For K1C 2025 revision, set Host Type to CrealityPrint, hostname 192.168.x.x (no port). The Device tab will show 404, which is harmless; uploads still work via Creality's API. Port reference K-series ports that matter: Port 80: Creality web UI (stock pre-2025) Port 4408: Fluidd via Creality nginx proxy. This is the OrcaSlicer connection port. Port 4409: Mainsail via Helper Script (after install) Port 7125: raw Moonraker. Often firewalled by Creality on stock firmware. If port 4408 returns "Connection refused", your firmware is too old (≤ v1.2.x) or you've installed Mainsail (which moves to 4409). Update firmware via touchscreen, install Helper Script, or point OrcaSlicer at the right port. Common K-series connection errors "Update OctoPrint to 1.1.0 or higher" This is the most common K-series error message and it's misleading. It means you forgot to append :4408 to the IP address. Add the port suffix and the error goes away. HTTP 405 / OrcaSlicer crash after sending print Active bug as of February 2026, GitHub issue #12340. Affects rooted K1 Max running vanilla Moonraker on OrcaSlicer 2.3.2-beta+. OrcaSlicer attempts a Creality Print legacy /upload/ endpoint after the Moonraker upload succeeds. Workaround: use OrcaSlicer 2.2.x stable until the fix ships, or disable "Send-and-print" and use "Send only" then start the print from Fluidd. "Unauthorized" / 401 after installing Helper Script Cause: Helper Script Mainsail/Fluidd has CORS lockdown enabled by default. Edit ~/printer_data/config/moonraker.conf on the printer and add: [authorization] cors_domains: *.local *://* trusted_clients: 192.168.0.0/16 10.0.0.0/8 Then restart Moonraker. OrcaSlicer can talk to it after that. "404" on file upload Wrong port. Used 80 instead of 4408. Or installed Mainsail (4409) but pointing at 4408. Check what's installed and align ports. Device tab blank but slicing works Expected on any K-series in stock config. The Device tab needs Mainsail or Fluidd installed via Helper Script for full live console. Without that, OrcaSlicer can send G-code but can't show real-time temperatures, manual G-code commands, or bed mesh visualizer. The Creality OS caveats vs mainline Klipper Creality OS is a fork of Klipper that diverged from mainline mid-2022 and was never rebased. Things that affect OrcaSlicer behavior: No EXCLUDE_OBJECT before firmware v1.3.6.0. OrcaSlicer's "Skip Failed Objects" feature silently no-ops on older firmware. Fixed in newer firmware. No BED_MESH_CALIBRATE ADAPTIVE=1 in stock builds. OrcaSlicer's adaptive bed mesh checkbox does nothing unless KAMP is installed via Helper Script. SET_INPUT_SHAPER works, but the printer reverts to defaults on next boot unless SAVE_CONFIG is run. OrcaSlicer's input-shaping start G-code injection is per-print, so values persist only for the current job. Pressure Advance override on K1 Max: Creality runs PA = 0.04 by default and overrides slicer values via their own PA macro on K1 Max LiDAR firmware. To use OrcaSlicer-calibrated PA, untick "Enable Pressure Advance" in the filament profile, OR add SET_PRESSURE_ADVANCE ADVANCE=0 after the START_PRINT line. Old Moonraker version (~0.7.1 era): some Orca features that need newer Moonraker endpoints (per-object cancel, thumbnail upload v2) silently fall back. Filament sensor / runout uses Creality's binary GPIO, not Klipper's [filament_switch_sensor] syntax. Don't try to override it through OrcaSlicer. If you root the printer and install vanilla Klipper (the ballaswag mod is the most popular path on K1/K1 Max), all of the above goes away. But you also lose Creality's AI features and the touchscreen UI. Most users keep stock firmware plus Helper Script and accept the trade-offs. Calibration sequence for K-series Order matters because some steps overwrite others: Step 1: PID tune hotend and bed (one-time) From Mainsail or Fluidd console: PID_CALIBRATE HEATER=extruder TARGET=240 PID_CALIBRATE HEATER=heater_bed TARGET=70 SAVE_CONFIG Stock Creality OS does this at first boot but values drift after a hotend swap. Re-run after any hotend change. St [truncated] ---