@Carsten_Pfundt, you’re exactly the user McNeel should pay attention to. Ready to switch, just waiting for this one dependency to resolve. There are lots of reasons to make the jump. Personally, workflow efficiency is huge (tiling window managers like Hyprland keep me in flow vs. constant window shuffling). But as an entrepreneur, digital sovereignty through open-source has become a strategic necessity, not just preference. And that’s the key point for McNeel: In Europe, this isn’t a niche hobbyist request anymore. It’s becoming a business requirement - public tenders, institutional policies, even corporate IT strategies and my small company are shifting toward open-source infrastructure. The users McNeel risks losing aren’t leaving because Rhino isn’t excellent. They’re leaving because the platform dependency conflicts with organizational strategy. That’s an opportunity McNeel could seize - or competitors will.
Windows developer explains why Windows 11 is garbage: https://www.youtube.com/watch?v=C44iCr6czAo
While probably interesting I don’t think this is the correct thread for that.
what are the specs on your machine so others can see if it’s feasible to run rhino?
IF the object were to run Rhino on a UNIX-class operating system with an Open Source foundation, then Rhino has already been running on such an industrial-strength operating system for several years now. It’s called MacOS (since Rhino version five).
It’s hugely interesting, its about privacy (and the absence of it in Windows 11). The former Windows developer is now largely a Linux Developer.
A Linux compatible port of Rhino would enable larger distributed cloud workflows. Removing Windows server from Rhino compute/Rhino would make management of Rhino compute so much easier with the ability to spin up Ubuntu spot instances and imaging without all the overhead of Windows Server. I think this would open up new revenue streams for McNeel in the cloud computing market for offices that are creating centralized computational analysis workflows using Rhino Compute and don’t have the staff/interest to write their own geometry libraries.
Note that Rhino Compute is different from a Rhino desktop app.
would you mind sharing your machine’s specs so that others could compare it to their own machines?
Who are you asking?
I can share my case and 2 cents…
I use Windows only for Rhino as well. Off-line that is. Occasionally connecting to the internet for the licence (and update check and download). I do not email, nor surf under Windows (and I dislike its UX anyway).
ALL my desktops run Linux only (in the past SuSE, OpenSUSE, and nowadays, Ubuntu, before Linux, Unices like IRIX and OSF) for normal work, writing my own code, running simulations, post-processing etc etc, and laptops (always dual-boot because of Rhino). My last MS Office came bundled with a Sony VAIO laptop running W98. No, I do not use MS Word, at all… all my (large) documents, some of them with hundreds to thousands of pictures, are made with LaTeX. For a letter, I use StarOffice → OpenOffice → LibreOffice.
As Rhino also runs on iOS it should be easy to port it to a (Debian) Linux variant. Mr Steve Jobs grabbed the ‘other Linux’, called BSD, Berkeley Software Distribution, an open source Un*x variant, and adapted it for his NeXTSTEP computer. When Jobs went back to Apple, Apple acquired NeXT. Today’s iOS is based on BSD, FreeBSD, and NeXTSTEP.
McNeel could look into Snap, Flatpak, and AppImage, as universal package management systems. I kinda like, and use AppImage (for programs like Darktable, RawTherapee, Siril, XnView). Other codes (for example exiv2, libraw and Gmsh) I download the source and compile it myself.
I would really, really, REALLY, be happy with Rhino as an AppImage.
it would be person i originally replied to (ie you).
however, it would be good if everyone who successfully runs rhino day-to-day in a virtual environment to post their machine specs. baby steps towards benchmarking?
I wasn’t sure if the reply was to me or not, since you had replied to the whole thread instead of one of my posts. Anyway, here is an overview generated by inxi:
╭─2025-10-23 17:07:57 0 1996 3
├ (py3135) jesterking@jesterAmd24ThreadsKub:~$
╰─◒$ inxi
CPU: 12-core AMD Ryzen 9 7900X (-MT MCP-) speed/min/max: 3007/415/5738 MHz
Kernel: 6.17.0-5-generic x86_64 Up: 4d 1h 43m Mem: 14.03/125.01 GiB (11.2%)
Storage: 22.74 TiB (32.2% used) Procs: 555 Shell: Bash inxi: 3.3.39
i meant here, reply 10 :
also, how much of your physical system do you dedicate to running windows?
Wow, that’s a lot of memory and storage space. I’m impressed.
Most of it is just external drives on USB ports - just a couple of 2TB and 4TB M.2 SSDs in external casings. Actually my main Linux is on one of those, I boot it directly from such a device.
I built it several years ago with running lots of virtual machines in mind. Not that I have needed it much.
Right now 48GB, but I’m planning to drop that down to 16GB unless I think I need to run huge models.
But for those who can build their own system I would find a cheap 3xxx or 4xxx series card second-hand and use that to dedicate to the Windows VM and keep your nicer card for Linux. Probably a 2xxx would still do fine if you don’t do rendering as the main focus, like I have (since I do Raytraced and Rhino Render development).
I’m not trying to be negative as I think a Linux version of Rhino would be good. But in the realm of most user’s reality, much of the commentary/words/terms used here will be “tech-ese gobbledygook”. In my world, most people never heard of Linux, and of the few that have, none would ever think to use it. The majority are on Mac, “because it’s easy, always works, and I don’t have to know all the tech stuff like windows.” Just the mention of multiple Linux distro names glazes over eyes – no matter how many YouTube videos explain which is best for whom.
This discussion also seems to revolve around Windows, and ignores Mac users. I wonder how many Mac users are in the world, and how many could be convinced to switch to Linux? Most don’t know the difference between a CPU and GPU, and don’t want, or need to know.
When I installed Mint, my fingerprint reader and webcam would not work. (They do on the Windows partition I later made) I read everything I could and downloaded and installed every possible fix. This required getting to the depths (for me) of the OS, and Terminal typing. With my limited knowledge, it was all meaningless to me. I just needed to get every character and switch right. But… like many users I know, I’m a two finger typist. I cannot type in terminal without a misspelling, or space/slash error, etc. I did get the fingerprint reader to work once, and then on further restarts I always got errors and had to resort to the Text password. Eventually, I removed all the “fixes” and just gave up on having the camera and reader work on Linux. That took about a week.
On Mint, initial installations are easy with the App Package Manager. However, even with these improvements, I cannot use Linux without NEEDING to use the terminal. Maybe I’m wrong and missing something?
Other than Rhino and Reaper, all my other software is open-source. Pretty much Nathan’s list above.
Again, I’m not trying to be negative. I’m just pointing out, that what may seem easy to the mavens here, is very complex to the average user. People who click an icon in MacOS and Windows and things just work for them.
Edit: 11/14/25. I just updated Linux Mint and the fingerprint reader and camera are now working. Hmmm? Progress.
Edit: 01/26. Fingerprint readers have stopped working on both Linux Mint Machines.
@CalypsoArt You’re absolutely right about the complexity barrier. That hardware experience you describe is exactly the friction that keeps people away. But this thread isn’t about converting everyone - it’s recognizing that **Linux adoption is already happening** (for performance, workflow, or preference), and strategic factors in Europe are now accelerating this trend.Universities, public institutions, and European companies (like mine, small engineering firm) where platform independence is becoming important. Hardware support and UX have improved dramatically - modern Ubuntu/Fedora handle most hardware out-of-box, and many professional users never touch Terminal. **The ask:** Support users already migrating to Linux, not convert new ones. Like Mac support - McNeel doesn’t need everyone on Mac, just enough users with valid reasons. But your point stands: if McNeel does this, UX needs to be “click icon, it works.” CodeWeavers’ PortJump (mentioned earlier) could achieve that.
Hey everyone!!
I’ve really loved this thread, even though I still have a lot to learn. I’d like to contribute to the discussion based on my experience as an architect, a digital product designer, and after working on Linux for a few years now.
I started back in the MS-Dos era. Around 2006-2007, I made my main migration to Linux (Ubuntu 6.06 LTS). At that time, trying to work as an architect and designer on Linux was a significant challenge… a reallyyyyyyy significant challenge. In fact, back in 2007, I published an article titled “Software Libre in Architecture and Design Careers,” exploring exactly those difficulties and opportunities.
Since then, as you all said, “a lot has changed.” The ecosystem has matured incredibly. However, to have a productive discussion about Rhino on Linux, I think we need to start with the “why” and the “what for,” and break the challenge down into two distinct tracks: the challenge for McNeel (the product) and the responsibility of the end-user (us).
**McNeel’s Challenge (**The Product: let’s always remember that as spectacular as Rhino is, it’s a digital product with a value proposition, this is just my personal opinion, of course):
Porting an application of Rhino’s complexity isn’t just a technical challenge; it’s a massive strategic decision.
-
The Steam Comparison: Steam and Proton get mentioned a lot. It’s a fantastic success story, but we have to remember that Valve (Steam) didn’t just adapt the software; they invested millions in developing Proton (based on Wine) and created their own optimized distro: SteamOS. I doubt McNeel is going to create a “RhinoOS” (which would be awesome!!). Therefore, relying on Wine/Proton will always be a short/medium-term solution, and it will always come with its own stability issues, especially with plugins.
-
Which Linux to Target? The “fragmentation” argument is real for a developer. It’s one thing for the community to get something working; it’s another for a company to officially support it. If McNeel decided to make a native port, it would have to pick its battles. A realistic roadmap might look like this (and this is just my personal opinion, of course):
-
Option A: Officially support a specific version: Ubuntu LTS (Long Term Support). This is what many CAD developers (like BricsCAD) and professional software vendors do. It’s stable, predictable, and has corporate backing (Canonical). Full disclosure: I don’t use Ubuntu, nor do I work for Canonical.
-
Option B: Target RHEL (Red Hat Enterprise Linux, or its derivatives like Rocky Linux/AlmaLinux). This is the standard in the VFX and animation industry, but it has a slower software update cycle, which could be a problem for new Rhino versions.
-
Option C: Target universal packages/containers like Flatpak. This would solve the distribution problem, but it adds another layer of complexity and potential graphics performance or hardware access issues.
-
The User’s Responsibility (Us)
As a Linux enthusiast, I’d be the first to celebrate a native Rhino. But I also have to be realistic with users considering the switch: Linux gives you incredible freedom, but that freedom comes with responsibility.
-
Pros: You have total control over your OS, better security, (potentially) superior performance, and access to a wonderful FOSS (Free and Open Source Software) ecosystem for other tasks.
-
Cons: You largely become your own tech support. The learning curve is real, and hardware compatibility is the most important factor.
A practical tip for those who want to test their current hardware’s compatibility with Linux:
Before you format your machines, I invite you to test your hardware. Boot a “Live” version of the distro you’re interested in (Ubuntu, Fedora, etc.) from a USB drive. Or, if you have a spare drive, just use that to avoid any dual-boot complications. Open a terminal and run this command:
sudo dmesg -t -l crit -l err | sort | uniq
This command shows all the “critical” and “error” messages that the kernel (the heart of “Linux” 
) has logged since booting up. sort | uniq simply cleans up duplicate entries.
How to interpret the results?
-
If you see nothing or just a couple of lines, that’s an excellent sign. Your hardware is (probably) 99% compatible.
-
If you see a long list of errors (commonly related to
ACPI,NMI watchdog,iwlwififirmware, ornvidiaerrors), it means the kernel is having trouble communicating with your hardware.
Even though many of these errors can be fixed, if a machine shows a lot of issues right out of the box, it might not be the ideal candidate for a stable, professional work environment where you just need things to work.
My takeaway:
For years, my professional workflow was a constant juggling act: Linux (Ubuntu, Debian and later Arch) for personal works, development and general tasks; macOS and Windows (yep :)) in the office.
In short, the challenge is both technical and strategic for McNeel, and a challenge of adaptation and technical responsibility for the user.
I’m all for having Rhino on Linux, but I think we need to be realistic about the effort required from both sides.
Happy to help anyone who wants some advice or a “How-To” based on my limited experience (seriously, still so much to learn). It’s 100% free, and that’s the best part of the Linux community .
Thank you for sharing this!
These were always the reasons why I do not like to work with Windows (and yes, I too had the MBR overwritten and I had to reinstall grub). TPM is switched off in the BIOS, so is “fast boot”. I work offline to prevent unwanted (premature) updates (and intruders — connect, get the Rhino licence, disconnect). No Bitlocker, no MS Office (365), Teams, Excel, no surfing… nor an X-Box. I do NOT want to use, let alone rely on, co-pilot “AI” (a Dunning-Kruger regurgitation technology which results cannot be trusted). “a true AI companion, hear what to hear, see what you see, … talk to you… seamless fluid very, very smooth conversation interaction”? That means continuous pushing data to MS. No, thank you.
I really love using Rhino, albeit I use only a subset of its possibilities, just cleaning up of CAD data, simplifying the design for my (simulation) purposes. The data is downloaded using Linux, then reboot to use Rhino, followed by the rest of the work under Linux again.
I really, really, REALLY, love the idea of using Rhino natively under Linux (as an AppImage-like package).