I still find it puzzling since pretty much all technology in this space is developed for Linux primarily. No one runs this at scale on Windows. That said, I can see hardware for the desktop marked being more of a challenge than the regular hardware used for computational work. It usually requires a lot of tinkering when building code yourself. Since this is all proprietary tech, they won’t let us do that of course.
Topaz Video AI is a wonderful tool already, but many users do not have access to either powerful or multiple computers. The temptation is to use our main or most powerful computer to do the video rendering, but the downside is that this computer is now tied up, many times for days or weeks (months?) trying to clean up, or upscale, and render one, or more, video(s).
I know many have already requested this, but I’m also requesting a Linux/Unix version which supports both GUI and CLI, much like other video processing projects do, such as HandBrake, and of course ffmpeg (mostly CLI).
Many users have one or more Linux/Unix boxes because it is less expensive, and allows reuse of older computers which have outlived their usefulness as desktops.
If Topaz Video AI could offer a Linux-based version which opened up CLI access, it would allow background, and possibly batch, processing of video files using our older, non-primary computers, so we could continue using our main/higher powered computers for day-to-day projects.
Network Rendering would be another request. If the program has already been optimized for multithreading, it should be well on the way to take advantage of implementing network rendering, so we could use multiple, older computers to help with reducing the rendering time.
The master computer could either be the desktop/workstation (Windows or Mac based… possibly Linux), or could have a Unix box designated as the master for the render queue, to render by itself, or with the help of slave computers to network render the queue jobs. The job or queue itself could be moved as needed back and forth between the desktop/workstation or the Unix master box on the network. To simplify interaction with Topaz on model updates, the designated master would be the computer to pull updated models from Topaz, and could share or handoff those updates with the desktop/workstation.
Of course, any feature/version request is normally very simplistic compared to all of the work involved to get the program working, but the additional features could be monetized to offset the expense.
It would be nice to have all of the features included in a one-program price, of course, but additional levels of enhancement could be offered as an add-on, or a more complete version for additional cost.
This would also position Topaz Video AI to not just be a workstation-only program, but also have the option of being adopted as a workflow component for larger content companies looking to batch process multiple videos.
Thank you for considering these suggestions!
Big post producing company use Linux (think Company3, picture post, fotokem, light iron, harbor).
It’s a shame that we can’t use this technology
Why not ?
There is a Linux version.
Interesting! Do you know where to find the Linux version?
Sign up to be a Beta Tester and it will be avail to you.
I see, thanks for the info. I’ll keep that in mind when I start my next video project.
Edit: On second thought, I might pass for now. I checked the membership requirements for video beta testers and it turned out that you need to hold a valid license for Topaz Video AI. I only have a license for TPAI and it seems too risky to buy a license for the video app if it is only in beta and I guess there isn’t a demo version to try first.
I would very much welcome it if there were a full Linux version (with GUI). Because my hardware (Mac) is getting on in years, I am planning to buy a new workstation. Because apparently even the fastest Mac available can’t match the performance of Nvida GPUs and a lot of the software I need is designed to run on them anyway, I’m seriously thinking about switching to Linux completely. The most important programs, all run on Linux: Davinci Resolve, Blender, DFL, VC etc. Windows is not an option for me (partly because of the questionable handling of privacy).
There is !!!
It’s in the beta section currently.
So the time has come where all my Topaz products are up for renewal. And VAI isn’t on the list for products to renew since this team isn’t prioritizing Linux.
Linux support is the sole reason for me to consider renewing the VAI subscription. The stagnated progress since V4 with the focus on just messing (up) the UI isn’t the type of changes I’m willing to pay for.
Take a page from the Gigapixel team’s playbook and listen to what customers need and their pains. So this will be my last VAI forum interaction until (if) the day would arrive where at least the CLI for TVAI is able to run the models at comparable speed to Windows. The ~100% performance efficiency hit presently in the Beta is a no-go.
Just here to say.
Id love to have proper Linux support.
as i wanna get rid of Windows on my computers before end of win10 support.
and the only thing that is holding me back is PhotoEditing programs. as the once made for linux just dont feel right to me.
I agree, native Linux versions would be great. Having said that, I’ve settled for a slightly less optimal solution i.e. running TPAI under Wine. There are some occasional hiccups and sometimes a new version of TPAI won’t run until a new version of Wine has been released due to some undisclosed change under the hood. For example, for more than a month I was unable to upgrade beyond TPAI v3.0.3, but then (after a few new versions of Wine had been released) I was able to install and run TPAI v3.2.2.
There may also be performance issues with some GPUs, but I don’t find it particularly slow. However, I have not seen how it performs on Windows/macOS so I can’t really compare.
Anyway, in my opinion TPAI runs well enough under Wine and since I don’t need to run the very latest version at all times, I don’t consider getting a Windows PC nor a Mac. TPAI under Wine on Linux is good enough for my needs.
For completeness, I should point out that the install process is a little rough in that it always fails on the first attempt, then you have to kill the installer (e.g. using xkill) and then rerun the installer. Topaz Labs really ought to fix that because I suspect that a lot of potential users will assume that it won’t install when it consistently fails on the first run.
Hi,
how did you manage to install Topaz VAI with wine ? I tried installing the last version 7 on my Debian Workstation with wine 10. I used that kill install trick as well but get lot of dll missing errors. I downloaded them and placed them in the path Topaz was missing them. No luck.
For the Linux client…I was in a mail conversation with tech support of Topaz VAI. They said that it’s very unlikely to receive further updates…
BUT
they will give Linux support for user with valid licenses.
Anyway…
In another thread i saw that it’s not needed to use the GUI and ‘ffmpeg’ cli is enough. If that’s the case then it should be easy to build docker container and run it on Google Colab with cuda support.
Did someone maybe tried that out ?
Sorry, I haven’t tried to install VAI. I would have tried it had there been a version for Linux that was available without committing to buying a license before trying it out.
What I managed to install under Wine was TPAI v3.2.2 (and some even older versions of TPAI).
Thanks for your reply
Yeah i checked https://appdb.winehq.org/ and saw that TPAI seems to work fine.
Well for Topaz Video i will try to write some code for Google Colab ( Cloud ) and see if the Linux client works. I have a Colab Pro sub and can use the NVIDIA GPU ( T4 GPU ). Will see how fast that is.
I will try to rent a GPU Server for testing with Windows OS and compare the Video quality between Version 5.5 ( Linux Topaz VAI ) and 7.x ( Windows Topaz VAI ). GPU renting is pretty cheap and specially if you pay in Euro and rent a US Server 
@Topaz, please consider acting on this “Idea” now, since the prior beta stability and performance issues are no longer blockers.
All you’d need to do is compile TensorRT variants of the later models, which I’ve already done myself as a PoC for the Beta, with resulting performance equivalent to or even slightly better than on Windows.
If you were to just tensorrt compile the currently missing models I’d consider renewing my subscription once more. That’s the only thing preventing CLI use of the product on Linux, something I’d expect is what most users of that platform are looking for anyway (not the GUI bit).
PS. If you’re having problems getting your own models to compile correctly (e.g prob4, iris-2/3), I’m even willing to tell you how to do it, for free.
I’m interested in automating upscaling via the CLI. I’d appreciate it if you could provide CLI support for Linux.
I’m assuming you’re talking about the linux beta. If so, then yes, it is supported, just like on windows.
Yes, I knew there was a Linux version, but I gave up because there wasn’t a way to automate the login process. Since a new version seems to have been released, I’ll have to test it out to see if any improvements have been made.
Based on my test results, it appears that the current version does not work with my RTX 4090 graphics card. Despite trying numerous versions of drivers and libraries, the GPU remains unrecognized. Additionally, the beta version has not been maintained for a year.
