Yes, I understand that the Linux Beta has been discontinued. It’s all us Linux users have at the moment. I have tested running the Windows version on Wine, and unfortunately Wine does not support tensorrt, so it encounters the same issues as the Linux beta on this card.
When attempting to run the Proteus model on the Linux Beta, I can inspect the command line of ffmpeg and see that it repeatedly shows errors such as RESULT for https://veai-models.topazlabs.com/prob-v4-fgnet-fp16-576x384-2x-rt809-8517.tz is 404 until it is finally able to successfully pull an onnx version of the model, which is presumably because the RTX 4080 Super does not have models compiled for this version of tensorrt. The problem is that the onnx models are 50% slower than the tensorrt models, which I verified by running in Windows natively. But I don’t want to use Windows just for one singular application and have to constantly dual boot back and forth any time I need to run Topaz. It would be massively beneficial if I could actually use an albeit outdated version of the application on Linux with the same processing speed as I can on Windows, instead of continuously trying to find an alternative that works on Linux properly.
