Now that I finally have a GPU again (went 7 weeks without one waiting for a RMA replacement) I can say that I quite like the v4 changes. The 1/2 preview system works well for me to see what settings produce the best results and while it would be nice to have a live export preview, I don’t need it as once I get the settings I like I’m not going to sit and watch the entire thing being processed.
I still would like to see all the major requested features being implemented (such as scene detection and pause/resume support) and of course continued bug fixes/polishing, and new models!
I purchased a year of extension because despite all the groaning in here, it’s still the best video enhancer available, by far!
Kudos to the Topaz team and please keep improving Video AI, and bringing in the requested features.
You know some people if you give them a pile of gold, they will probably complain it was to heavy…
The majority of the reason why we use this tool, works and it works great. yes, there are some UI issues, previews are not smooth, lacking documentation and some hiccups.
but the AIs, the quality (to the limits of what AI could achieve at this time), overall, it works and it works great and that is what most important.
if we had a great UI, amazing previews, Pause/resume, outstanding user experience, warp speed, etc. but the AIs and quality would be crap, all of those wouldn’t matter…
think about it (whoever is reading this response)
Do you know of the 80/20 rule?
Putting 80% of your energy on the 20% that matters / significance and 20% of your energy on the 80% significance.
Sometimes I have that feeling that people are in that spot.
My biggest complaint against TVAI 4 is I don’t think Nyx 2 works as intended. Unless it really is not intended to be used on DVD resolution sources. Some official statement on the matter would put my mind at ease.
The AI upscaling is solid. I can get it to do what I want it to do. The AI frame interpolation is the best I’ve ever seen. When I use the UI, I never look at the live preview, so I didn’t even notice it was gone. For serious comparisons, I use a script and create several clips.
The UI is still good for quick tests of what can be done to a movie. How much improvement can be achieved. I like the direction the UI is headed toward. It looks like it’s moving in the direction of being able to define different AI models and settings for different sections of the video. I might be wrong about that direction, but that would be the most useful way to use TVAI. It would do away with the limitation of finding the least bad AI model for a whole movie. Or painstakingly splitting the movie into clips, finding the right model and settings for each clip, then joining them all back together.
I remember that being said about Nyx. I decided to try it anyway. Nyx with 0s in all the sliders, doesn’t do much to DVDs. Nyx 2 on the other hand, oversharpens them at those and any other settings. If Nyx 1 had the ability to only adjust the noise removal with the Relative to Auto setting. It would be perfect. I was hoping Nyx 2 would be that. It’s very not.
