The VEAI uninstaller is not very elegant and seems to leave a ton of orphaned files/folders behind, especially when updating without a complete uninstall first. When the uninstaller failed to remove everything, I ended up manually deleting folders and then used CCleaner for the rest. The new V3 installer broke naming convention so it wasn’t grouped with other Topaz products.
I dunno why but when I want to use the frame interpolation the result looks awful. I get either heavy flickering with strong pixel patterns and all sort of that. 
There are 2 (maybe 4) environment variables that need to be set when running ffmpeg in a shell.
You can find them if you open the shell from Video AI. In the window, set | clip to copy all of the environment variables to the clipboard.
I’m certain on these two:
VEAI_MODEL_DATA_DIR
VEAI_MODEL_DIR
These are also present:
SESSIONNAME
ZES_ENABLE_SYSMAN
I’m writing a batch file to run my Video AI jobs that will have a notification when they’re done - over growl and rocketchat.
It should be including all tracks especially if audio is set to “Auto”. I suspect though the ffmpeg parameters are only specifying one audio track instead of to just copy all.
(The ffmpeg command string Video AI 3.0.0 builds has a number of issues.)
UPDATE:
Topaz Video AI 3.0 does indeed specify “-c:a copy” (well "-c:a" "copy" which is weird), it is not specifying -map 0:a, I’m not an FFMPEG expert but I think it needs that. (.e.g -map 0:a -c:a copy)
I’m on Windows and I only had to set VEAI_MODEL_DIR as per the documentation. I do have VEAI_MODEL_DATA_DIR set, but I think that’s leftover from the early access releases.
I have a python script that has grown into a monster, but it runs all the passes I want of TVAI and encodes the results into something more stream-able.
Thanks for the update and thanks for supporting non-square pixel sources. One note: If you choose to use a preset to upscale from non-square to HD or UHD, the software should automatically assume you want Square Pixel output. Currently, the software defaults to Original Pixel Type, which would give you wonky results.
What’s MKVToolnix GUI? When it comes to video processing, I’m a complete newbie. I’m just upscaling a bunch of lectures and training videos from the last 30+ years, and excerpting just the best scenes. Is MKVToolnix GUI part of Topaz?
No. It´s not related to Topaz in any way. It’s a very known software for remuxing video, audio and subtitles into a mkv container.
You can use google. It’s a good friend.
I like the direction this is going! One workflow issue: I have jobs that are taking dozens of hours. I need to be able to pause & restart them so they don’t completely overwhelm my computer.
Under 2.6.4 I would just stop processing, set the start trim point to wherever the cursor started, then restart processing later. I would then join the separate segments into one output file.
I’m not at all sure how to do this with Topaz AI 3, but it’s unfortunately a deal breaker for me–I can’t afford to tie up my computer for 2 days at a time without being able to pause for video meetings, etc.
How can I get the same effect in Topaz Video AI 3?
Would the workflow be that I would remux just the low-res excerpts I want and then upscale the MKV container?
I suggested it because it does splitting and reassembling very well, by specific frames or timecodes.
Hi !
my view and comment here on the 3.0 is very different, most people here will not agree with me,
but, the feature you ask will be available soon on the software and you’ll not have to do anything like trim on frame 0 to 10.000 etc… like we all did it in 2.6.4.
A “videoai” tag has been added to each videofile processed in VEAI 3 with all the settings used to make it. from what i understood, it will be used for the future feature of being able to stop the software and resuming the rendering / process at a later date.
this new “update” version must be, i think, seen more as a new software… A Video Ai 1.0, instead of Video Enhance Ai 3.0. The software has totally been redone from scratch. seeing it as an update is maybe not the best way.
for now, the only way i see to make what you ask without using a video editor is to use a software who cut the videos in several parts, and render each of them, and after, add them again in one single video until the pause/stop resuming will be added. the timecode / trim way is not good for that actually.
Are you saying the new version does not have the ability to create “In and Out” points on the file you are rendering?
Also, many times I just want to render a 5 minute section of a video file. Do I have to create the mini clip outside of this software?
I hope that is not true. I have 2.6.4 on many Mac Mini M1 machines. I am afraid to check out this new version.
There are in and out points for the section you want to export.
To Everyone:
It would be sooo helpful, that at the beginning of your complaint or comment or praise that you describe your hardware, your file input and output parameters.
Finally — PLEASE tell us your frame per second rate speed that you receive in performance. No one is talking about performance.
Because we KNOW they haven’t really started working on performance yet. There’s not much point when we know it will be changing in the near future. It’s mostly irrelevant right now.
this version doesn’t have yet the speed improvement that everybody think it has /must have/maybe have… it will be added in a future update, certainly not in a very long time, but this speed improvement is not in this version.
Ok - so it is slower from what I am reading.
Is there a significant difference in output quality between 2.6.4 and 3.0?
Many are reporting lesser quality output compared to 2.6.4. I can’t really say. It doesn’t produce anything I’ve been able to use…2.x or 3.x.
on my machine with my files, i have the same speed than 2.6.4 and even a bit faster. but it’s my machine, component, on a windows 11 system (insider beta version) with a 1050ti graphic card etc…
so i can only talk for me. i don’t know what the other have, and the results, what they do etc…
in my case, I uninstalled 2.6.4, I don’t need it anymore. i only does upscalling , and use proteus in Auto or manual. i don’t use it for de-interlacing, and use the fps conversation as well. i don’t do anything else wit the software except for testing during the beta.
about output quality, it’s the same for me, but i don’t have a good tv, not a high end monitor, and i’m not as “picky” about high end quality as some other people here. h264 intel encoding is ok for my use, mainly because my hardware media player doesn’t support any other format for now. certainly if i had a 4K monitor and better gear, i’ll change my comment.
so it’s really a matter of what each experience and want. the best suggestion i can tell is that you try the software and see for yourself.
the software will evoluate, as new version will be available certainly pretty quickly because a lot of issues are reported since yesterday. so we expect/hope something much better soon, for now, it’s working ok for me, but can only talk for me, for what it worths ;).
