Topaz Video AI v3.3.3

weird…

In properties, select ‘device manager’ and check if the adapter is switched on or not.

Potresti eseguire due istanze dell’applicazione o renderizzare due metà del tuo video contemporaneamente attivando il massimo di processi “2” nelle impostazioni. Questo potrebbe velocizzare l’elaborazione!

Avete un tipo di scheda minima da suggerire per avere prestazioni accettabili?
Grazie

Depends on how much money you want to spend but at least an RTX 4070 that will cost around 790€ or if you don’t want to spend so much maybe an RTX 3060Ti that will cost less 300€. Even with an RTX 3060Ti I’m sure you will get at least twice the performance.

You are welcome

How about .avs file support for those who want to streamline their production pipeline with AviSynth scripts? This was never officially supported but at least it worked until v2.3.x. Version 3 completely removed it and that still stings.

Not so soon… But we can always use some workarounds… https://community.topazlabs.com/t/how-to-load-avisynth-scripts-into-tveai/
I’m currently using AVS2PiPEMOD64 to pipe AVS scripts to Ffmpeg

1 Like

I´m getting “ERROR” at certain point of certain videos when using GAIA. I´ve tried both on an Intel IGP and on my AMD RX 580 4GB. Both have the same issue. I´m using it to upscale 360p / 480p to 4K music videos.

Here goes my error log:

If you need anything else just tell me!
logsForSupport.tar.gz (124.3 KB)

Thanks for the tip , 3.1.9 runs great with no problem with Intel Arc a770.

1 Like

I have a similar issue with an i9 10th gen. Here are the specs
specs

More of a requested change than a bug, but:

  1. Minimize Topaz Video window
  2. Type Command-W w/o changing focus to another app
  3. Watch your job in progress quietly go away.

Very maddening. Command-W has a huge impact, and an “are you sure” dialog would be appropriate at a minimum.

In versions 3.3.2 and 3.3.3, there are 2 major bugs. Firstly, if I want to import an image sequence, for example a PNG sequence, the software crashes instantly. If I want to export an image sequence from a video file, well, I don’t get the preview generated on the right. It’s really very annoying. I don’t know if this bug was present on 3.3.1, but I don’t think it was on 3.2.9 and maybe 3.3.0.

Hello, is there any option for reduce RAM usage when export it ?

Feedback
I’d like to see a photorealism filter to process characters and architecture. Things like sub surface scatter and displacement(an ai corrective feature to grow/pop out wrinkles, pores, target moss when scene details seem flat and dull close up. This kind of feature would definately change the scope of requiring a photorealistic renderer. The AI could potentially cartoonify buildings, characters, lighting, etc. Maybe the features exist already? Haven’t yet fully explored the program.

When I open Video Ai, the “trial yellow box” shows up for several seconds - then turns to the feedback button?

WHY, when it’s a purchased copy and account is signed in?

Is this only occurring with Chronos?

Please post in the ideas section of the forum!

Not reproducible locally, appears to be machine or file specific. Please open a support ticket and we will look into this :slight_smile:

Yes, lower memory usage File > Preferences

This is expected and what you are seeing is the app connecting to the server and confirming the log in

Strange. I didn’t have this problem with previous versions.
My configuration is as follows: Windows 10 Professional 22H2
Intel(R) Core™ i9-9900K CPU @ 3.60GHz 3.60 GHz
32 GB ram
RTX 3070 8 Go.

Here are the crash logs
2023-07-10-20-37-39-Main.tzlog (24,1 Ko)
logsForSupport.tar.gz (9,0 Ko)