Please add Vegas Pro support for OFX plugin! Vegas Pro handles OFX plugins, but Topaz Video AI does not currently appear in the Fx list.
1 Like
Thatās been my experience, too.
It doesnāt matter what I set my output to, it comes ouit at much lower quality. At a quality that negates what the program is supposed to do.
My fix right now is that I export to 16-bit tiff and reassemble with FFMPEG, which is a pain in the a**.
Iāve been experiencing the spinning beach ball on Mac. Iām finding that reloading the projects is very slow as the system is repopulating the old previews. I think there needs to be an option to delete these as I rarely refer back to the older versions. They fill the system disk.
Iām considering deleting and reinstalling 5.2. Perhaps it just me, but Iād be interested to hear if any other Mac users are having the spinning beachball problem.
While I appreciate the ability to load the previous project, I donāt think itās quite there yet. If you are doing critical client work Iād roll back to the previous version.
Also I have some issues with renders with ProRez std which canāt be read as the compiling is incorrect. I hope the next update fixes these issues.
I get more failures than successes now days 
logsForSupport.zip (2.1 MB)
This is a basic Anime Music Video going from 1080p to 4k and 25 to 50 fps aion with protege auto.
Why is the pause button not showing for me? I would love to use it, but itās no where to be found in my version of 5.0.2. I am not using stabilization and I have all three options checked in the export menu.
Same here, I get ERR fps on all models.
How Well Do the NVIDIA GeForce RTX 40 Series Perform in Topaz AI?
Overall, the NVIDIA GeForce RTX 40 Series GPUs do well in the Topaz AI suite, but their benefit highly depends on the specific application you are using. Topaz Video AI and Sharpen AI tend to give you the biggest performance gains when using a more powerful GPU, while Gigapixel and DeNoise AI appear to be more CPU limited.
If you already have an RTX 3080 or above, the new RTX 40 Series GPUs are likely not worth an upgrade for any of these applications. However, if your current GPU is an RTX 3070 or older model, you could see a decent performance increase ā especially in Topaz Video and Sharpen AI.
The question of AMD vs NVIDIA is even more complicated. The RTX 40 Series GPUs have a terrific ~25% performance lead over the Radeon 7000 Series in Gigapixel AI, but AMD holds a smaller 5-14% lead in the other applications. On the whole, this washes out to almost identical performance between NVIDIA and AMD overall, but what it really means is that the faster GPU brand is going to highly depend on the specific Topaz AI applications you are using.
If you are looking for a workstation for Topaz AI with an NVIDIA GeForce RTX 40 Series GPU, we recommend checking out our Photoshop or Premiere Pro solutions page or contacting one of our technology consultants for help configuring a workstation that meets the specific needs of your unique workflow.
Topaz AI Suite: NVIDIA GeForce RTX 40 Series Performance | Puget Systems
I should have known before but good thing is my previous was 3050 RTX..
2 Likes
Quick question: when will there be a promo video for Video AI version 5 like there was for v4?
Two things Iām still unsure about:
-
Am I wasting my time trying to get the Aion frame interpolation model to work on a 3070 8GB? It randomly fails at a very high rate with no helpuful error message maybe 90 percent of the time. But I do like the way it looks when it succeeds, Iām just not sure if it was designed to work with a GPU with only 8GB of VRAM.
-
Is there any way to see a live A/B comparison of the final conversion in progress like it used to be possible to do with versions 2.X of this app? I used to start my conversions and by monitoring the results of the conversion in progress I would either decide to continue or not.
I think thatās a much more efficient system than creating previews that you always discard. Am I fundamentally failing to understand how to use the preview system these days?
I did notice that thereās a āpreview in to outā option. Would that give me functionality like the preview system had in 2.x? Would a āpreview in to outā preview result in a useable end product?
i7 9700K+2080ti+16G,
v3.2.2 input 1920x1080, x2, Artemis, output 8.2fps.
v5.0.2 input 1920x1080, x2, Artemis, output 4.5fps.
laptop i5 13450HX+4060+16G,
v5.0.2 input 1920x1080, x2, Artemis, output 6.3fps.
laptop 4060 > 2080ti ? fuck
Could be partly explained by the vast improvements with tensor core operations between the 2000 series and 4000 series cards. But, yeah, even a desktop 4060 is slower than a 2080ti, (at least for gaming) and the power limitations of the laptop 4060 should have the 2080ti winning by a big margin (you would think).
The most troubling thing I see there is how much better the Artemis performance was with 3.x vs. 5.x.
5.0 seems like an overhaul in some ways. Maybe they have some kinks to workout yet performance wise with some cards.
Itās at the very top of the new Video AI 5 page!
Topaz Labs | Video AI 5ā¢ | Cinematic superpowers. Ultra smooth. Sharp. Steady.
1 Like
@garblah are you currently running Aion with a VRAM limit set?
-
For an 8GB card, Iād recommend about 60% on the VRAM slider since Aion is built with 12GB+ GPUs in mind. This may not solve 100% of Aion errors but depending on the input resolution the VRAM limit could be a good temporary solution.
-
Something weāre actively working on and have planned for the v5 series of updates. In general, we still recommend spot-checking areas of large files using the Preview button instead of exporting and cancelling. This avoids repeated renders and can be used to specifically preview tougher areas of an input (dark scenes, text, film credits, etc.)
Previewing the full In - Out area does produce a readable video file, but those files do not carry audio. If you are editing your clips after processing in Video AI, you could re-merge the audio track from the input.
2 Likes
Running both Aion and Proteus at 4K will definitely require more VRAM than available on the RTX 3080. Iād suggest splitting these tasks by running Aion first and then re-importing that output for Proteus.
1 Like
Nyx is realy slow with resolve, proteus is much much faster.
1 Like
I really hope you will work on AIONās problems related to the use of GPU/VRAM resorts and CPU/RAM for the next versions, as I keep having problems even having a very powerful ASUS TUF RTX 4070 Ti SUPER with 16GB of DDR6X VRAMā¦ if I try to convetire a video in 4K, it is practically impossible: the program immediately gives error!
Of course, I already reported the problem some time ago, but things have not improved at all in these months:
By the way, you can extract the icon file from the original exe by navigating to the ICON folder after opening the exe in 7zip and extracting 7.ico, then you can make your own shortcut with that icon and copy that shortcut over the one in the start menu.
What kind of Mac? Specs? Iām running a Intel iMac 2020 128 RAM, 3.6 GHz i9 10 core, 10 gen, AMD Radeon Pro 5700 XT. 16GB, exporting to a fast UD 90 nvme. I Had 1 crash after several tests and exports. Speed of processing has increased and quality has improved. What are your Topaz settings for Processing. Try reducing your memory allocation, especially if you have it somewhere around 80-100%. I have mine set to 20%. If that doesnāt help try selecting Low Power mode as well.
Export from resolve and then use TVAI to enhance is much faster.
But when doing so, there is a big shift in color, not with enhance in resolve.
Processing speed about 6.3fps
