Are there some benefits to output in 16 bits tiff videos that are 8 bits ? Does it improve the dynamic range, the colors ? for a better color correction postwork.
I tried once a 10 minutes SD 480p clip at 200%, it has bring a folder about 180GB of size 
and to just learn after that that you can’t import 16 bits TIF image on Premiere Pro 
Fortunately VirtualDub2 can do it and then you can export a 10/12/14 bits output video. But does it really worth it knowing you will have a gigantic master file in size ?
I read a post in the FB VEAI beta testing page where someone did a comparison between 8 bit and 16 bit output. If memory serves, his conclusion was that there was a small difference where 16 bit gradations looked a bit smoother, but for him it wasn’t worth the much larger file sizes.
What annoys me is that Gaia HQ and CG often adds unwanted noise to skin textures, especially non-white skin. This makes young skin apper like old skin, and the unwanted dotted texture. The older CG mode has less of this issue as it had more smoothening.
1 Like
VEAI is fine for beta testing purposes and it looks promising, but it’s not ready for serious upscaling yet. I would give it 8 to 10 months of further development; if it’s still not improved much by then, I’m sure another competitor will come along with a better software offering.
Hello,
I had my hopes up for trying this product but it doesn’t seem to work with my computer with the following specs:
GPU: RTX 2080 Super
CPU: AMD 3700x
RAM: 16GB DDR4 3600mhz
OS: Windows 10
I just get the “AI Engine Failed” message and changing the resource percentage makes no difference. Any ideas?
2 Likes
Be sure on Windows to NOT run the program with admin rights.
For DVD quality scanned film, I would say Artemis is definitely amazing for serious upscaling. Results look like actual HD film scans for the most part. Distant objects of course lack detail (i.e., it can’t guess words on a distant sign, but instead generates “fake” text"). But I feel that’s to be expected. Unfortunately, the superior Artemis modes (notably Artemis-MQ) are in version 1.2.0, so I hope users saved their downloads so they can revert to that version.
But things tend to fall apart when scaling anything from standard definition that didn’t come from a film source, such as standard definition DV/Mini-DV (like the 28 Days Later sample clip), digital video shot in standard definition NTSC or PAL, or VHS/SVHS (which is extremely low resolution by today’s standards, recorded on analog magnetic tape). It would be really nice to see additional scaling models trained on such sources added. (I’m guessing they are going to have to purchase equipment… old SD cameras and VHS camcorders from eBay, then shoot footage with those sources along with modern HD sources to create training data.)
Topaz products have surprised me and are beginning to really put the expression “you can’t polish a turd” to the test! 
1 Like
Yes, Topaz has impressed me enough to buy most of their AI products, including this one. I can’t wait for VEAI to develop further into an even more capable and refined product.
1 Like
Will there be Linux support in the Future?
Not sure but i believe there is a online version available.
I am also having it crash like a lot.
Also an RTX Card: 2080
CPU: 8600k
16 GB RAM 2600 Mhz
short version: svchost (22288,R,98) TILEREPOSITORYS-1-5-18: Error -1023 (0xfffffc01) occurred while opening logfile C:\WINDOWS\system32\config\systemprofile\AppData\Local\TileDataLayer\Database\EDB.log.
I can’t paste XML due to the link limit
Hope this can get hotfixed and yes, CPU only works without any issues, it rly just seems to crash with the GPU usage. I am not the only one with crashes that has an RTX. Everybody else has zero issues.
this is the xml friendly view:
System
|||- Provider
[ Name ] ESENT|
| — | — |
|||- EventID 455
[ Qualifiers ] 0|
| — | — |
Level 2
Task 3
Keywords 0x80000000000000
|||- TimeCreated
[ SystemTime ] 2020-04-24T13:37:35.260876500Z|
| — | — |
EventRecordID 21262
Channel Application
Computer DESKTOP-VAL58TU
Security
- EventData
svchost
22288,R,98
TILEREPOSITORYS-1-5-18:
C:\WINDOWS\system32\config\systemprofile\AppData\Local\TileDataLayer\Database\EDB.log
-1023 (0xfffffc01)
That would be good and maybe could speed up the rendering process knowing Windows squeeze something like 2GB of your GPU Memory.
So far it is working pretty well for me, apart from a few small bugs. Where can I report them ?
Also, it would be fabulous to have a “pause” option if this is possible. for some videos, the processing time is quite long, and sometime, I’d like to stop the process, to be able to resume it later (after a computer reboot for updating other apps for example).
Cheers !
4 Likes
I know, I am currently using Windows, but Linux would be more convienient because Windows is not my main os.
That would be great!
I am a new user and enjoying the software so far - realizing very quickly that the render times make certain projects I was hoping to use if for a little…intimidating? I think it would be great if Topaz (or others here with deeper experience with the program) would post a true recommended “Best, Better, Good, Minimum” set of system specs for achieving better render times on projects. What type of dedicated render machines is Topaz using for their online service for example? Is the CPU option designed to work best with specific chipsets/# of cores, is the GPU option significantly improved by moving from a 1080 to a 2080 - what about specific editions of cards - OC, ROGG etc.? What are the actual render time differences using OpenVino? If I wanted to build a decent dedicated render box for basically just using VEAI and upscaling my old SD projects - what components/configuration would Topaz recommend? Sorry if I missed this information somewhere else. Again - fascinating software - really getting some nice results on my (short by necessity) initial experiments…Hope someone has suggestions! Thanks again!
For my impression the smaller in pixel size the source video is the faster it will be processed, where the overall processing speed will be very similar no matter how far you enlarge the video.
