And just like every version since the “gpu enhanced blah blah” this version crashes as soon as i try to load any video.
I’ve never had an issue with any version of topaz in the past 3+ years.
Could you DM me your log files from this installation?
To gather logs, please select Help > Logging and make sure that there is a check mark next to File Logging. Next, recreate the issue and then return to the Help menu and select > Get Logs for Support and attach the “logsForSupport” file to your reply.
Done, would be great if i could get some feedback quickly on this one. I haven’t been able to use any of the new versions that keep being released.
Its taking quite a lot time for Initiallazing resources in Iris model with same resolution settings its over 10 minutes now still not started processing. Estimated time for video output was 12 min when i hit cloud button
When previewing things locally, previews will not be added to the list of queued tasks.
Steps to reproduce:
- Import a video
- Adjust the enhancement settings on it
- Select the “Render Xs” preview button.
- Notice a preview is being rendered, but it’s not added to the
Preivewslist in the middle bottom of the UI.
1 Like
Like 20rushtonj has stated, the preview list in the UI appears to be non-functional. I’ve also received no extra credits for this beta.
EDIT: Once a local export has been carried out, the content of the export list is mixed with the cloud export list when you view the cloud export list:
I installed 5.5.0.2.b but didn´t find those 1000 Topaz Credits.
In TVAI there´s a 0 in the right top corner.
An in my profile in the community and topazlabs.com there´s nothing called Topaz Credits there.
What it´s needed to get those 1000 Topaz Credits ?
So an interesting crossroads you are at here with cloud processing coming in. Give me a quick sec to explain, this is not a negative post.
I have zero clue how they look to making this cloud processing work, but this is a crazy good opportunity to kick start possibly a new bitcoin like opportunity. Selling processing power is a great idea and will allow people to get work done quicker utilizing larger farms. That said, there is a much LARGER opportunity here if you think outside the box for a quick moment.
Make a separate program, or a new mode in current app, that allows people to pool and host up their GPUs to allow them to process customers work loads and get paid in credits/cash for it. Think bitcoin mining or seti and it is not a hard stretch to know that people love to be a part of something more than just running an app, and when they can offer up their own equipment and get paid (in whatever currency) it makes a lot of people rabidly loyal for some reason. I think it would be a totally new side of this app, allowing others to process stuff, obviously you would want to split up the jobs between many users sitting ready so parrallel processing can happen, and the customer pays a flat rate based on speed (which GPUs would be used) and the users get flat $ based on how many segments they processed. I assume the easy way dev side would be to split up the media into many small segments and number them. When they finish they are uploaded back to topaz and then reassembled into a finished product.
This would allow Topaz to provide a cloud service to it’s customers with literally unlimited growth potential at VERY little cost to Topaz beyond bandwidth, storage, and CPU to reassemble the media. The users would be the ones to allow their idle GPUs to earn them some credits or money, maybe those credits can apply towards yearly license fees(of which STILL needs to be dramatically lower priced) or something like that, or obviously they can use their credits to get their own media processed.
This would allow people to use the cloud service WITHOUT having to own a app license, which greatly opens up your customer base to anyone who can afford processing fees. This opens up earning credits to apply towards whatever for users with decent equipment (like maybe a 3000 series or higher type req)
meaning even more users will stay with Topaz or even get even more involved.
Just a thought, but this could really help grow your business while adding bitcoin style rabid fans with a win win method to give unlimited processing power for Topaz with zero up-front cost and unlimited scaling potentional.
I just imagine someone wanting to transcode something that would take maybe 10 hours, but instead paying to cloud processing fee which then splits that media into 100 slices and sends it to a bunch of users farms to be parallel processed and then finish in 30 minutes, which each user gets a few credits and Topaz still pockets like 25-50% outright. This is like free money for Topaz at that point.
3 Likes
Same here: no 1000 credits in balance.
BTW, macOS Sonoma reports a crash each time I close this beta, even if no video is loaded. Running on Mac mini M1 16GB RAM.
2024-11-16-10-38-47-Main.tzlog (76.7 KB)
I can PM send the macOS’ Console report if needed.
Oh, and I get an error when I click on Cloud export:
Also, SDR to HDR disables Coud export.
Same here, 100 credits only due to Gigapixel/PhotoAi subscription ahah 
Hello I have just 100 credits.
only 100 credits for me as well, i do have an active TVIA subscription. Would be great if you could add for testing. thanks!
No extra credits for me - only the 88 left from Gigapixel.
We have thought about this too, but we cannot do it. Not due to a technical reason but due to privacy and security. Also, the large volume of data from the processed videos will result in a lot of upload traffic similar to torrent networks. Network bandwidth is a significant component of the overall costs.
If we were folding protein, making some random calculation etc. it would be fine to distribute the work. Since we are dealing with actual user data, it is very important that the data is secure and can only be viewed by the user.
We do break the videos up into chunks and process them in parallel on our secure cluster, so for larger videos the processing should done in 30mins or less. We are also working on reducing the initialization/queueing time to a few seconds. Between the processing speed up and reduced initialization time even smaller footage should process much faster.
1 Like
For the GP users who didn’t get the free 1k credits to use with video ai beta. Just post here from your video ai owner account and @tony.topazlabs can add the credits to you.
2 Likes
Yes, we should have SDR to HDR in the cloud eventually.
Which makes cloud-rendering de facto undoable, by the looks of it. For one, because (near) lossless formats, like ProRes, require you to then download hundreds of Gigabytes back to your own harddisk (which is extremely unpractical); and the h265 options for TVAI are too minimal (just ‘high quality’, with fixed bitrate). And no support for Rhea?
All-in-all still looks promising.
