I have noticed the ETA times are very wrong, they get close to correct only when the process is mostly done. Perhaps there can be a way for the program to keep track (locally) of typical time to process X frames at Y resolution given the current PC and the model used - then extrapolate time to complete for the video and even for the batch. Giving more accurate ETAs the more Topaz is used?
If we upgrade or change something on our PC then we can wipe this metadata so the estimates are re-learned.
I had an issue where the Video Enchance AI 2.4.0 kept crashing constantly within few seconds or minutes. I tried pretty much everything I can think of until I finally found the culprit, which was RivaTuner that comes with MSI Afterburner. Uninstall RivaTuner and it works like a charm. IĀ“m using the new RTX 3060 TI with the newest 472.12 studio drivers.
Iāve got a feeling that we wonāt see any major updates before Black Friday. Why? Itās not a good business. Many of the VEAI users bought the software on Black Friday sale, and why not save the best new features after that for another subscription period.
Topaz 2.4.0 in macos Catalina, hangs in āLoading AI modelā. Tried with admin privileges, granting files/disk permissions, does not work. Can not use it.
Iāve got a strange problem since a few days (after i updated to the latest windows 10 version and the latest driver for my nvidia geforce 1070 ti):
whenever i start a new rendering job with VEAI (proteus v1), my whole system crashes after some time (randomly between 1 min and 2 or 3 hours) and windows restarts after one sec. i never had any crashes with topaz or windows 10 so far, but now i canāt use VEAI anymore at all and i donāt have any idea, what i can do.
and iām especially wondering, why not only VEAI crashes, but whole windows.
iām not sure, if this might be a hardware issue, but i donāt think so.
i tried different gpu benchmarks and several GPU intensive games, all is fine, except VEAI.
what iāve tried so far:
i checked CPU/GPU temperatures ā they never exceeded 68 Ā°C
increased the gpu fan speed to maximum ā crash at gpu core temp. of 64 Ā°C
checked power consumption ā everything in normal range
i checked windows protocoll ā it only sais, that windows restartet after non proper shutdown.
reducing maximum VRAM to middle / to half ā still crashing
activated ārecude machine loadā ā didnāt help
turned off undervolting of my GPU ā didnāt help
tried all available older GPU driver on the NVIDIA homepage ā didnāt help
made a complete rollback on windows updates ā still crashing
tried version v2.3 of VEAI ā same issue
i have no clue, what might be the reason for this issue.
do you have any idea, how i can fix that?
Thank you very much
Try pushing your GPU fan to 100% - If something other than the GPU core if overheating (e.g memory) that should fix it. Does the windows event log record any errors before the reboot?
This is the main reason I disable updates on my apps and OSes and test new stuff on a VM or non-production PC. How did you roll back Windows Updates? With Windows or with a drive imaging program like Acronis True Image? I would not trust Windows to roll back or uninstall anything properly.
I had that happen too some time ago. I think it was related to maybe a faulty video source file. VE AI probably did not like variable bitrates for audio and video.
i kind of fixed the issue by limiting the VRAM in msi afterburner and reducing the power limit.
iām still not sure, if this is a software or hardware issue. strange thing is, that benchmarks and games are running well.
