Did raise a support request and the only advice I got was a fresh install after reboot that unfortunately does nothing.
I experience the same problem with: Adjust AI and Sharpen AI.
For the moment I’m back to the old Adjust and get faster and better results with much more control over detail. Till now I’m not impressed at all with the AI plugins I have, including the latest Mask AI.
AI means nothing to me when the results are not good and the workflow 300% slower.
Application & Version: Topaz Sharpen AI Version 1.4.3
Operating System: macOS Mojave (10.14)
Graphics Hardware: AMD Radeon Pro 580X OpenGL Engine
OpenGL Driver: 4.1 ATI-2.11.20
CPU RAM: 40960 MB
Video RAM: 2047 MB Total, 2047 MB In Use
Preview Limit: 4492 Pixels
Just got an message from “Topaz Operations”:
I’ve noticed on several support tickets that this particular graphics card has been involved in some slowness complaints, and have confirmed with two developers (independently), that the Radeon 580 is apparently known for not operating as expected. It seems to be a flaw in that particular card/driver combination that is known among software engineers. In particular they referenced that it would show it was using full VRAM but in reality was using a fraction of it.
That being said we have an Open GL related update coming that is related to some errant crashing on some machines running Gigapixel AI and Sharpen AI, so it’s probably worth seeing if it affects you as well. We are trying to release that update this week.
If you can hang tight for that update, I think that’s the best next step to see if we can find a resolution for you.
I have a RX 580 GPU with 4GB of Vram on a Windows 10 machine. I noticed today that gigapixel was running slow so I check the Radeon settings. It seems that the video card was stuck in a slow mode 300MHz for GPU and a 300MHz for the memory (usually can go to 2000 MHz). I ran gigapixel and the frequencies stayed low.
I loaded a saved profile and everything came up to normal and this greatly sped up the processing in gigapixel. I wonder if you have looked at the Radeon settings > Gaming > global settings > Global wattman to see if the memory and GPU are going to max speed when needed?