Version 2.0 has tone and light adjustment as beta version. However, if those new features are not even turned on, I can hardly process a one photo without crashing. If there are more photos than only the one for processing, the crash is occurred and the software shut down.
Steps to reproduce issue:
Topaz Photo AI [v1.x.x] on [Windows/Mac]
Seems to crash due to face detection 
.
Same problem but on Windows 11, I haven’t been able to export a single photo yet without hard crashing.
One more thing noticed:
New features are not yet included in a bulk processing. Only the very first photo is processed by ligth and tone adjustments.
Hoping, that also those nice new functionalities are added to bulk processing, too.
Same Problem, crashing as soon as you view pics, or if you just click save all, it got thru 3 pictures
HI Purkka0206,
I am running Windows 10 and I have no problems at all !
Everything is working fine.
I am running also latest NVIDIA drivers (537.13).
Could it be possible it comes from graphic card driver ?
Kind regards.
Vincent.
Same crash also with Win11PRO 32GB and INVIDIA 3060TI, last 536.99 Studiodriver. So i hope Topaz give us a update. In the meantime, I am trying to install the old version again.
However, the first impression was very positive and brings better results than the old version.
I too was impressed with the new features and how useful they were. Cannot wait to get them work in bulk processing.
Exactly the same problem here too. I’m in Win 11
If I open Topaz to work on an individual image, it closes itself down midway through enhancement BUT if I open Lightroom and choose to edit a photo from within that programme, it works perfectly. Weird and extremely frustrating
I experienced same. What I also found, that Photo AI works pretty fast with LR.
I have not managed to process a single photo since the update. A day after the update completely hosed my installation, I received an email telling me the update is available. Someone is asleep at the wheel.
Version 2.0 stürzt bei mir bei nahezu jedem Bild mit Gesichtern ab.
I used sharpen, denoise, color balance, light adjustment separately and together fine. Takes a while to load them initially but didn’t have issues. But as soon as face detection is turned on and begins to attempt to generate the results, I get a quick crash to desktop. I’ve attempted 4 times now.
I’ve tried many different driver versions for my Nvidia RTX 3090 and have had the most success with the older 835.98 Studio Driver. It lets me batch-process up to 74 photos with 2 faces before crashing. Testing a photo with 20 faces still crashes. Newer driver versions either crash immediately or after 1-2 photos.
For the 535.98 driver, Google: nvidia studio driver 535.98
For my 20 faces Top Gun test photo, Google: CinUIp_U4AABkHP
Me too
Same here on Windows 11
Nvidia driver updated, but still same issue occurred if face detection is used.
If face detection is turned to None, no issues.
Best regards
Matti Purojärvi
Hi purkka0206,
Thank you for your quick answer.
On my side, I do not experience any issues with Windows 10 and NVIDIA latest drivers.
But I have a friend on another location in France that do experience one issue.
All functions seem to work fine except the Balance Color one!
If he enabled it, there is a warning message just under the preview viewport that I remember as:
An unknown error occurred
Naturally the function does not work but the application is not crashing, just the previewing and the output do not work.
When the function is disabled, everything is okay again.
So if I understood well, in your case it is the Preserve Face functions that is crashing the app.
Strange that everything is okay on my computer.
Do you have a file test to share? Is it JPEG as source or RAW?
In the past, I experienced issues with previous versions of Photo AI and Gigapixel AI for instance with JPEG images compressed with the Progressive Optimized compression.
Kind regards.
Vincent.
Yo, same here…
2.0 seems to have a hard time with face detection and batch editing…
I just reverted to 1.5.4
Hope Topaz team can resolve this soon!
