Example of RAW file from iPhone 15 Pro Max - Noise Remove only…
This tint is unacceptable…
1 Like
Agree with you, they were working fine for me.
My preference is Both!
I’d like to start with the Comparison View to choose the model to work with - like I do in the previous products.
Then, once I pick the model that produces the best initial view (comparatively), I move to working with the Split View for actual settings refinements.
That’s the value of Comparison View - it’s for making that initial choice of model side x side vs the other model options.
3 Likes
I agree, I’d like to have both the split view and comparison like in topaz sharpen etc. Also when processing a batch of files, the other applications give a progress report (eg saving file 12 of 34 files), but this is missing in AI, can it be introduced.
5 Likes
Lenscorrection for Olympus ORF Files does not work in my case with PhotoAI. Is there a possibility to reload or reinstall lenscorrectionfiles?
I am using Windows 10 Intel CPU 7700 with Radeon VEGA56.
I think AI is getting too “fancy” now - it seems to be making things more unrealistic than ever and I find myself using the older DeNoise more and more. This is not the case with everything for sure but is happening a little too often now. For instance I took photos of a Morepork Owl on Monday night - as it was night, some denoise was required for the high ISO used. I deleted nearly every AI photo and all are just run through DeNoise low light - much more realistic photos than the smear that the feathers turned into with AI -
Saying that, for some photos it works wonderfully
I would like to suggest a feature enhancement. When upscaling, the default value is set to pixels and I operate in inches. Can there be a toggle in preferences where I can set inches instead of changing the value every time I upscale? And the same for PPI? I always upscale to 300 PPI. Thanks.
Ich habe nun genug mit 2.0xx ausprobiert …
und bin mit meinen Macro-Fotos zurück zu 1.5.1 .
Anfang Nov. ist mein Upgrade fällig, ob ich das kaufen werde steht momentan in den Sternen 
I can relate to that. I often use the even older version 1.3.2, since the detected autopilot values there make more sense in some situations, as seen in this example: Question about autopilot values - #3 by PWC
I have the feeling: the newer the version of Photo AI, the more the look goes towards a more artificial look (people’s skins more plastic, less remaining detail,…). Shouldn’t it go in the opposite direction?
I just downloaded v2.0.3. It crashes while doing the first task (denoising). I tried several different images and different file formats (tiff and jpg). Same result. So I went to your page and tried the download again - this time with the repair option. The program still crashes. Suggestions?
I’m also back at version 1.5.1 as everything after this became incredible slow and the results are often unuseable
You could try a complete uninstall (when it asked to delete settings and models let them be deleted) and then just to make sure also manually remove the complete c:\ProgramData\Topaz Labs LLC\Topaz Photo AI folder and then do a fresh install. It should then detect your graphics card and download the models for your hardware. If it still doesn’t work then I would think you found a new bug (which you can post with sample images).
Ja, schade, ein ähnliches Gefühl habe ich auch. - es wird von Mal zu Mal schlechter…
Die Version 1.3.2 werde ich mir nochmals ansehen (Danke) - denn LensBlur und MotionBlur sind nun, breits bei allen Schärfe-Reglern auf Stufe 1, von vornherein zu krass eingestellt.
Da benutze ich lieber wieder das gute alte Denoise - die Ergebnisse sind dort besser.
Thanks. I uninstalled v2.0.3 then reinstalled it. Same issue. It just crashes doing the first task. This is why I loathe product updates. At some point you get a version that doesn’t work. And honestly - I’ve had Topaz AI for a year and a half and although I faithfully do each update, I can’t say I’ve noticed any improvements. I see that v2.0.3 now has lighting and color balance features, which might be useful, but I imagine I can fix those issues with other non-AI programs. Now I’ll see if I can just go back to the previous version that worked just fine. Btw, I’m using Windows 10 on my Lenovo PC.
1 Like
I agree. The only reason I am using 2.x is that they added some additional autopilot presets (which are stored in the registry). That allows me to use the CLI with specific models (using the workaround to apply the saved reg files before starting the CLI exe) - less than ideal, but better than nothing… But I still can’t force “always sharpen” and auto-detect never works…
Being a senior software engineer myself with a few decades of experience I have to say that that doesn’t have to be the norm. You can incorporate functional and integration tests, that ensure the functionality. Of course you can’t predict every eventuality (new picture input), but if you continually cover your bases, chances are very good that new versions still work. And the CLI is an excellent base, which Topaz Labs could use to build tests. I suggested something in this idea:
But I can only suggest things 
1 Like
I’m finding I choose Standard sharpening over Standard v2 more often than not, for whatever it’s worth. It seems to balance sharpening with looking natural a little better.
1 Like
Thank you for the detailed answer. But… Yikes. I’m a photographer not a software engineer. What you wrote went way, way over my head. I need software that works and is easy to use. It looks like the previous version of Topaz AI is not available for download (none are for PCs, unless I’m missing something). So I’ve written to Topaz support. Hopefully they can help. Otherwise I’ve got a $200 program that doesn’t work.
Every previous version is available to install. Your PC probably runs windows or mac.
You can download them here: Topaz Photo AI - Topaz Community
Go to the version (thread) you want, e.g. " Topaz Photo AI v1.5.4" and then the top post:
If I understood correctly Standard v1 might be gone soon, as they are planning to replace it with v2. At least I think I read something along those lines…
