Free indeed 
Still having issues? We’ve had crazy server load recently so we’ve been uh… balancing our flux capacitors. Let me know if it works on smaller images too if you’re struggling with large uploads
1 Like
Thanks for your reply, I will do so. I was using Studio with the LrC plugin, but now tried as standalone and things do work.
@michael.basler - Scale factor to get to this pixel size would then be 1.56, can you input this value and let me know if it works? Can you also send an email to support@topazlabs.com with screenshots of this issue and logs (Help menu > Open logs folder) and we can create a development team ticket for the comma issue and see if we can add Dimensions in pixels as well in the UI. And thanks for flagging this!
P.S: the forum tag issue in the new section should be fixed now. But this releases thread is still a good place to post issues!
1 Like
Thanks - I figured that was the issue. I’ll try again in a couple days when things “settle down”.
We updated the AI engine for more efficient processing, glad to hear it’s noticeable!
I tested the subject mask on my device and it seems to perform fine on both Mac and Windows. Are you still having this issue in later sessions?
1 Like
Are you using the application as a plugin? You may need to give the host application Full Disk Access in your security settings so that the application can access the models as a plugin.
To find Full Disk Access go to the Apple menu > System Settings > Privacy & Security > Full Disk Access and use the + button to add your host application (Photoshop, Lightroom Classic, etc).
Then try using the plugin again.
What region is your device set to? We have an issue with foreign locales on Windows showing incorrect characters.
Looks like one of the new models is redownloading because the file size is wrong. Please go to this file path and tell me what the file size for the following file is:
C:\ProgramData\Topaz Labs LLC\Topaz Photo\models
irwn_dit-v1-fp32-128x128.pb
It should be 5.29GB.
My (Win 11) PC ist set to German - and I don’t want to reset it to another region, for several reasons.
I’ll send logfiles later.
I have 3 active subscriptions, but I don’t have unlimited access to the web based apps, the site wants me to upgrade from the free tier. When will I get unlimited access to Bloom and the other web based apps?
@kiprode - this confirms the transition is still underway for you. The full roollout will take a few more days! Once it’s done for your account, you will have unlimited for Bloom, yes.
1 Like
Amazing, that did the trick: plugin from LrC is working now. Thanks so much!
Am I the only one who finds it weird that an image takes forever to process in Wonder (Cloud or local)? As mentioned earlier, I have a 7900 XTX with an AMD 5900X. No matter what it says ETA 33 seconds… I walk my son to school, come back, get a coffee, sit down and 20+ mins later nothing… Maybe an hour later it’s finished? What’s going on? These aren’t big photos and just photos (JPG) for socials. Is there a bug? Am I missing something?
1 Like
The question then is, if I am going to get a new Mac (my current one is intel) what specification should I be looking at to run Topaz Studio ? Want to build some future proofing in too.
Here it wasn’t that slow (on a Mac, so your rig should be faster).
Although I wasn’t too impressed by the results.
Both, Redefine and Recover did a better job here.
But then I just tested three images…
I switched from my AMD GPU (was high end 3 yrs ago when purchased) to NVIDIA two months ago for that reason. I wasn’t able to practically use AMD for the more ‘creative’ AI models.
I have got one more question. When I process an image using Super Focus and Wonder Upscaling afterwards, I see a note “1 feature flattened”, which was obviously Super Focus whic disappeared from the list. When I postprocess using “Upscaling” instead of “Wonder”, “Super Focus” remains.
So, am I right to assume “Super Focus” and “Wonder” are incompatible?
You can’t really “future-proof” a Mac. You buy one at high cost with extra specs, a minute later the next model comes out with a better GPU and double the RAM for the same price… You cannot upgrade these internally yourself like a PC. And several years later Apple will forget ever having produced it and then cut off OS and app support.
And yes, I am mainly a Mac user going back almost 35 years…
Some people advise buying the best-deal base Mac models more frequently, supplementing the smaller internal storage with external as needed.
The silicon Macs will get you into good graces with the new Topaz Studio, and with cloud processing, the local specs shouldn’t matter as much, so long as you meet their new minimum (ie, not Intel).
2 Likes
They still sauter in the RAM, right? I remember an old iMac I had that allowed me to swap out for more RAM, saving like $200 buying it from Crucial… sigh…
Unfortunately it’s their hardware strategy, same with iPhones. Pushing you to buy the latest devices every 3-5 years.
We do what we can to support, but our AI engine is cutting-edge, and often these cutting-edge ML features are strictly NOT supported by old hardware. We work directly with these manufacturers in many cases. I don’t think many companies develop AI models on desktop like we do.
