Max resolution

I have a scanner for photos, which ofter gives out pictures more than 45000x14900, I can not open it in topaz photo ai. Is there a limit in max resolution? or can I solve this by increase memory?

Topaz Photo AI [v1.3.8] on MacBookPro M1 Pro, 16G mem

Hi,all the information you need about Photo AI can be found within the TopazLabs Docs page.

Topaz Photo AI will upscale your image width and height up to 6x, increasing the image pixels by 36x. The maximum image size it can output is 32,000x32,000 pixels (1024 megapixels).

Hope this helps

Is the limit can be extended in future?

Hi again Pin.pan. I’m afraid I don’t work for TopazLabs, so I don’t know if they intend to increase the limit size in the future or not having said that the sizes they do offer, are huge we’re talking high resolution billboard sizes I mean they are massive.

The files you’re working with are also massive and quiet why you would want such large files I can’t imagine.
Although I would be interested in finding out.

However, if you still want to process them with Photo AI there is a workaround you could use, this involves using an image editor such as Photoshop for instance.

The first step is to divide your image up into smaller segments (tiles if you like), makeing sure each segment has an overlap (this will help with realigning them later) then process each segment in Photo AI and finally rejoining the segment’s within your chosen image editor

Hope I could help you

thanks for your replay!
we works for museum and our tasks is to digitize some historical remains. which means more is better. In this case, we tried slice the data into 9 part, but we find the topaz treat each part differently, the result is not good enough to be used in exhibition. currently we process the picture in 30000x14900 with topaz and upscale it in PS.
we really hope topaz can extend the limit in future.

1 Like

Hi @pin.pan, I’m afraid we do not have any current plans to increase the max resolution of our apps, but I will pass this along to our team and let them know about this use case! Apologies this is not currently possible. Please let us know if you have any other questions.

Hi again Pin.Pan. Firstly I was fascinated and intrigued by the information within your reply, and secondly I’m pleased that Preston is passing information about your case over to the TopazLabs Boffins, hopefully they may have a solution for you.

Apologies for my late reply, I’ve been thinking long and hard about your predicament, therefore may I be so bold as to offer some suggestions of my own.

Firstly, before we start anything please make sure you’re working on Copy’s of your images and not the originals thank you.

Now I know I’m probably stepping on well trodden ground here but, do you have access to Gigapixel AI, I say this because with Gigapixel you have a lot more flexibility when it comes to manual control.

For instance you could load all 9 segments (parts), find out which upscaling settings works best on the first part, then eva write down the settings you have and use the same settings for all 9, at least this way you can be sure your adjustments are consistent, or simply apply to all.

Secondly, I suspect with Photo AI’s Autopilot and Subject Detection, would indeed treat each part differently.
Therefore, may I suggest making some changes to the configuration within the Preference Menu, which I believe will help to make the upscaling more consistent.

1st. Go to and select Preference (on the Mac it’s under the Topaz Photo AI drop-down menu)(on the PC it’s under the Edit drop-down menu), within the Preference menu you’re find two tabs General and Autopilot.

2nd. Click on the General tab, then under AI Processor, you’re have three options - Auto, CPU and GPU - the GPU will probably have the name of your GPU already there, please select GPU only.

Then please make sure Lens Correction is disabled

Next option is - Close Images After Saving -, this is particularly useful in the Standalone version of Photo AI therefore I leave this one on the default, so Photo AI doesn’t close the image (in other words, I often find I have to save multiple instances of the same image with different settings and blend them together in Photoshop to get the best results) and that’s it for the General tab.

3rd. Click on the Autopilot tab, the first menu is Subject Detection click the down arrow you’re have four options, normally I would leave this on default as Photo AI has an excellent subject detection algorithm. However, in your case it might have difficulty identifying the subject, so, I would select None (why you may ask, because Autopilot will always pick something and as you know Autopilot will only apply Sharpening to the subject, where as Noise Reduction will always be applied to the whole image) by selecting None sharpening is applied to the whole image

Note: obviously you do have the option for tuning off Subject Detection within the Autopilot panel under the Refine option and in the Sharpen panel by deselecting Subject Detection, then again, if you turn off Subject Detection within the Preference menu by selecting None, you won’t have to worry about turning it off for every image or forgetting to turn it off altogether.

The following option is Face Detection, again I would select None this may not have an effect in your case however, it’s better to be safe than sorry.

Next option is Sharpen, under this menu you have three options - Low (Soft), Medium ((Blurry) and High (Very Blurry) - please uncheck these three options so, Autopilot doesn’t attempt to automatically sharpen your image.

On to Auto-upscaling, you can if you so desire pick a predetermined Upscaling from two to six if you like or for more flexibility it’s better to select your scale from the Upscale panel within the application itself.

And that my friend is it for the Preference Menu.

Note: Here’s the thing TopazLabs recommend working with RAW images in both their Photo AI Standalone application or accessing Photo AI, from a program like Lightroom through the Plug-in Extras option. Personally I wouldn’t recommend this in your case, because when Photo AI return’s the images they might be underexposed, overexposed or have a strong colour shift/cast, but also if you’re happened to have left Lens Correction on and say Lightroom applies Lens Correction as well this will ultimately result in a distorted image, and make re-alignment a nightmare for you.

Which will involve a lot of unnecessary work on your part correcting the images, therefore I’d like to offer an alternative workflow for you working with Photo AI

Lightroom Classic: Post-processing (I’ve never tried or looked at Capture One so if that’s your weapon of choice I’m sorry I can’t help you with it).

Once you have the (9 segments) images selected in Lightroom, I would first of all apply Lightroom’s own Denoise from the Enhance panel (the easiest way to get to the Enhance panel is to select all 9 images then right click on the first image and select Enhance from the pop-up menu) apply denoise at 50% to the images (why mainly because Lightroom’s denoise is convenient quick and easy to use and besides Adobe’s Denoise algorithm’s are really quiet excellent) now I did say 50% this gives a good balance between reducing the noise and grain (digital noise).

Next continue processing the images within Lightroom as you would normally do and when you’re ready highlight the images you want to carry over to Photo AI
Right click on the first highlighted image and from the pop-up menu select > Edit In >Topaz Photo AI (that’s right pick Edit In and not File > Plug-in Extras > Topaz Photo AI, for the reasons I’ve mentioned above).

So another pop-up menu will appear please select Edit a Copy with Lightroom Adjustments, this will create 16bit TIFF files, which will retain all your Lightroom Adjustments plus the Lightroom’s denoise settings as well as all the tonal qualities from the images.

Once you’ve in Photo AI the first thing you’ve see is a pop-up about working with RAW images close this.

I think you’re be pleasantly surprised just how well Photo AI work’s with Lightroom’s Denoise, because we’d only applied a value of 50% within Lightroom’s Denoise, you’d be able to continue removing the rest of the noise without any loss of detail and because I suggested turning mostly everything off in the Presence Menu you will have practically full control over the DeNoise, Sharpen and Upscaling panels, which I believe will help.

You probably will have noted I haven’t mentioned Photoshop that’s because at this present time when you enter Photo AI from the Photoshop Filter menu, Photo AI’s Upscaling is disabled. You can if you want Upscale with Gigapixel AI within Photoshop from the Automate menu, quite why Topaz haven’t implemented this option with Photo AI is beyond me, maybe in the future I hope.

Pin.Pan, I do sincerely hope any of these suggestions help you in your endeavours to solve your upscaling problems.

And I do hope the incredible team of geniuses at Topaz gets back to you and are able in some way to help

As for myself, again I hope some how you find a solution, and may I ask if you’re not too busy in the future, maybe send a reply let me how you got on


This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.