I’m using Lightroom CC Classic and Topaz Denoise AI, Clear AI, Adjust AI and Sharpen AI.

What is the best workflow?
Rgds Per

Disregarding JPEG to RAW and GigaPixel AI this is what is recommended:

In other words - no editing in Lightroom?

Nope, that is a workflow that you would use calling from any raw editor or, if you don’t have one, you can call using the plugins in Studio 2 … or simply standalone.

I understand and am ok with the upgrade changes; however, it would be VERY helpful to provide information on 1) when to use which product for those which have some degree of overlapping functionality, and 2) what the recommended workflow is for those of us who own ALL of your AI (and predecessor) products! Thx.


I’d like an answer to point 1, too.

I’d like an answer to both questions as well. Have been searching for exactly these points. Thank you.

I just purchased DeNoise and want to figure out the best workflow prior to enlarging small files (pixel dimensions) in Gigapixel and then sharpen in SharpenAI.
I shoot Raw files pretty much all the time and process them in ACR (adobe camera raw). Am I correct in understanding that it’s best Not to sharpen or Reduce noise at all in ACR? Only do exposure, col, contrast etc. And then open the raw file in Denoise? (Does it open All raw files - Sony, fuji, Panasonic etc?)
And then in Denoise to use the DenoiseAI model not the AIClear. Start with auto settings and tweak from there, but turn off Sharpening and low light mode and do that later with SharpenAI or Photoshop?
And a side note question please -
100% preview is Much larger than 100% view in Photoshop - this is confusing for evaluating noise or sharpness etc
Thanks for any help

No, opening the RAW file in this case will do nothing except create a output image that has DeNoise applied.

In your case it would be best to open the image in ACR without applying Sharpening or Noise reduction as DeNoise has a pre-sharpener. In DeNoise use the suggested noise reduction and sharpening and, if necessary, use the low light mode for images that are either night or high ISO shots, or under/over exposed in camera.

The AI Clear model can be used but there is less control over the application to the image.

Sharpen AI can be used as a final step before output.

I can’t see that because here is the CS6 view:

And here is the Sharpen AI view:

And here is the DeNoise AI view:

Seem to be exactly the same to me, note the navigator as there are different viewing spaces between PS and the items.

Thank you for your info. I tried to reply via the email notification for your post that I received. But since it’s my first foray into the Topaz forum I’m not exactly sure how it works so if you get two replies I’m sorry.

If I understand you correctly, you’re advising me to stay on the workflow I currently use which is to process my Raw files in ACR and save them as Tifs (16bit) but to exclude any sharpening or noise reduction in ACR
Then open the Tif in Photoshop and use Denoise. Save the file and use SharpenAI (or if its to be enlarged then first use Gigapixel and then Sharpen AI.)

You said -“In DeNoise use the suggested noise reduction and sharpening and, if necessary, use the low light mode for images that are either night or high ISO shots, or under/over exposed”

By suggested I assume you mean the Auto Detect values. Or, I assume I can choose to use the manual sliders (Noise reduction and preserve detail) accordingly but without the sharpening slider or not, depending if I am planning to use SharpenAI afterwards or not. Low light mode can additionally be on or off as well as the manual sliders.

I still can’t figure how to get my Ps and Topaz previews to match at 100% as you have illustrated you do. Perhaps this is a separate post/question.

DeNoise has a pre-sharpner which is not the same as Sharpen AI.

If you are using ACR just use DeNoise from PS, no need to save then go to DeNoise, use it as part of the process.

Note that not every image needs noise reduction as it depends on the sensor, camera settings and lighting conditions.

"If you are using ACR just use DeNoise from PS, no need to save then go to DeNoise, "

This isn’t clear to me.
When using ACR I always have to Save and then click Done in order to have a Tif to open in Ps

I’ll practice on a few images and maybe it’ll become apparent how to establish a workflow

In Photoshop you can duplicate your layer (name it DeNoise or something like that) and then go to Filter—>Topaz Labs—>Topaz DeNoise AI. Do your adjustments in DeNoise and click Apply. Then duplicate the DeNoise layer and name it Sharpen AI. Go to Filter—>Topaz Labs—>Topaz Sharpen AI. Make your adjustments there and click Apply.

Yes thanks. That’s exactly what I’ve been doing tonight on a few images.

And if I need to enlarge the image in Gigapixel would you say I do that in-between those two steps

  1. Denoise 2. gigapixel 3. sharpenAI

My understanding is Gigapixel is always the final step. So you would save your work in Photoshop as a TIFF and then open that TIFF in Gigapixel. Someone might want to confirm that.

Hello, I’m a new Topaz Gigapixel AI user and have similar questions to workflow as discussed in this thread. My general workflow is to do various image corrections/adjustments in LR or ACR/PS, resize the image for the specific print output size (e.g. 24"x36" @ 300 dpi ), then do final output sharpening on the resized image. If I’m going to use Gigapixel AI to do the upscaling/resizing, would you recommend that I do final output sharpening prior or after Gigapixel AI? The general Topaz workflow suggests using Sharpen AI early in the workflow (pre adjustments) but I’m not sure that would make sense for output sharpening. Also, I’ll probably continue to do the output sharpening in PS (e.g. Smart Sharpen) for now but may try Sharpen AI down the road. BTW, I’m starting with high-res raw files in my workflow. Any suggestions from experienced users appreciated!

Sharpen AI is general sharpening and not output sharpening. Output sharpening is specific to the media being used for the final output such as monitor, gloss paper etc., etc.

GigaPixel AI should be there final process before output sharpening.