Unsharp mask more than once?

When messing with USM also try the trick of using the largest radius choice along with wildly varying Amounts. Sometimes some surprises (not my taste though). It helps the effect if the program's USM allows up to a huge radius like 100 pixels or 250 pixels.
That used to be known as HiRa-LoAm (High Radius, Low Amount) sharpening. It acts just like the Clarity slider in Lightroom.
Clarity is a md-tone contrast booster. It’s neither sharpening or Unsharp Masking.

USM has been functionally obsolete for photos since about 2004. Yrs you can make it work but there are better tools in the box
Don't know how Clarity is done, but what gscotten described [I was used to calling it Local contrast Enhancement] certainly used USM at that kind of settings. It may even be in the book you mentioned.
 
When messing with USM also try the trick of using the largest radius choice along with wildly varying Amounts. Sometimes some surprises (not my taste though). It helps the effect if the program's USM allows up to a huge radius like 100 pixels or 250 pixels.
That used to be known as HiRa-LoAm (High Radius, Low Amount) sharpening. It acts just like the Clarity slider in Lightroom.
Clarity is a md-tone contrast booster. It’s neither sharpening or Unsharp Masking.

USM has been functionally obsolete for photos since about 2004. Yes you can make it work but there are better tools in the box
While there are better tools I believe they are more recent than 2004. AI sharpening in newer software produces better results than USM.
 
Even though the book is 16 years old and digital tools have evolved, the principle of a multi-step gradual > Targeted > Output method is still the gold standard.
But, are you actually doing this on a daily basis ?
On a daily basis? I always do capture sharpening in Lightroom or with ACR. And If I know how and at what size the photos are going to be reproduced, I do output sharpening using Smart Sharpen as a layer )so I can control the opacity in Photoshop.



if I think a photo needs it, Ido the second step. It’s easy enough to do as a masked smart sharpen layer in Photoshop
No thanks, two is enough for today's world.
Given you can only need all this for some incredibly demanding product, most just pre-sharpen in RAW and one pass the JPEG with USM or the like. Job done.
And now, you can play with AI, they didn't have that in 2009.
 
Neither do I. The fact is that I rarely use unsharp mask sharpening anymore. I like AI sharpening (properly controlled) better.
Well, that's news to me. I wasn't aware that there's such a thing as "AI sharpening".
It's been around for some years. Adobe, ON1, Skylum, and Topaz Labs all offer it. There are also web-based apps and mobile apps that use it. DxO has a proprietary sharpening method that's not AI powered, but is based on analysis of how specific cameras and lenses perform, applying different sharpening in areas where the sharpness of the lens itself differs.
Thanks. I'll have to look into that.

.... john
If you have Photoshop, use the Smart Sharpen Tool. It’s found in the Sharpen menu tab under Fikyer. Path: Filter>Sharpen>Smart Sharpen. You can change the Amount, Radius, and Reduce Noise values, but I leave them at their default settings. You can also choose between removing Gaussian Blur, Lens Blur, or Motion Blur. And you can also choose separate Fade Amount, Tonal Width, and Radius values for Shadows and highlights. Again, 99.5 times out of 100 I don't touch the Shadow and Hight values.

My preferred way of using it is to first do all of the other processing steps a photo requires (raw processing + whatever post-processing steps in Photoshop that I found necessary) and if that involved creating layers in Photoshop, creating a new layer at the top of the layer stack* and then applying the Smart Sharpen filter. This allows me to decide where I want and do not want the effect applied by creating a mask (when painting in a mask, white reveals the effect and black conceals it), and control how intensely the effect is applied by adjusting the opacity of the layer.
Here are four examples of how Smart Sharpen's default affect the appearence of an image.

1. Before applying Smart Sharpen

dead34b2e0af4f69be5fd2f6d1393518.jpg

2.) Smart Sharpen Default Setting (Lens Blur)

df25ce829336416997fd548f22dddec6.jpg

3. Smart Sharpen, Gaussian Blur (Default settings)

c1b330eb68a146beba355e2614354073.jpg

4. Smart Sharpen , Motion Blur (default settings)

631c6b045737463a95b009c8fa693dee.jpg

To be fair , there was no motion to be cancelled in that last example !

The base image I used is a full-resolution 770x770 pixel crop from a focus-stacked ( 13 frames) 45MP raw image shot with a Nikon Z7 II camera and a TTArtisan 75mm f/2 lens. Exposure settings: ISO 64, 1/60th second, f/5.6. The raw NEF file ws processed in Lightroom Classic and exported as an 8-bit TIFF in the sRGB color space. These examples were then saved as level 12 JPEGs after processing in Photoshop Beta (release 27.0.0)


Now for the big question, do you or I need to do any of this? For most photos, no.
.

*applying anything other than mild capture sharpening to a raw file sharpening and then doing more post processing on top of that raises the risk of creating nasty artifacts if you are working with 8-bit files.
--
Ellis Vener
To see my work, please visit http://www.ellisvener.com
I am on Instagram @EllisVenerStudio
 
Last edited:

Keyboard shortcuts

Back
Top