Question re: Old Photoshop Workflow (Blending Mode)

[joneh]

Hello All,

I hope I can bother you folks with a newbie question.

I’m coming from one of the older non-subscription versions of Photoshop (not sure of the version; that computer is no longer accessible). I’m a photographer – definitely not a power user – and I had a fairly standard workflow for adjusting image contrast in Photoshop. I’m curious whether I can replicate it in Affinity 2.

Assume a jpeg image. My Photoshop steps were as follows:

1. Command-click on the RGB channel (“marching ants” appear)
2. From the Layer menu, select New Layer via Copy
3. On the Layers tab, select either Soft Light (usually) or Hard Light (occasionally) blending mode

This almost always gave me a reasonably subtly contrast enhancement – good enough for my purposes. Hoping I can get the same results in Affinity.

I appreciate your help!

PS I tried to search the forum for this question, but I didn’t come up with anything; if it’s been asked and answered elsewhere, I’d be grateful if someone could point me to the post.

[firstdefence]

Affinity doesn't have a RGB composite layer so you would need to select each RGB composite channel.

From the Channel Panel

1. Right Click on the Composite red and select load to pixel selection
2. Then Right Click on Composite Green and select add to pixel selection
3. Then Right Click on Composite Blue and select add to pixel selection
   Now you have your RGB selection...
4. Now do a Copy and Paste and change the ensuing layer to soft or hard light.

[firstdefence]

Or you could play around with a Channel Mixer Adjustment Layer set to soft or hard blend mode and drop the opacity to about 25%.

[walt.farrell]

1 hour ago, firstdefence said:

Affinity doesn't have a RGB composite layer so you would need to select each RGB composite channel.

From the Channel Panel

1. Right Click on the Composite red and select load to pixel selection
2. Then Right Click on Composite Green and select add to pixel selection
3. Then Right Click on Composite Blue and select add to pixel selection
   Now you have your RGB selection...
4. Now do a Copy and Paste and change the ensuing layer to soft or hard light.

After step 3, don't you simply have all the pixels selected? And if so, why not just do a Ctrl+J and duplicate the original pixel layer, and set the Blend Mode of that new layer?

[smadell]

In Photoshop, command-clicking the RGB channel creates a luminosity selection. You can do the same thing in Affinity Photo by (a) command-option-clicking on the thumbnail in the Layers panel, or (b) choosing Selection from Layer Intensity from the Layer menu. Once you have the selection, choose Duplicate from the Layer menu (or press command-J). Set the new layer’s blend mode as appropriate. Same result as you saw in Photoshop.

[firstdefence]

8 hours ago, walt.farrell said:

After step 3, don't you simply have all the pixels selected? And if so, why not just do a Ctrl+J and duplicate the original pixel layer, and set the Blend Mode of that new layer?

Actually no, I see a similar effect to the command click in photoshop, albeit slightly more intense. I'd be interested to know why my method is more intense, when I am effectively doing the same thing but in individual stages. Maybe there is a layering effect on selection making each pixel more opaque?

[smadell]

@firstdefence - First, I'll admit that I'm leaping without a net here. But, I think the difference you're seeing is because Command-clicking the layer thumbnail (what I'll call the "smadell method") creates a selection based on Luminosity, and sequentially loading and adding the individual channels (what I'll call the "firstdefence method") is based on adding RGB channel values. And, I have to assume that Luminosity and (R+G+B) do not give the same values.

To test this out, I made a single pixel layer. It has a white background and a dark blue rectangle. The white background has an RGB value of (255,255,255) and the dark blue rectangle has a value of (40,60,80). I then created two additional layers, one by command-clicking the layer thumbnail and then duplicating the selection (the smadell method), and the second by choosing Load to Pixel Selection in the Composite Red channel, then Add to Pixel Selection in the Composite Green and Composite Blue channels, and then duplicating that selection (the firstdefence method). The result is shown below.

Clearly the results are different, and I think that the difference lies in the method of selection (luminosity vs RGB values). Also, I would have expected the "firstdefence method" to give a rectangle with the color (180,180.,180) since 40+60+80=180, but clearly that didn't happen either! At this point, only the Serif developers (or someone with way more color theory info than me) can take the explanation further.

[joneh]

Hi All,

Thanks so much for all the replies, and all the helpful suggestions.

I've been experimenting with the two main options, and to my eye at least, in practical application, the results are indistinguishable – and perfectly satisfactory. I think the results may be better than my old tried-and-true Photoshop method, for whatever that may be worth.

This will greatly ease my transition to Affinity Photo, and I look forward to learning more about its (very impressive) capabilities.

I so appreciate your help! Thanks again.

[lepr]

The firstdefence selection method gives a selection intensity of the maximum of R, G and B for each pixel. I wouldn't use that method because it can result in a perceptually dark colour being considered as having greater intensity than a perceptually bright colour.

The smadell selection method gives a selection intensity of the perceived brightness (maybe luminosity or maybe luminance or maybe luma or some such measure of "brightness") for each pixel.

[firstdefence]

Thanks for the explanations people, got a clearer understanding of whats happening.

[firstdefence]

Interesting Video on luminosity and Affinity Photo, although it's showing it's age a bit now being 4 years old.

[James Ritson]

On 11/18/2023 at 10:36 PM, firstdefence said:

Or you could play around with a Channel Mixer Adjustment Layer set to soft or hard blend mode and drop the opacity to about 25%.

To expand on what @firstdefence has said here:

A good non-destructive option would be to use a Channel Mixer adjustment layer and set its colour model to Gray, then use any blending methods you wish (e.g. Soft Light with a low opacity value). This produces the same weighted intensity/luminosity result as CMD+Option-clicking on a layer, except it will update dynamically if you modify layer content underneath it.

If you just want to apply a blend mode to a duplicate of the image content, you can use an adjustment layer such as Channel Mixer rather than duplicating the image layer. Any adjustment that doesn't immediately modify the pixel values will suffice, e.g. Curves, White Balance, Levels (adding them but leaving the parameters alone will produce an identity result). This is useful for preventing redundant copies of your image content.

@joneh as you experiment further with Photo, it's worth noting that it has some non-destructive functionality that isn't immediately obvious—so whilst your techniques from PS will generally port over, please don't hesitate to ask on the forums as there may be a more effective approach in Photo. Live filter layers, for example, let you apply various filters as layers rather than having to merge your work into a single layer and apply a smart filter to it.

One example would be recreating something like the Orton Effect: instead of duplicating your layer and applying a blur to it, you can instead go to Layer>New Live Filter Layer>Blur>Gaussian Blur and choose a suitable radius value. You can then set the blend mode of this live gaussian blur layer to Overlay (or whichever blend mode you prefer), then fine tune its blending using Blend Ranges (similar to Blend If): https://affinity.help/photo2/English.lproj/pages/Layers/layerBlendRanges.html

This will allow you to blend the effect away from the darker tones but keep it in the brighter tones. You can position this layer above your other layer work and modify it at any point. For photography, there are also live implementations of filters like Clarity and High Pass, which are useful for for texture/structure enhancement and detail enhancement respectively.

Hope the above is helpful!