• Please use real names.

    Greetings to all who have registered to OPF and those guests taking a look around. Please use real names. Registrations with fictitious names will not be processed. REAL NAMES ONLY will be processed

    Firstname Lastname

    Register

    We are a courteous and supportive community. No need to hide behind an alia. If you have a genuine need for privacy/secrecy then let me know!
  • Welcome to the new site. Here's a thread about the update where you can post your feedback, ask questions or spot those nasty bugs!

Photoshop - blending modes - great reference

Doug Kerr

Well-known member
New member Jonas Wendorf dropped in while I was mucking about in the area of Photoshop blending modes and, among other useful contributions, gave a link to a marvelous reference resource on the topic:

http://www.simpelfilter.de/en/basics/mixmods.html

The site, by Ralph Altman, gives concise but illuminating and rigorous descriptions of the blending modes, with the corresponding equations. (I was relieved to find that his equations for Multiply and Screen agreed with the results of my own reverse engineering!)

The site contains the modes available in PS as of CS2, but there is a companion site (in German) that supposedly brings it up to date through CS5.

Thanks, Jonas, for this reference, and for joining in the fun here.

Best regards,

Doug
 

Asher Kelman

OPF Owner/Editor-in-Chief
Wonderful link. So logical. But why on earth is there not a look up function in Photoshop to let us in on these facts and explanations!

Asher
 

Doug Kerr

Well-known member
Hi, Asher,

Wonderful link. So logical. But why on earth is there not a look up function in Photoshop to let us in on these facts and explanations!

Indeed! And the Photoshop help in this area is written as if in the private language of some secret society.

Best regards,

Doug
 

Ben Rubinstein

pro member
I'm glad PS doesn't use that link to explain the modes. Although accurate, to someone wanting to actually use these methods it's somewhat lacking in readability and practicality for a user. Sandrine's link is far nearer to a 'help' guide than this however valid it may be.
 
I agree with Ben, first you want to use things, and of course, later you want them explained (and you don't count on the PS help for that). I have my own theory:

Look at all those tutorials, schools, degrees, books, videos and so on that companies market and sometimes at huge prices to "master" Photoshop. Some of them are also free, and since, like me, not everybody is a scientist, some need a straight forward approach, and some are OK with formulas. I can't see why Adobe would invest more in the help file since we got the Adobe.com community and experts (Deke, Andrew - Hello, you- Russell Brown and the likes) because it would never be comprehensive.

some examples:

http://www.photoshop.com/tutorials
http://tv.adobe.com/watch/the-russell-brown-show/advanced-masking-with-photoshop-cs5/


The entire www. is an help file :) .
The problem remaining: where to find them?
 

Jonas Wendorf

New member
Hi Doug,

I'm glad the link is of help to some members :).
From my own experience I know that some people will learn better with visual examples (as in Sandrine's link), some will learn better if they truly understand what's going on under the hood of Photoshop.

Unfortunately Adobe seems to be quite close about the real formulas (the ones that are shown on Altman's site are empirical measures, not necessarily what's really going on).
Maybe they don't want any competition, I don't know (though sometimes they give a sneak peak at some of their modes, e.g. here: https://www.adobe.com/devnet/flex/articles/pixel_bender_basics_flex_air.html section "Using blends").
 

Doug Kerr

Well-known member
Joachim,

Well, Adobe has this covered in her online-help section. More and more programs do so, to keep the help-information up-to-date. It's quite easy found on adobe.com

http://help.adobe.com/en_US/photoshop/cs/using/WSfd1234e1c4b69f30ea53e41001031ab64-77eba.html

Yes, this is essentially identical to the local "help" on this topic.

Many of those descriptions of the blending modes are not particularly helpful to me (although they are better than I at one time thought!), and some are just not correct.

For example, the entry for the Screen mode starts:

Screen Looks at each channel’s color information and multiplies the inverse of the blend and base colors. [The rest of the item is devoted to discussing the visual result.]

In fact, this function multiples the inverse of the blend and base colors and then delivers the inverse of that.

Best regards,

Doug
 

Joachim Bolte

New member
@Doug,

I see what you mean:

In the formula the site gives:

C = 1 - (1-A)*(1-0,5) = 1 - (1-0)*(1-0,5) = 1-(1*0,5) = 1-0,5 = 0,5 , so B is unchanged in case the tonal value of A is 0 (or black)

C = 1 - (1-A)*(1-B) = 1 - (1-1)*(1-0,5) = 1-(0*0,5) = 1-0 = 1, so B will be turned up to full tonal value in case the tonal value of A is 1 (or white)


In a formula I coocked up according to the Adobe site and your info (remembering that it should be the exact opposite of the Multiplier mode)

INV:C = INV:A * INV:B = INV:0 * INV:0,5 = 1*0,5 = 0,5, so C = 0,5 (same result, black gives back the original color)

INV:C = INV:A * INV:B = INV:1 * INV:0,5 = 0*0,5 = 0, so C = 1 (same result, white gives back white)


I like fiddling around with formulas, but could you explain to me what the added benefit for this knowledge is when I know what the blending mode does? I really never felt the need to go and calculate what the effect would be, I just applied the blending mode and played with the settings untill satisfied.
I think most people work in this 'visual' way, I have never seen any tutorial about the mathematical part of things.
 

Doug Kerr

Well-known member
Hi, Joachim,

@Doug,

I see what you mean:

In the formula the site gives:

C = 1 - (1-A)*(1-0,5) = 1 - (1-0)*(1-0,5) = 1-(1*0,5) = 1-0,5 = 0,5 , so B is unchanged in case the tonal value of A is 0 (or black)

This is evidently for B=0,5.​

To be rigorous, B is not changed. I think you mean that in this case C=B.

C = 1 - (1-A)*(1-B) = 1 - (1-1)*(1-0,5) = 1-(0*0,5) = 1-0 = 1, so B will be turned up to full tonal value in case the tonal value of A is 1 (or white)

To be rigorous, B is not changed. I think you mean that in this case C=1.

And yes, that seems to be how it works.

If A=0, C=B

If B=0, C=A

If A=1, C=1

If B=1, C=1

This follows from:

C = A + B -AB

which is of course equivalent to:

C=1- (1-A)*(1-B)

the form that best presents the Screen function as what I call the "dual" of the Multiply function (I don't know what the "opposite" of Multiply would be - Screen is like Multiply with an inverted scale for A, B, and C).

Note that the text description for Screen in the Photoshop Help implies this:

C=(1-A)*(1-B)

Thanks for your insights.

Best regards,

Doug
 

Doug Kerr

Well-known member
Hi, Joachim,

I like fiddling around with formulas, but could you explain to me what the added benefit for this knowledge is . . .

Well, I would hesitate to say what benefit my writing is to any given person.

My basic motivation in this case was this. The descriptions I have seen of the Screen blend mode are not very specific, and some of them just don't work.

I know that if I take two separate images and combine them in the Screen blend mode, in the composite image I will see all the stuff in each, and the whole thing will seem pretty light.

If I take two copies of the same image and combine them in the Screen blend mode, in the composite image the midtones have been substantially lightened, almost as if I had applied a curve like this:

Curve_screen_01.gif


But I still have no idea what principle the Screen function is following that causes this.

So I did some reverse engineering to find out. My answer came out in the form of an equation. And I reported it that way in my concise note here on my findings.

Now a little later, I got an even further understanding of the significance of that equation. And, with that in hand, I could now say:

In the Screen blend mode, for each color channel, for each pixel:

• the color coordinates (R, G, or B) of the two source images are inverted (in the sense of taking, for example, 255-R, not 1/R)
• the two inverted values are multiplied (doing what we need to to deal with the fact that the values are on a scale of 0-255)
• the result is inverted (in that same sense)
• the result of that is the value of the coordinate for the pixel in the composite image.

. . . when I know what the blending mode does?
I'm sure you do, as a result of extensive use of the function. But could you articulate that understanding - pass it on to a colleague?.

Suppose I came to you and said, "Joachim, I am new to Photoshop, and I have encountered this thing called the Screen blend mode. Could you tell me how it works?"

And in fact, I would appreciate it if you would do for me exactly that.

I really never felt the need to go and calculate what the effect would be, I just applied the blending mode and played with the settings until satisfied.
And that's fine. I think few people would want to calculate the effect of applying this function. But the general nature of what is going on can be valuable. And I don't know any way to report that except "technically".

I think most people work in this 'visual' way, I have never seen any tutorial about the mathematical part of things.

Sure. But I think some people want to know in some broad, but true, way what the knobs do.

If you don't, that's no problem to me!

By the way, and I'm sure you know this, the result of combining two copies of the same image under the Screen blend mode is in fact exactly like applying the curve I showed above.

Best regards,

Doug
 
By the way, and I'm sure you know this, the result of combining two copies of the same image under the Screen blend mode is in fact exactly like applying the curve I showed above.

Hi Doug,

Which would make it much easier to remember and predict the result, and correct it in the same step if we don't like it. Like you, I don't particularly like black boxes where it's hard to predict the result that comes out, and potentially with unknown side effects. A fundamental understanding also allows to seek an alternative approach, which might have other benefits as a bonus (e.g. being indifferent to the workspace gamma settings).

Cheers,
Bart
 
Top