Applied Color Theory in September

by Dan Margulis on April 30, 2014

Update 25 July 2014: The class described below is sold out. The sponsor will waitlist on request.

In September 1994, I took three days off from my day job in New York, and spent them teaching color correction to six people in Atlanta. We were using Photoshop 3, with no adjustment layers, no multiple undo, no actions, and computers with 16 mb RAM.

Sterling Ledet gave the name “Applied Color Theory” to the class. Certain features of it may seem familiar to various list members. The concept was that we learn by seeing how others approach the same originals that we do, and whether we prefer their results to our own. When the entire group decided unanimously that a certain version was bad or good, we took that as an indication that it was, in fact, bad or good. When there was disagreement, we discussed why we disagreed and how a client might take either position.

The reactions to that first class were so positive that we scheduled another, and then another one after that, and so on. By the time I decided to wind down my career a few years back, Applied Color Theory had been offered more than two hundred times, in close to a dozen different countries, in four different languages.

And never the same way twice. Every group has its own character, faces its own challenge, exults in its own successes. I never got tired of it, because I never found it repetitious.

As time went on, the curriculum changed. Exercises were replaced when more instructive ones were found. But a few have stood the test of time. Of the dozen images that today’s classes work on in the first day and a half, three date from the last century, due to their consistently high ratings on evaluations that ask the students which images they found most useful.

In 2008 the curriculum was overhauled to emphasize the Picture Postcard Workflow, which of course was a work in progress at that time. Classes got to play with early versions of the PPW panel, and in 2011 in Italy, we expanded the format to four days.

In September, we’re going to run a 20th anniversary class, the only one I’ll offer this year. It will be at Ledet Training in San Diego, Wednesday 10 September through Saturday 13 September. Our industry doesn’t work 9 to 5 and neither does this class, so this schedule implies 40 hours of class time, limited to eight people so that everything is hands-on.

It’s a big investment in many respects, but it’s also been a life-changing experience for many over the years. If you’d like to consider signing up, click here.

Who’d have thunk it, twenty years ago?

{ 13 comments… read them below or add one }


Hello Dan!
I just purchased your video tutorial on Kelby’s training (Introduction to Photoshop LAB Color), and i am disappointed by the quality of the video (even in Hd 720).
It is quite blurry, so much that it is hard sometimes to see the changes that you make on the pictures. Also, the recording of the screen is so laggy that it prevents from seeing the effects of some of your layer’s visibility On/Off (like on the part with the american flag). I asked a refund to kelby and they will hopefully grant it.

Is there another website hosting a more recent and more high quality version of your course? I would rather invest in such a thing, or even buy your book since you seem to have updated it last year.
Your teaching is a gold mine, so i thank you so much for your work and dedication in this field!

Dan Margulis


People sometimes complain about speed and streaming ability from the Kelby site but it is rare to hear about fuzziness.

The Kelby videos were state-of-the-art when they were made, at least as far as technique was concerned, but some of them no longer express my thinking. Part I of the PPW set is AFAIK still 100% accurate. Part III is largely obsolete in view of subsequent developments. Part II is a little of both.

For buyers of the new book, friends and I recorded many new videos and samples can be seen around the site. They were recorded in late 2012 so they are still very current. The resolution is the same as at Kelby but the production values are a little better.

Thanks for the encouraging words about the gold mine, although I must point out that gold has been rapidly decreasing in value in the last year!


Haha!! Let’s say then that your teaching is a good bottle of wine, it never go down in value, quite the contrary (especially for the thirsty one). By the way, Kelby refunded me instantly. So i will get your book as soon as my bank account breathes again.

Your new videos look very good. Your capturing software is for sure very much better!

I am at the time quite amazed by the discovery of the Lab mode.
I am a digital painter, and look for ways to use its power for this purpose. The first very interesting point was the possiblity to choose a color without changing its luminance. If i want to paint a shadow, i can add variation by just changing this parameter and it won’t mess up with my base values.
Though there seem to be a limitation in the sense that a saturated color influences our perception by making them look lighter (because of the Helmholtz–Kohlrausch effect i guess).

This property of the lab is well used by a plugin called Coolorus, which is a colorwheel for photoshop that allows a luminosity lock. By the way, the friendship with the munsell color system used by painters is quite obvious suddenly.

It also feels like the color blend mode of lab works a bit differently than the RGB one. But it looks like it still affects the values. A way i found to counter this was actually to mask this color layer with the image of my values.

Though, I am far from understanding all this stuff and what is going on. I have only been in the lab world for a week :p

Maybe you have some thoughts of secret recipees that would be interesting for the digital painters workflow?
In any case, have a fantastic day Dan!

Dan Margulis


If there were a recipe for “art” we could let a machine do it. It’s harder than standard color correction because of what you are producing is considered “art” and not “photography” then you aren’t limited by what the viewer might take to be realistic.

The PPW actions are valuable for this because they deliberately exaggerate effects. The idea is it’s easier to cut back on excessive color than it is to force color into a bland file. By the same token, however, it makes it easier to see where certain obviously exaggerated effects may nevertheless be artistically desirable.

I’d especially point out the options windows for the MMM and also the MMM + CB actions, which allow you to try and compare several different approaches. Also, if you are interested in counteracting the Helmholtz-Kohlrauch Effect, the H-K action’s options will permit that. (The action itself does the opposite, it caters to the effect by darkening neutrals, but you can configure it to do the opposite.)

For tricky moves in LAB, my book Photoshop LAB Color may be ten years old, but the “artistic” portions of it are not out of date.


Oh thank you Dan for these precious ideas.
I am not really looking for some recipes, more trying to find ways to simplify or go over some technicalities in order to focus on the artistic side.
I will definitely take a shot at all your books and i am sure they will stay on top of the pile of color correction for quiiite a while :)


After enhancing colour in LAB mode I usually find channels in RGB has been detroyed, being totally black. Am I wrong to use the tool?

Dan Margulis


*In Camera Raw you move the Exposure slider to -2.00 and all shadow detail is destroyed. Are you wrong to use Camera Raw?

*You drive your car at 180 kph, and the police stop you and take you to jail. Are you wrong to drive the car?

If your RGB channels are plugging, it indicates that you have used LAB to try to force colors that are out of the RGB gamut. The book strongly cautions against doing that and to monitor the Info palette to make sure it does not happen.


When I shoot something with bright colour, the photo I take has already gone over the RGB gamut. What should I do? Should I force it to nomal in Camera Raw by adjusting “black” and “white” or set a special white balance to achieve complete RGB channels ? And will an image of 16 bit help to solve this and control noise?

Dan Margulis


A camera cannot capture something that is out of the camera’s own gamut. So there is an incorrect usage of Camera Raw. In your first message you indicate that you did something in LAB that destroyed the image by forcing colors out of output gamut. Now in this second message you say ythe same thing is occurring in Camera Raw. I cannot guess at what has happened because there are thousands of ways to drive colors out of gamut in LAB and thousands of other ways to do it in Camera Raw. All I can say is to repeat the advice of the book: it is the user’s responsibility to take care that it does not happen.


Thank you very much for your patience. In camera Raw, the blue channel of a yellow follow sometimes become totally black. I try to get it back but in vain. I do not know it is normal or not, for I lack the experience in dealing with raw image. Your book is very helpful to me, it is time for me to read it again and again.

Dan Margulis

Optimo, I would not worry too much about a bad blue channel. We cannot see detail in yellows that bright. If you said that the flowers were magenta rather than yellow, and that the green channel was very bad, this would be much more serious and you would need to find a solution. But a bright yellow flower is different, we can appreciate its color but not so much its detail.


Thanks, your word remind me of what you told me early in the book, blue channel only compose one percent for the whole contrast. Most of my troubbles occur in bule channel. And there is another tiny question, why different RGBs have different gamut? They are all have 256 in each channel.

Dan Margulis

Optimo, different RGBs have different gamuts because our language is not precise enough. If you say “255r0g0b” I know that it is “red”, but I do not know what “red” means specifically. If you say 54L81a70b, that is unambiguous, we know exactly what red is being referred to, which happens to be the same as 255r0g0b in sRGB. But in Adobe RGB the equivalent is 63L80a78b, and every other RGB has a different definition as well.

Leave a Comment

Previous post:

Next post: