Color Management, What is it really??

Skinflint

Well-known member
Color management seems to have many components all playing a roll at different stages.
Can anyone explain the process from beginning to end?
From within design applications, monitor, proofing, press etc...
How does it really work? Are there any good books that explain the topic?
 
Lots of books are recommended in various posts on this forum. Any reply to your message is necessarily going to be too brief or abstract to give you any practical details of what you're asking. Agree with Stephen, above, you should visit Amazon - the book by Abhay Sharma is also good (it seems most people get both books - Real World & Sharma - eventually).

See if you find anything of use on the icc website... INTERNATIONAL COLOR CONSORTIUM ... They basically orchestrate the whole technology.
 
There are several parts:
Understanding colour
Understanding how we see colour
Understanding the limits of each medium and colour model
Managing the variables so that each medium is as consistent as possible (to the standard for that medium so that it is consistent with itself, and other media of the same type)
Making the decisions how to convert colours, and the philosophy behind the various rendering intents

Colour management is the last of those, the decisions.
 
Color Management: what is it, really?

Color Management: what is it, really?

I'd just like to second Stephen Marsh's post on November 27. I have read Real World Color Management, and I have found it to be an excellent tool that attempts to explain the process in 'real world' English! This book has helped me a great deal and I wholehartedly recommend it.

Regards,

John Stanzione
Manager - Technical Support
Pantone LLC
 
Re: Color Management, What is it really??

I would sum up Color Management in the following way: Color Management is a consciousness that, behind every digital value, there is a mathematical visual value associated to it. Every CMYK, every RGB, every CMYKOG has some kind of correlated visual value. You simply need to be conscious of that and... when you don't know the answer... seek it. There is always an answer. If you can correctly determine it, you can successfully manage it. Once you achieve this consciousness, issue like "why does this RGB value look different on my monitor and on my printer?" start to become a lot clearer, as well as both the solutions and limitations of communicating color.

When I started at Monaco Systems in 2002, I was told "In every case, if you use the software [color management] correctly, it will not let you down." In my experiences to date, for the most part, this has been true.

My 2 cents.

Marc
[email protected]
 
It's a scientific attempt to control something that is completely subjective. Agree or not...if a customer says it's too red...it's too red...even if you know better because that's what they see...and no level of expertise and no measurement device will change that.

It's a pretty good way to control a process with so many variables.
 
It's a scientific attempt to control something that is completely subjective. Agree or not...if a customer says it's too red...it's too red...even if you know better because that's what they see...and no level of expertise and no measurement device will change that.
That's what contract proofs are for. So that the customer will sign off this is the colour I want…*then i can match it. This is just because colour is subjective, and even more we know that (physical as opposed to hypothetical) people have poor colour memory.
 
That's what contract proofs are for. So that the customer will sign off this is the colour I want…*then i can match it. This is just because colour is subjective, and even more we know that (physical as opposed to hypothetical) people have poor colour memory.

Let's be realistic. The "contract" in "contract proof" is all but meaningless.

I regularly have signed proofs that become garbage once the customer arrives on press with an additional target that they found later and like better. In this cut throat economic climate, it's not really prudent to argue the point of color. The customer is always right. Sorry, but that's the fact. I remember the days we we could practically pick and chose our customers...now we're lucky to have who we have.

BTW...there's no such thing as color memory...there are memory colors, but visualizing color is relative to the moment....it's dependent upon everything from the environment, to what you're wearing that day, to what you ate that morning. A trained eye certainly helps, but no one can consistently recall a particular color.
 
Last edited:
There is also a bit of psychology involved... We had this girl, marketing specialist, coming to press approvals. In flexo, color management is all but press-feasable. Signed contract proofs would do no good, the girl always pulled a rabbit from her purse and press approvals would last forever. We realized that this girl just would'nt want to travel back to her office, so she'd fuss around till about 4 0'clock. We decided to schedule the approvals at 3h30. All of a sudden all the colors were matched perfectly !! (true story)
 
There is also a bit of psychology involved... We had this girl, marketing specialist, coming to press approvals. In flexo, color management is all but press-feasable. Signed contract proofs would do no good, the girl always pulled a rabbit from her purse and press approvals would last forever. We realized that this girl just would'nt want to travel back to her office, so she'd fuss around till about 4 0'clock. We decided to schedule the approvals at 3h30. All of a sudden all the colors were matched perfectly !! (true story)

Not uncommon, LOL. That's why I always say it's all smoke and mirrors. Science is for the lab...very little applies to the bottom line.
 
Color Management--What it is Not

Color Management--What it is Not

All these anecdotes are amusing and doubtless true, but irrelevant to the question that was asked in earnest. Besides, saying that CM doesn't matter because customers are often irrational or demanding is like saying speed laws don't matter because so many people ignore them. Right--try repealing them in YOUR town and see what happens. Unfortunately, this attitude often masks an aversion to learning color management because it appears difficult to grasp. Sometimes it is. Marc's answer was a start. I'll try to add to it.

Color management should properly be called digital color management because it was the digitization of color that made it necessary. Before we started scanning films and prints and snapping digital pictures we had only "real," physical color in the form of photographic prints, films, and so on. These got duplicated and transferred as physical objects that looked pretty the same no matter where they went or who looked at them--all one needed to do was fix the lighting conditions. Whatever one did to adjust color in cameras, on press, and so on, it was evaluated directly by looking at the printed products themselves, generally side-by-side with some "original."

When these devices started outputting numbers rather than film or paper the problem immediately arose that because they behaved differently the mixture of red, green, and blue (or cyan, magenta, yellow and black) needed to create a given color was different for every device. The "red" of one monitor did not really have the same appearance as the red of another monitor, and a scanner would produce a different red still, and so on. And of course there was the problem of equating CMYK with RGB. Color management is a system of reconciling these differences and getting a consistent color appearance from all sorts of digital color devices that behave differently. The core of the system is something called a Profile Connection Space, or PCS. This is a system of color definition that expresses how a color actually appears. The most common PCS is L*a*b*, and it defines how reddish, greenish, yellowish, or bluish, as well as how light or dark a color is. This is quite different from RGB or CMYK definitions, which tell us only how much of each monitor/scanner colorant (forgive me, Marc, but I'll skip abstract working space "colorants" here) or printer ink is in a color but tells us little about what those particular colorants look like. As we said, one monitor's red isn't like another's, one printer's cyan is not the same as another's.

Color management works like this: The output of every digital color device gets measured--not just its RGB or CMYK primaries, but hundreds or thousands of color combinations, and a kind of translation table is built that gives equivalency between the "device color," which is RGB or CMYK and the PCS, usually in L*a*b* units. This table says, "When this device is given so many units of R, G, and B, or C, M, Y, and K, it produces a color that looks like THIS." Or, when I feed the device a color that LOOKS like THIS it produces THESE RGB or CMYK numbers. This little color "dictionary" is an ICC profile. This is the basis of the entire system. From this profile the color management system can interpolate and predict millions of colors--the entire range of possible colors each device will produce. No matter what the color device is and how it formulates its colorant mixes, so long as its output is "tagged" with a correct ICC profile, the correct color meaning is passed along to the next device's profile and retranslated to look correct on that device. Needless to say, the goal is to remix the RGB or CMYK "recipes" to keep those L*a*b* numbers, and image appearance, as consistent as possible.

For the whole system to work every device in the color reproduction chain must have a valid ICC profile, and every application involved must be able to "read" them, or, if they are missing, allow the user to tell it what the missing profile might be, and convert the colors using, once again, an ICC profile.

Color management is critical because without it matching a large number of colors between disparate devices is a manual and pretty hopeless task.

Enjoy your reading.

Mike
 
Last edited:
Mike, that was a great post.

My 2 cents:
Color management is the answer to the question: "Why don't these colors match?" The only OTHER answer is the paper. Of course one can't get bright vibrant color on uncoated paper like they can on coated paper (although we are asked the impossible ;)

It can be as easy or difficult as one wants to make it now. The easy solutions are here. Go to X-Rite. They have acquired GretagMacbeth and PANTONE. Ask them what you need. Just aim to ISO standards (GRACoL 7 for coated, ISOuncoated or the newest one for uncoated, since one is not available in the U.S. as official profile to aim towards, or SWOP, SNAP, etc., all depending on paper really, but all aiming to the same tone curve (aka NPDC aka Neutral Print Density Curve) under the 50% tones as close as possible, so the appearance is as close as possible between all printers and countries.

Using ICC profiles, the designer can know while they design (while soft-proofing) what the final outcome will be, no matter if for web (sRGB IEC61966-2.1, or maybe the new sRGB profile), or print (described above). No paper wasted. No many rounds of color correction. Efficient and worthwhile more than I can say, not to mention making it easy, finally taking the "mystery" out of printing and giving the customer a stake in the ground that they can design to.

If they soft-proof while designing and approval before it gets to prepress, we never would hear "Why can't we make that look brighter like that?" Limitations of the device or paper are the only answers left. And these are seen when soft-proofing and knowing when to assign (keep RGB or CMYK numbers the same as before, but assign different device to those numbers, changing appearance) and when to convert (change the RGB or CMYK numbers to keep appearance), but always using Relative Colorimetric Intent (if prepress) for production files (going to press).

The ICC profile needs assigned at the beginning of the workflow. The conversion to final output (press profile) comes at the end when sending the page to the device the color needs to be separated for (and if for press, would then get first sent to proof, converting from press profile to printer profile using Absolute Colorimetric Intent to show color of press paper on proof, and the same imposition would then get screened and sent to plates). Up until then, soft-proofing shows us what the image will look like when it is output on that device (press in my case), so we can soft-proof a CMYK device or Black Ink (grayscale) device from an RGB or LAB image.

It really couldn't get much easier now. Just convert RGB (use sRGB IEC61966-2.1 if an RGB image is untagged coming in to you) to CMYK, and leave CMYK and grayscale alone when coming to you. They are most likely using the old SWOP profile, which was the basis for the worldwide tone curve we're all aiming towards now, so these images will print close to the appearance they were intended to.

I haven't had color problems in over six or seven years except for a handful if you count where I was asked to match a customer's color LOL where it already didn't match well enough by the procedure I described above: Basically using the defaults in programs works, but critical color will need soft-proofing for the actual output device and using that profile to separate the image also.

Regards,

Don
 
No one said CM doesn't work...it is not the end all to all questions of color.

With the onset of CM there is now an even greater confusion and misunderstanding of color in the industry. It is not plug and play and requires education...the concepts, the terminology, the expectations, all need to be clearly understood by everyone in the process. More reality...it's just not understood. It has become a Pandora's box. Selling the concept is easy...actually applying it is another story.

Color Management is a tool to define and in turn control a subjective and variable process.

I could not begin to communicate the concepts with designers on one end, or my pressmen on the other, so I control the portion of the process that I can which in turn creates a more consistent result within my environment.

The concept of designers monitor to the printed sheet...meh...sorry...understand the concept, but I've never seen it practically applied. Great concept...maybe one day.

I just think all the sales pitches and standard creations, etc., lead to an unrealistic expectation that is very often laid on the lap of the printer, so I try not to make it seem prettier than it is.
 
For what it's worth?

For what it's worth?

Nicely written Mike!

It seems to me that what's underlying most of our differences of opinion with regard to color management practices and policies. Is that our own individual perspective's and experience's influence the standpoints that we often reflect in our posts.

In other words, specific practices that work well for, lets say a National Geographic Magazine type publication, often do not work well for, or apply to the shop who's specialty is direct-mailers printed on un-coated stocks.

Perhaps another way of saying this is that when we share our experience's with the forum but omit the context of where we acquired our experiences, it is not likely to be as beneficial to others as it could be if we hadn't omitted these contexts?

Just a sentence or two like, "I work in a High-end sheet-fed offset print shop and have frequent customer press-checks" or " I Work at a cold-set, web offset 4/C daily newspaper", etc etc. Should suffice.

I believe that Mike's post sticks with the fundamentals and framework of color management and therefore should be common to all of our experiences.

I think that perhaps the word "translation" might be a bit more informative than "equivalency" in the quote below snipped from Mike's post? Maybe not?
"Color management works like this: The output of every digital color device gets measured--not just its RGB or CMYK primaries, but hundreds or thousands of color combinations, and a kind of translation table is built that gives equivalency between the "device color," which is RGB or CMYK and the PCS, usually in L*a*b* units. This table says, "When this device is given so many units of R, G, and B, or C, M, Y, and K, it produces a color that looks like THIS." Or, when I feed the device a color that LOOKS like THIS it produces THESE RGB or CMYK numbers.
To use the analogy of translating between two different languages to further emphasize the boundaries we sometimes encounter between Color spaces.
So as an example, lets say that one language(LAB) has a vastly larger vocabulary than the language that we want to translate it into? And lets say that the language with the larger vocabulary contains the word "trogans" and it means essentially "my eyeballs are swollen and that my hair hurts in the morning",when translated into English okay?.

But the language that we really intend to translate the word "trogans" into, has only 400 words in it's lexicon(CMYK) and that it's lexicon is simply a collection of grunts, groans, quips and moans manufactured by Bruce Willis, Arnold Schwarzenegger and Sylvester Stallone during their most notable movie appearances;)
So what I'm trying to say is that, omitting the context and scope of the translation that defines our expectations and provides us with a reasonable yardstick with which to measure our results by, is required before we can intelligently discuss anything outside the scope of theory here.

Color management provides the best possible equivalencies admirably for the migrations between color spaces when properly informed, no doubt and no argument from me!

But if our clients are expecting to see the same gorgeous scarlet and crimson Sunset that they photographed at the beach the other night on Kodachrome film, to be matched by our 4/C web offset press running newsprint stock. Then we have failed to set the proper expectations for our customer.

Color management, be it digital or analog, be it well informed or not, has never been able to make a device with a feeble dynamic range match a device with a far superior dynamic range.

Best Regards
OT
 
I agree...depends on our personal situations. I'm in packaging, so it differs quite a bit from commercial and pub.
 
Software help

Software help

Nazdar has a photoshop plug in that can be very helpful. I use it almost daily. We print a lot of solid colors where a customer wants a PMS. Using CMYK to get PMS is in many cases impossible. Nazdar's CATZper is a tool that allows you to enter a LAB value. Then, it creates a chart that starts with that LAB color and then gives you similar ones near it in the spectrum. Then you can convert to a CMYK profile (I use SWOP) and print the chart. Once your customer has chosen the square they like you can pull the CMYK values from the chart and plug them into you art. Obviously this does not help with Photos, but it is great for logos and or solid colors.

You can even get a LAB value reading off previously printed or painted substrates and then start with that. Say, if someone wants to match a coke can.

You would need an eye one or other color reading instrument to get your LAB.

You can download a trial here.
Nazdar Consulting Services - Download Catzper©
 
Interesting thread

Interesting thread

It's hard to argue much with many of these good answers. Having a marketing, advertising, printing and business management background I come at the question from a very different (and non-technical) perspective.

I view color management as a set of production controls and standards (in the pressroom) and an ability to set customers expectations (the whole concept of a contract proof).

My answer to the second component has been to encourage printers to provide a remote proofing system to their better customers so that both the print shop and the customer have similar color expectations.

As a little shameless plug, we've recommended our free certified proofing system to handle setting expectations and enabling customers to get and send proofs instantly over a network. Free download at Veroproof.com. However, there are many other qualified solutions that accomplish this goal --- they whole idea is simply to talk the same color language.

My two cents,
-Scott
 

PressWise

A 30-day Fix for Managed Chaos

As any print professional knows, printing can be managed chaos. Software that solves multiple problems and provides measurable and monetizable value has a direct impact on the bottom-line.

“We reduced order entry costs by about 40%.” Significant savings in a shop that turns about 500 jobs a month.


Learn how…….

   
Back
Top