'Pantone Colour/CMYK/etc Matching For Idiots' advice, please..?

[jitterbug]

Okay, so my educational background (Photography/Illustration/Design) has not been a particularly technical one. I've just taken on what I thought would be a very simple job of finding CMYK/RGB/Web colours for a number of given pantones to be used in branding.

I thought, in my innocence, that I would simply pick each colour in Photoshop (for example), go to the picker, and copy down the different values from there. (I know and accept, of course, that it can be difficult to get an exact visual match for certain colours.) However, I noticed that the numerical values I got this way varied a little from the example given with the job.

Then, I found a couple of examples from Pantone's Color Bridge book; I tried (in Illustrator this time) picking the colours I had found and then converting them to CMYK, to see how close they came to the 'official' conversions. The answer is, not very.
One example: Pantone 1767C. Illustrator found: C-0/M-27/Y-12/K-0.
Color Bridge has this as: C-0/M-32/Y-10/K-0.
Photoshop thinks it's: C-0/M-38/Y-9/K-0 (!!!)
This is enough to make quite a marked difference.

What I do not understand is the inconsistency. Surely if Pantone is an industry standard, then there should be a set of standard values for RGB/CMYK/Web that are simply pre-loaded into software such as CS3 etc. Why the variation..?!

I've tried looking around on the web, and I'm just getting a migraine. I cannot find a simple answer to my question: why are they all different, and how does this work?

If you had this job, then how would you attempt to find the values?

Many thanks in advance...

[Soilworker]

Jitterbug,

I understand your frustration. I've always wished (naively) there was a reliable standard sort of conversion for Pantone solids to Process CMYK... it would make things so much easier. But, I understand why that's not the case.

Pantone inks are meant to be special colors that can't quite be hit by a simple CMYK mix. They are typically out of the the CMYK color gamut. What this means is that there are only so many colors that can be faithfully represented by using just CMYK.

The inconsistency has to do with how far out of the CMYK gamut the Pantone is. Some are closer than others. The closer they are to the CMYK color gamut they are the easier it is to simulate using just CMYK (you need to be careful to not surpass a certain amount of ink coverage in doing so, though). Some colors you just won't hit, though, so you have to settle for getting as close as possible and that's a matter of opinion.

As for the conversion inconsistency between applications that's a good question. They might just not care because it's not going to be exact no matter how you convert the color. However, it would be nice if Adobe would at least keep it consistent throughout their software.

The only way to find a good conversion value (IMHO) is to color correct. That's what we do when we need to simulate a Pantone used in one of our printed products in a CMYK catalog/mailer etc. In order to do that you should have color managed proofers running special icc profiles and have a reliable Pantone source to compare your converted output to. Even then you should have special lighting so that you can view the two side by side correctly. Also you have to take into account the type of paper the final product will be printed on since paper color affects the overall appearance of color... that's what the icc profiles do... simulate the paper color etc. No matter how white you think your paper is it's going to be at least a bit darker than what you see on screen.

Well, enough of my convoluted explanation. I hope it helps you more than it confuses you.

~Soilworker

[jitterbug]

Thank you very much Soilworker- that was very helpful, actually.

It has helped to clarify things somewhat; I did understand (sort of!) the gamut issues with CMYK reproduction, it just does my head in that the various bits of software have such wildly varying ideas on how to match each colour. I just want to be able to provide the information for my client properly without them finding down the line that I've done it 'wrong'.
*sigh*

Thanks again.

[Werby]

Soilworker makes the excellent point that most Pantone colors cannot be hit by ANY combination of CMYK, and therefore any conversion is just a rough simulation. Pantone makes their money by creating colors that are visually different from colors that can by created by CMYK, so they have no real motivation to make it easier for you (or Adobe) match those colors. They want to sell INK, not just swatch books.

One reason for inconsistent CMYK values is that there are different kinds of CMYK. Every combination of press, inks & paper can produce a different color with the same CMYK values, so CMYK numbers only have meaning within the context of the printing condition they have been created for. That's why there are publicly available datasets and profiles published by groups like FOGRA, SWOP & GRACoL, which define a sort of average visual appearance for a given type of printing on a given class of paper (#1 coated, #3 coated, uncoated, etc.). So any software that is going to convert a PANTONE into CMYK has to consider what kind of CMYK printing you are going to be doing. First it has to look-up the L*a*b* value of the pantone color, then it has to translate to the color profile you have selected in your color settings using the rendering intent you have specified. If you pick a different profile in your color settings, you will get different CMYK values out (try it!).

I really don't know why Illustrator and Photoshop can't agree - They don't even agree on the LAB values of PANTONE colors, so their CMYK or RGB translations are of course going to be different. You would think that at least they would have the same LAB look-up tables, but apparently not. Maybe someone on this forum knows why they have different LAB start values.

Another factor is that even within a single CMYK "color space" (like GRACoL2006_Coated1), a given LAB color can be rendered by a variety of CMYK values, especially muddier greens and browns that use all 4 inks. If you want to see wildly divergent CMYK values, lookup 385 C in Photoshop and Illustrator! The point is that just because two programs give you different CMYK values, doesn't mean one is better than the other or that they will even appear visually different when printed.

Wow, what a minefield! Just as PANTONE wants it to be! Soilworker is correct in that you may just have to print out a bunch of swatches on a proofing system that is set-up to match your printing condition as closely as possible and pick the one that works best for you. IF you can get your hands on an official LAB look-up table from PANTONE, then you can convert to the correct profile and get good starting CMYK values, but you may want to do further tweaks from there.

By the way, PANTONE isn't an industry standard, they have just been around for a long time and are very well known. They are in it to make a buck, not make color matching easy.

-Todd Shirley

[jitterbug]

Hi Werby- and thanks for the further clarification.

I had noticed that selecting different CMYK working spaces produced different results, but wanted to keep my question simple, in the hope that the answers would then be simple enough for me to follow!

As far as opting for one colour space or type of printing in order to aid the selection process? Here is the problem; this is for a set of guidelines that will be sent out internationally by the client so that different 'branches' keep to the brand colours consistently. Even if we were to specify the workspaces/printing processes to be used along with the various values/logos etc, I cannot imagine these sorts of specifics being adhered to. Meh!
The best I could do, I suppose, is aim for average- i.e., which is the most commonly used colourspace?

On the up side, I'm finally filling in some of the gaps my degree left me with, with your generous help. Thank you both!

[Lukas Engqvist]

Originally Posted by Soilworker

Pantone inks are meant to be special colors that can't quite be hit by a simple CMYK mix. They are typically out of the the CMYK color gamut. What this means is that there are only so many colors that can be faithfully represented by using just CMYK.

Sorry I think a slight missunderstanding. What makes a Pantone colour a Pantone colour has nothing to do with the gamut. Many Pantone colours are reproducable in CMYK. The difference is that they are out of the bucket colours. Yes there are Pantone colour that cannot be translated to CMYK, but it is not the lack of ability to reproduce a colour that gives it a "pantone"ness.
Some pantone colours are used to print with 2 colours or large solid areas, or match with textiles, all sorts of reasons.

Neither Pantone colours nor CMYK colours are independant of the substrate, paper. (Compare Pantone 116 C and Pantone 116 U)Same ink on different paper gives different results. This is why in recent software there is the possibility of using standard LAB values for pantone. The LAB is then converted colometrically to the CMYK values of your output profile, which if you have a good output profile describing your process will give better accuracy.

[Soilworker]

We are all astounded by your vast knowledge, Lukas, but I was only speaking in general terms considering the background of the audience at hand and for the particular question. Sorry if my phrasing might be a bit off or not quite in depth enough for your liking.

The bottom line is that in order for Jitterbug to obtain a good match, regardless of what makes a Pantone a Pantone, he/she should, as I stated, color correct taking the paper into consideration, and proof on a color controlled proofer with profiles set to simulate various papers until a desirable simulation is reached. Even if LAB values are used.

We're all here to help each other, not thrash the way questions are answered.

[Lukas Engqvist]

Sorry if you took it as getting at you, I wasn't getting at you in any way, it was not intended.
Seen the argument that "Pantone is not reproducable with CMYK", and just wanted to stop the missconception spreading (seen it in other forums too).
The idea comes from I want a bluer blue so I have to go Pantone and add a fifth colour. But we often print gray pantones, not because they are not reproducable, but because we do not want raster points or missregistration. This is why at times it is good to know that a Pantone colour can sometimes replace a CMYK with identical colour value.
Was not at all questioning your mothodology for gaining a proof that would predict the results. A proof is after all the only "dressrehearsal" we get before printing a job, and usually the contract that binds the customers expectations.

My other half of the message was meerly to point out that there is a general sloppines in using pantone scale and often find customers quoting a pantone coated colour even if printed on uncoated. To be aware of what the pantone system is and how it works.

[Soilworker]

I understand what you're saying, Lukas. No hard feelings... just having a crabby morning... I apologize. Perhaps just a miscommunication/misunderstanding then. I may have partially mis-spoke about CMYK/Pantone matching but given the nature of the post I didn't feel the need to go so in depth but you did bring up good points and believe me, they are one's that frustrate me on a daily basis in my dealings with certain folks.

Cheers!

~Soilworker

[Padzilla]

I've found, in my experience, that the Color Bridge conversions are more accurate, compared to any app. We did a bunch of tests on our Epsons, and even the Color Bridge colors needed some tweaking, depending on the substrate. There's a lot of variance, even when you look in the Color Bridge swatch books.

We created color palettes, in InDesign, with our custom brand palettes. About 70 percent of our design files ultimately end up in InDesign. In theory, we can just have anyone working for that brand load the palette, and work in those colors.

If you're picked a printer for this particular job, they may be willing to run some pres tests. We've run a number of 6 and 8 color press tests, to compare PMS/CMYK variations on specific paper stocks. I have seen brand swatch books, for companies, that have their colors printed, in CMYK and PMS, on their brand stocks. Sometimes instead of CMYK values, they have LAB values printed in the book. It seems like a lot of work, but in the long run, it saves a lot of headache.