Pantone 295U is off from my pantone book

[Guest]

I am working with a client that uses pantone 295u as their primary color. I looked it up in my physical pantone book and it is a nice dark blue even in the uncoated guide. But...when I use the color book in ilustrator and pull up PMS 295 it looks like a 75% tint of the color, its not the dark blue thats in the printed guide book. I tried using overprint preview to see if the LAB preview might help, but with this color it makes no difference. Has anyone run into this? I'm not sure how to work with this as I cant show this greyed version in my proofs with the client as they will think I have not selected the right color.

Thanks

-KC

[Terence879]

When Adobe and Pantone made a huge formula shift back in 2000, at least there was a warning that came up regarding a color mismatch when opening a legacy file. You had the option of applying the new or keeping the old version. Adobe also changed the suffix on the library from CFC (or CFU) to just plain C (or U). There was no such change when updating to the Plus colors and their lab values in CS6. The program is seeing the PMS number in the old files and changing the values to LAB automatically.

Adobe could have done a better job of pointing out this major shift in a CMYK color system that has been in place for decades and was working just fine, even though to some it was imperfect. Something a little more than a paragraph buried in an online list of features. They could have also done a little better on making the transition a little more user friendly, ie by allowing the keeping of old book values in a dialog box on opening a pre-CS6 file. Those values are not meaningless after decades of use and to cast them aside as though they are is very presumptuous  of both Pantone and Adobe. They could have made sure the Book Values button worked consistently and that it was present in InDesign.

Not everyone is mixing spot inks using Pantone mixing colors. I'm sure Pantone wishes that were true, but it is not.

Quite frankly, I don't care what it looks like on my monitor. I really don't. Its the print result that matters.

Adobe always forgets about the wide format printers who happen to rely on the CMYK libraries.

The presses in wide format are not limited to the dull CMYK colors of a regular 4-color plate press. The goal is to be able to hit the Pantone color using a build based on the machine and the media. The outlay to do the updating on all the profiles for all the media is somewhat of an unexpected expense.

That aside, the fact that we are getting color conversions on opening an older file in CS6, with no conversion warning or option to even select a color space caused a lot of reprints in pre-workflow test runs on release of CS6. It is even possible now to get a legacy design or logo file with 3 different methods of calling out a single PMS color.

Is the file built incorrectly? Yes. 

Should we as digital printers be responsible for catching this? No, but we are.

We are slowly getting this up to speed. I had actually argued for a LAB color environment in the past, but didn't exactly expect to implement it in quite this way.

Most designers aren't even aware they are creating a problem.

Some digital printers still aren't aware of all the color shift ramifications.

As media manufacturers update online profiles there is going to be some confusion.

Until the dust settles it is very important for designers to work closely with their printers to get the results they intend.

[Monika Gause]

Problem is: Adobe can't keep the old values, since they are not contained in the book anymore which is delivered by Pantone.

[Terence879]

They weren't when they did the change in 2000 either. But there was an option for the first release of Adobe programs with the new colors.

Sure it was phased out on the next version.

I think what most people here are objecting to is that there wasn't a choice or a warning.

Just an arbitrary end to CMYK as it had been used for most people's careers up to that point.

[Cam_Reesor]

Hi Terence:

You really sound like you've got your S together (and your H,B,L,a & b). I don't suppose you'd like to chat about this on skype sometime? The problem is really more deep rooted than all of this. The core color models that we use in CS are flawed:

Just make a color cirlce in photoshop (RGB or LAB). Give it any pure color (H=whatever, S=100%, B=100%). Save and go back in. Now go to the L value and just change it. Watch what happens to the H value. It freakin' changes. A change in L value should not affect the H value. Whenever I'm speccing very specific colors: the only way to do it is to go back and forth changing L and H values until CS finally gets it that I want H=125º, L=45, S=100 for example.

The only efficient/intuitive way to spec color is by H value to start. If you're working with even the most basic of color symetries, you have to work with the knowledge of geometry.

CMY is intuitive becuse its basic addition, but you have to be a memory genius to move from color to color. And there's no intuitive way to work with symetry.

RGB is harder because subtraction is just plain harder than addition.

That leaves LAB

Any pure color has an inherent Luminosity value that is ignored by HSB (HSL), so you have to use L. You have to be a rocket scientist to spec color in LAB (No mere human can do it), so you have to use H. None of the models are satisfactory environments for speccing color. We dearly need an HLS (Hue, Luminance, Saturation (grey value)) model. And it's entirely possible.

Make any gradient in whatever program from whatever compliment to whatever compliment. You get grey in the center. CS is thinking "as the crow flies". When I'm goin' from pure complement to pure compliment according to the Laws of Color: I'm getting one or the other of black or white in between, not neccessarily in the center. If I don't know that, I can't possibly spec color wisely. Black and White are as unique as any other color. They both happen once at the intersection of all pure colors at there respective L values. There is no redundancy whatsoever. A color model that requires a spectrum of black and white to make it work is inadequate.

You can't make a purely subjective evaluation of any two colors without first removing their inherent differences in both Saturation and Luminance. The color cirlce we all work with provides the fabulous geometric symetries we need. It also balances pure color by Saturation. BUT, it neglects to address the inherent Luminosity of each. That circle is an axonometric projection of a true color model. It's an optical illusion. A parlor trick. Pure colors do not co-exist on a circle. Pure Hues do.

The only way I can do advanced color work in CS is to use a grey scale proofing setup and adust colors visually to known grey value references (backgrounds). I can't do that kind of work the way I want to: by dialing them in. I don't want to rely on my "eye" for Luminance and Saturation differences in colors, I'd rather just dial them in, but there are no dials.

If you don't think this is a big problem: then you also believe that balancing an image, based on a neutral reference, is not changing an images intended H values. Whether you're in LAB or not, the freakin' Hues are changing. We all do this everyday at professional levels and fabulous images are being corrupted.

If you want my Skype ID, let me know.

C

[Terence879]

Cam, I don't purport to be a color whiz. I'm more a mechanic than a theorist.

I'm very adept at color matching in the CMYK color system for the simple reason it is so easy to visualize. I can work RGB to get the results I need when printing to Lambda or Lightjet. Plus I've been doing it for quite a number of years now so I'm sure things can be adapted to work with whatever system I'm presented. It just would have been nice if Adobe (and Pantone) had reached out to their user base a little more emphatically on such a major concept shift.

Gradients are a pet peeve of mine. As with most things Adobe, they figure if the algorithm works for something the size of a magazine spread, it works for everyone. Vector linear gradients are the bane of a wide format printer's existence....(well, that and transparency.) Not only do you get the fade to gray, you get step banding that has nothing to do with whatever rip might be used.

Another pet peeve of mine is the propensity for reliance on neutral reference limits when it comes to color control. My object is to get the best color match and image depth possible. Sometimes I'm limited by the file material I am sent, but I never knowingly work with an outsource printer that uses a system that cuts their color gamut to fit a 'streamlined' work flow.

Which brings us back to the OP's topic.

The shift in color systems at this juncture is more than a little extreme. It has caused major and significant changes in pre-flight of files and it will most likely cause an increase in cost. As I noted previously, designers, a lot of them, are not even aware there has been a major upheaval in the way they apply colors now. They are corrupting their earlier files by opening them in CS6 and they are wondering why print jobs that went fine before are now being "messed up by the printer."

[Cam_Reesor]

A pure color gradient in CS can't be specified if it's more than 60º.

Even then, you have to watch out that your gradient doesn't cross over a primary or secondary.

If you add any variation of B or S, you're gonna get a mess.

In the top gradients, CS turns what should be a pure color spectrum into greys. It also misinterprets the Hue path.

In the bottom gradients, I've told CS not to do that.

This is not a problem with color. This is a problem with CS.

[Cam_Reesor]

If Adobe wanted to make serious dough, they would stop pandering to Pantone. They would specify fundamental pallettes based on H and L values and Pantone would create their pallettes based on that.

When I first hear about Pantone Goe, I thought things would get magnificent. But no one bought into it, and rightfully so.

There are fundamental compliments that are easy to spec. If you choose any Pantone color, just try to find it's compliment in the library, let alone it's split complement etc.

[Kals]

emil emil wrote:

No one can tell how each person comprehend colors but all people with normal vision will agree if a color is the same. So no color profiles needed and no one will use your patent . The colors are different wavelengths  that reach the eye and these wavelengths can be measured with a colorimeter. So defining a color with such measurement guarantees that all people will agree it is the same color and that's what Lab color space is based on.

On the other hand different devices don't agree that the same numbers are the same color and this is the problem color management is designed to solve.

No… If these devices could talk, they would all say that it's the same colour! The only reason we know the colours are different is because we can see what each one sees and compare them—I mean that's the whole purpose of a computer display or printer, to show us what they see. Unfortunately (or perhaps fortunately!) no one else can see exactly what our eyes see. I think a device which lets us see the world through someone else's eyes would make me very rich indeed. But I'm getting a little off topic now.