Pantone Color doesn't match the CMYK values?

[Lisa Herter]

I am trying to create a standards manual for several designers to use. We have 7 Pantone colors that will be printed in CMYK from Photoshop and Illustrator. It is important that the values are the same.

When converting Photoshop's Pantone Solid Coated to CMYK, the values are different than doing the same process in Illustrator. However, Illustrator's Pantone Solid Coated's values match Photoshop's "Solid to Process Coated" color book. What is the deal?

Why don't the 2 Solid Coated books match? Why isn't their a "Solid to process" swatch book in Illustrator?

Do I have to specify two different color books in my manual?

[Ann Shelbourne]

I am assuming that you are using Photoshop 7 and Illustrator 10.

Photoshop normally converts Pantone spot colors to Process by using Pantone's new Lab values and combining them with the CMYK profile that you are using to provide the best rendering for the press conditions that you specified when you selected your CMYK space.

You could, alternatively, use the Pantone Solid-to-Process Look-up-Table (LUT) which will not be affected by your CMYK profile.

Illustrator is not able to use the Lab space so you would use one of its color libraries instead. Among the available libraries you should see one called "solid to process" which is probably the one you want.

Probably your best course of action is to specify you swatches as CMYK break-downs; and use those values for all users, in all programs.

[dré as in Andrea]

Plus, yeah, what Ann said! :-)

[dré as in Andrea]

Hi Lisa,

I'm not familiar with the Photoshop and Illustrator books you're referring to. I have the Pantone Solid to Process swatch book, but they don't have anything to do with PS or AI.

If you want to convert your Pantone colors to cmyk, you may have to manually insert the values according to the Pantone book. Simply converting it will give you different values in PS and AI. For example, Pantone 287 in Photoshop breaks down to 100-83-16-5 for the cmyk values, and in Illustrator, it's 100-69-0-11. The solid to process book says 100-68-0-12. Go according to the book and you won't have any surprises on press.

Perhaps you can specify the values according to the Pantone book in your standards manual.

Andrea

[Chris Cox]

Lisa - you're using the wrong book in Photoshop. To match Illustrator and CMYK percentages, you need the Solid In Process book.

Solid Coated and Solid in Process Coated are two very different books. Solid Coated gives device independent colors (LAB), while Solid in Process gives approximations for those colors in a particular CMYK press condition.

Illustrator ONLY uses the Solid In Process books because it can't understand device independent color.

[Jay Chevako]

Don't use spot colors in a process job. Use the pantone process guide.
Jay

[John Slate]

Geez, I hate to play the devil's advocate, but it should be mentioned that if Lisa decides on a particular CMYK build to simulate her Pantone colors (which is the best advise at this juncture), and she sends this CMYK build to a variety of printshops, using a variety of contract proofing systems, and printing processes, she should expect some degree of variance in her signature color from project to project.

That has always been the nature of commercial printing, and an accepted fact of life.

Modern color management is an attempt to bring more consistency to color reproduction across a variety of processes by profiling them, but alas the current implementation only really handles photographic imagery reliably in the sense that a different CMYK is generated for each printing condition.

Once the RGB>CMYK conversion is done, color management is done.

Someday, there may be reliable in-print-stream CMYK to CMYK profiles, that all color-management-saavy printers will dutifully create and use to bring all their color worlds closer together, but until that day, (maybe never?) things will be done the old fashioned way, with each supplier basically reproducing the CMYK percentages you send them, with the occasional in-rip tranfer curve applied to a whole ripped page to account for higher dot gains...

[Mike Ornellas]

Someday, Adobe will stop sniffing glue.

[Tim Lookingbill]

John points out the hairball mess created with Pantone to process, bitmap and vector graphics using and not using CM.

I'm guessing AI wasn't designed to recognize tagged photos placed in AI layers due to the program's primary use in directly assigning CMYK values in vector data instead of using profile conversions.

But how does a designer control output color with tagged photos used in vector based logos built in AI? Build the logo in PS? Is PS able to tag images with it's vector tools? Place logo in a CM aware page layout program never using AI as a compositing layout program?

[Ann Shelbourne]

Tim:

Illustrator 10 and InDesign 2 are both CM aware programs: you can place CMYK images in either program and they will retain their color profiles. Just set all programs to use the same profiles.

The difference between the way that Photoshop can convert Pantone spot colors to Process by going through Lab, and utilizing the values of your CMYK profile to optimize color reproduction for your targeted press conditions, cannot be replicated by Illustrator because it does not use the Lab color space.

All you have to do, to make sure that you replicate Spot colors in Process inks the same way in both programs is to use the LUTs provided in the Spot-to-Process or Spot-in-Process color Libraries.

If you still insist on using QXP 4, you will need to specify those mixes with CMYK values because QXP does not use Pantone's new formulae.

[Jeffrey Smith]

I guess for those who subscribe to the idea that "color management" is a process in which to fabricate hyper-colors, trying to piece together vector and raster images, with or without cm, and attempting this task with the convoluted menagerie of cm utilities within software, could be a harrowing and troublesome experience...

[Ann Shelbourne]

Tim:
Your problem is that the programs that you are using (Pagemaker 6.52, AI 8 and PS 5) are simply NOT equipped to handle the problems that you are experiencing.

For color-management to work between programs and platforms, you unfortunately have no alternative but to upgrade to InDesign 2, AI 10 and PS 7.

Perhaps Adobe's Design Collection would be an economical route to take?

[Tim Lookingbill]

Ann,

Thanks for the reply. I'm not sure if you addressed my mixed vector/bitmap logo issue. I use Pagemaker 6.52, AI 8 and PS 5 and I'm financially in no position to invest in anymore graphics apps until I can understand and establish some kind of smooth workflow setup for all types of imaging for commercial purposes.

Right now, I'm not even producing anything. I'm still in research mode and this stuff is still hard to remember.

Regardless whether LAB or some other LUT library is used, it still appears to be a mess to have to deal with CM and assigned CMYK vector based values for smooth predictable imaging across apps, platforms and output. The meer mention of LAB and LUT makes my head hurt. It's hard to keep track of this stuff.

[Ann Shelbourne]

That might depend on how well the Someone understands the programs that they are using -- and if they are using the most recent versions.

Then it's up to you, the Pressman, to run your machines correctly (including the use of new blankets and properly washed-down presses).

[ Just know that if it is my job, I will be standing on the pressroom floor when you run my piece! ]

[Tim Lookingbill]

Wow, $900 for the Design Collection and another $1500 minimum for an Apple system to run it and THIS TIME IT WILL WORK AS ADVERTISED? Whoo! Hoo! Where's my checkbook?

Sorry, I need to get a job first.

[Jeffrey Smith]

"For color-management to work between programs and platforms, you unfortunately have no alternative but to upgrade"...

Follow that advice if you continue to believe the theory: color management is a process for magical color enhancement.

If you embrace color management with a realistic expectation, you will find better ways to invest your time, money, and effort...

[Ann Shelbourne]

InDesign 2, Illustrator 10 and Photoshop 7 run all on 9.2.2 -- what OS are you using?

And most of us can make Color Management work extremely well for us when using those three Adobe Programs -- Jeffrey seems to be the exception?

[Tim Lookingbill]

Vector and raster mixed?

I'm running OS 9.1 on one HD and OS 8.6 on another on a 9600/300. I just bought a used g3 500mhz powerbook for $800 on ebay and it isn't much faster than my 9600/300. I'm hoping to learn OS X on this, but I'm not sure.

But thanks for your feedback. I believe in CM. It's just making it work with what I've got for real world job scenarios is probably going to have to take a backseat to my other priorities I've mentioned.

[Ann Shelbourne]

<<Vector and raster mixed? >>

Yes -- in the programs that I mentioned.

But, as you said, Priorities first!

The programs will still be there once you have landed a job.

All good wishes for the hunt.

[Jeffrey Smith]

Ann, Don't get me wrong, I am not stating that color management [in whatever form you choose] can not be performed through the cm tools of those applications. And, I do not doubt for a minute, that people are successful in accomplishing their color management goals. However, I view the purpose of color management a little differently than the majority concensus of this forum. And the way color management is performed throughout my workflow, there is no need for this function to be accomplished at the image level by software.

[Ann Shelbourne]

If the Designers don't manage their colors correctly, particularly when building files from three different applications, how can you (or anyone else) possibly know what was going throught their little tiny minds at the time -- and what it is that they actually want you to produce?

[Bruce Fraser]

PageMaker 6.52 is fully color-managed. It has some quirks (like creating a temporary file for each instance of an embedded profile, which can eventually hit the OS9 OpenFiles limit), but it's workable.

The main problem with vectors is that .eps is not a format that's terribly amenable to color management -- the few products that do so have to parse the Postscript stream. Illustrator 8 is a special case in that it's possible to synchronize EITHER the interpretation of RGB or the interpretation of CMYK with Photoshop's, but not both. Illustrator 10 lets you do both. But if you build your vectors in CMYK, using the same CMYK as you do for your raster elements, the colors in both will match.

Your life will probably be easier if you accept the notion that the only invariant attribute of a Pantone color is its name. Try cracking the shrinkwrap on three supposedly-identical swatchbooks from different manufacturing runs and you'll see what I mean...

[Jeffrey Smith]

Ann, fundamentally speaking in terms of commercial offset printing, trying to produce what a designer wants, is completely backwards. If a misinformed designer believes that cm will magically enchance their imagery on an offset printing press, they are seriously mistaken. If an educated designer wants to take my profile - based on what press [web or sheetfed], based on what stock [coated or uncoated] and customized for their proofing device, then we are headed in the right direction. But, don't be too upset with how ugly the proof looks....

[John Slate]

But if you build your vectors in CMYK, using the same CMYK as you do for
your raster elements, the colors in both will match.

...and if you do it that way (I'm guessing this is the way Ann works) it doesn't much matter whether or not you use the latest Abobe Products with identical color settings. If you go by the numbers, the screen preview may not match vector to raster, but the print to a postscript device will (provided that the print process is not somehow further color managed).

Try cracking the shrinkwrap on three supposedly-identical swatchbooks
from different manufacturing runs and you'll see what I mean...

The only thing that has ever really been consistent in printing is its inconsistency. The processes involved are just as subject to the second law of thermodynamics as any other system.

On top of the standard variance through out a press run, or from this press to that, or this paper/ink to that paper/ink, is the factor of yellowing paper and fugitive dyes. I used to work for a print broker in Manhattan and my job included going to a bunch of different printers to OK jobs on press. My bosses had a real hard-on for matching previously printed material if any existed, and they would dutifully save office copies of every job they printed. Then they would pull out an old copy of something that we were reprinting and give it to me to use as a guide on press. It was pure idiocy. At times I almost had printers yellow up PMS inks to match an old sample! But I stood my ground.

To a certain extent, seeking this degree of consistency if a lesson in futility. Whatever the ideal color or condition of a printed image is (device independence?) is a fleeting thing once it exists on paper.

Everyone would be a whole lot better off if they accepted a certain degree of variance in print as a fact of life, knowing that the reality of a side-by-side comparison of all the pieces in a press run, for example, is not only unrealistic but pointless. I mean who really cares as long as the printing is in register, free of other defects and the color looks OK in and of itself?

Henry J Kaiser wrote, "The pursuit of excellence is gratifying and healthy; the pursuit of perfection is frustrating, neurotic and a complete waste of time."

Amen

[Bruce Fraser]

We generally try to aim for the middle of the barn door, but different shops have different tolerances, and different presses have varying degrees of stability.

If I'm printing 25k impressions on a 2002 Heidelberg Speedmaster with a CPC system, I expect to see a great deal less variation between the first and last sellable sheets than if I'm printing 500K impressions on a 1963 Toshiba web press.

If a particular pastel tint build on page 64 needs to match the same tint build on page 65, even though they're on different signatures, I'll grill the print shop as to whether or not they can hold a match, and if they can, how good a match they can hold. Generally, I pay for what I get, but I insist on getting what I paid for. Usually, that's a good deal more than "the printing is in register, free of other defects and the color looks OK in and of itself."

(Which means you may not want my business...)

[John Slate]

I consider "matching" color within a single publication "in and of itself".

Don't get me wrong. I'm not advocating anarchy, or throwing out the proverbial baby with its bath water.

I have always worked at high quality shops and have dealt with fine color adjustments often. I consider quality to be job 1.

It's just... hmm... I don't know WHAT I mean. There should be limits to people's expectations that's all...

OK I'll leave now...

[Bruce Fraser]

I often think that a Client Expectation Management System would be a killer app...

I hear what you're saying. It's just that some people use what you're saying as a lame excuse for not doing their job -- I don't number you among them!

[Mike Ornellas]

There is one word that can summarize the commercial graphics industry.

TURD

That's why color mgmt. at its current development, is a wax job.

[Jeffrey Smith]

Bruce, let me ask you this question: what image would you prefer to work with?

Image A - captured using a device with a low dynamic range, has remained in original RGB space, has not been altered.

OR

Image B - captured using a device with a high dynamic range, final output of image has not been determined, but has been converted to CMYK.

[Jeffrey Smith]

Mike, I understand your method of freezing a spot color within an RGB image for the eventual and reliable conversion to CYMK. But how does this resolve the issue and discrepancy of the two application's spot > CYMK conversion?

[Mike Donahue]

Lisa,

Back to your original problem. Chris and Ann both mentioned using the "Pantone solid to process" color book in PS to get the same values as AI.

However, if you use them in an RGB image as part of the RGB info, when you convert to CMYK by whatever means the values will indeed still shift. One way you can get a around this is to place elements that require matching process values onto spot channels.

You can create spot channel(s) (be sure to use the Pantone solid to process color book) in your RGB file, place elements on the proper spot channel(s), then when you convert to CMYK their values remain as spec'd by the Pantone breakdown. After your file is converted to CMYK you can merge the spot channels if desired back into your document.

Make sure you take care of knocking out the appropriate areas of the background. Spot channels overprint by default.

HTH

[Jay Chevako]

"Mike Donahue"
You want to use spot channels to make sure your cmyk values come out the
same.????
Jay

[Mike Donahue]

Jeffrey - if I'm understanding your question correctly there won't be a discrepency if you use the right color books. PANTONE solid to process coated in PS and PANTONE solid coated in AI.

Jay - the spot channel remains a spot channel when the conversion is done. This way the values don't change. You could even do this with custom values of your own making.

[Mike Donahue]

I guess I should say that this worked for me using:
AI 10.0.3
PS 7.0.1
OSX 10.2.6

If you're using older software with outdated Pantone books your numbers may not match between apps but the conversion technique will work.

[Tim Lookingbill]

Bruce or Chris Cox, could you direct me or provide a link where Pagemaker 6.52's temp file/embedded profile instances quirk with OS 9's OpenFiles limit resides? Where's it buried? I'ld like to read up on it.

Thanks.

[Bruce Fraser]

I don't know if anyone has ever documented it.

But if you keep placing images with embedded profiles in PageMaker, it will, at some point, crash, and when you reboot, you'll find the trash is full of temporary profiles.

Exactly when you crash will depend on what else you have open at the time...