CMYK Values, SWOP, Gracol 2007 SWOP

Bill W

Well-known member
Greetings,

We are working with an artist that is asking that we convert his PMS (Spot) colors to process mixes based on SWOP values. He also indicates that his proof, while being printed on a non-postscript printer, is color accurate to SWOP - YEAH!

We are a flexo printer and prefer to print PMS colors as opposed to CMYK equivalents, however we are willing to work with a customer. In an attempt to find out what SWOP values are I tried the following:

Checked the values listed in the Gracol 2007 spec, which for 310 TAC 150% #3 are:
C= 57.06, -36.97, -45.05
M= 47.99, 71.92, -3.10
Y= 88.08, -5.09, 87.90
K= 18.97, 1.10, 1.18

Created a file in Photoshop whose working space is U.S. Web Coated (SWOP) v2 (8bpc) and the values are:
C= 62, -44, -50
M= 52, 81, -7
Y= 95, -6, 95
K= 12, 2, 0

The artist proof values are:
C= 56.20, -32.69, -47.19
M= 48.92, 65.57, -9.56
Y= 85.23, -5.54, 73.44
K= 15.44, 0.41, -0.56

I am guessing this artist does not have a spectro because his "color accurate proof" is, CMYK respectively, off by dEs of, 5.62,5.33, 7.28, and 4.14.

If we accept this order I am sure I will need to discuss color with the artist and I would like to know what are the correct Lab values for SWOP solid CMYK.

While checking out values I decided to convert the solid CMYK U.S. Web Coated (SWOP) v2 (8bpc) file to Gracol2006_Coated1v2.icc and check some numbers. They do not appear to match the numbers listed for Lab values in the G7 How To Guide, nor the GracolSpec2007. Additionally, if I start from scratch and create a file using the Gracol2006_Coated1v2.icc profile as a working space its CMYK Lab values do not match the numbers I get when I converted from CMYK U.S. Web Coated (SWOP) v2 (8bpc) to Gracol2006_Coated1v2.icc.

Numbers, I am confused!! I thought the Gracol2006_Coated1v2.icc was based on the '06 Gracol G7 work - am I wrong? When one converts solids from one color space I would expect them to be the same Lab values as if I had started in the conversion workspace - am I wrong?

Are the SWOP Lab values for solid CMYK values using CMYK U.S. Web Coated (SWOP) v2 (8bpc) as a working space the "real" values for SWOP? If so how do they relate to the GracolSpec2007 SWOP values?

Thanks in advance to all that answer.

"your favorite flexo printer"
-Bill-
 
Re: CMYK Values, SWOP, Gracol 2007 SWOP

Your customer is asking you to convert his PMS (Spot) colors to process mixes based on SWOP values? Why not just use the canned (150 lpi SWOPish) values that Pantone provides? They will be just as irrelevant to flexo printing as his proof likely is to SWOP.
I'm getting the feeling that your customer may not have communicated clearly what his actual target for color is. If it is to use a generic recipe (i.e. SWOP) then he is not concerned with the final color. If it is to match his proof, then you need to create custom recipes based on your print characterization data by measuring the color patches on his proof. If the target is the closest match to the PMS color out of process then you need to generate the recipe from your print characterization data by either measuring the PMS patch from a swatchbook or from the CIEL*a*b* numbers from Pantone.

best gordo
 
Re: CMYK Values, SWOP, Gracol 2007 SWOP

Basically what you have to do is to convert the PMS spot colors to CMYK. By saying that the conversion is according to SWOP means that the CMYK numbers are obtained using the SWOP ICC profile. In my opinion those numbers are useless to you, given that you are a flexo printer.
The simplest path, as suggested above, is to *create an ICC profile for your printing process* and then use a software tool like i1Share (freeware) to calculate the best match, using your press ICC profile.

If you have to match the customer proof, measure the swatches from his proof with your spectrophotometer and use i1Share again.

Regards,
Gabriel
 
Re: CMYK Values, SWOP, Gracol 2007 SWOP

Hey Bill,

Regarding your numbers. Looks like your Photoshop colorsettings are set to relative colorimetric as the default intent, no? This would account for some of the deviation in numbers. If you have a 100% cyan patch in an image with USWebCoatedSWOPv2 assigned to it, it will show actual Lab values of 66, -44, -50 in your info pallette using Rel Col, but 56, -38, -40 for Absolute col.

You also mention converting from USWebCoatedSWOPv2 to Gracol2006_Coated1. The correct profile in this case would be SWOP2006_Coated3.icc, not the Gracol profile, as this (swop2006_coated3)will have identical numbers (rounded) as the Gracol Spec for grade #3 paper. Moreover, converting from USWebCoatedSWOPv2 to SWOP2006_Coated3, you'll get near identical values using abs col, but not so using rel col.

Noteworthy:

USWebCoatedSWOPv2.icc = TR001 (circa 1995)
SWOP2006_Coated3.icc = TR003 (circa 2006)

Quote of the day
>They will be just as irrelevant to flexo printing as his proof likely is to SWOP.
 
Re: CMYK Values, SWOP, Gracol 2007 SWOP

Greetings,

Thanks to all that replied. I think I confuse some of you. My real question was - what are the Lab values for SWOP? Meaning when someone says their ink is SWOP compliant, what does that really mean? And when an artist indicates conversions for PMS to CMYK are based on SWOP, what does that mean?

My effort is to help our prepress / press communicate accurately with artists when they say something such as my PMS are converted using SWOP and be able to come to an agreement as to what color the artist is referring to. As Michael pointed out, there are a few SWOP profiles available in Photoshop.

Gabriel - Thanks for the suggestion of using i1share. I have glanced into that program in the past but never really used it. I did the other day and I have a question. When converting a PMS color to process mix using my profile, are the process mixes based on rel col and the dE differences based on abs col? Seems to me this would be the only way to get an accurate dE difference and the correct mixes to work on the substrate in my profile.

Gordo - I am curious why you say:
"If it is to use a generic recipe (i.e. SWOP) then he is not concerned with the final color."

Michael - I think I confused you when I started to talk about converting from SWOP to Gracol. This was actually in reference to another exercise in which I was trying to correlate between creating a document in one color space; creating a document in another color space and converting the second document to the color space of the first document. My expectation was that for, let's say solid cyan, the Lab numbers from the first document would be the same in the second document once is was converted to the color space of the first document - but they weren't. Since you know I am on the "make a good flexo profile" stage and I know that once we have one I will need to have a better understanding of profile space conversion, I am now starting to look at what happens in conversions so I will feel more confident to talk about conversions to our staff and our customers.

I was curious as to how you got your SWOP abs col numbers. I discovered that the conversion option that is set in Color Settings overrides the conversion option one choses in the Convert to Profile dialog box - kinda of scary. Additionally it appears to affect the values in a new document too and documents that are already open - really scary. Am I right about this?

Thanks again to all of you that answered - you observations and tips are my education.

-Bill-
 
Re: CMYK Values, SWOP, Gracol 2007 SWOP

RE: Gordo - I am curious why you say:
"If it is to use a generic recipe (i.e. SWOP) then he is not concerned with the final color."

The screen tint recipes to simulate Pantone spot colors that are embedded with applications do not reflect your specific print condition and substrates. They are generic. If the customer is concerned with the actual final color then they would base their simulation recipes on the specific print condition they are working with. Another way of saying this is that a single application screen tint recipe (e.g. what is found in InDesign or Quark) can not apply equally to flexo and offset. If the customer is concerned about the final color they would base their recipes on the CIEL*a*b* value of the target PMS color according to their specific print process. I.e. the recipes for any PMS color would change according to the printing process i.e. the recipes for flexo would be different than the recipes for offset however the final color would be the same.

make sense? gordo
 
Re: CMYK Values, SWOP, Gracol 2007 SWOP

Bill,

When you make a conversion for production files (that will be printing on a printing press), you really need to use Relative Colorimetric Intent (unless doing a conversion for one patch of color to get closest appearance when printing, which I'll talk about in a minute). For the most part, we'll want to use Relative Colorimetric for the simple reason that when we convert files, we don't want what was white, or paper, in the source color space, to become actual dots in the destination color space. This is good for proofing (if you have enough nozzles on the proofer to make the paper simulation of the source when simulated on the destination not look like actual dots because they are real small). So absolute Colorimetric Intent is good for proofing (send press-separated files to proofing rip, which does conversion from press profile to proofer profile using Absolute Colorimetric Intent). Otherwise, it's best to use Relative Colorimetric Intent. Does this get as close in appearance when printed as when using Absolute Colorimetric Intent? It depends on the color. So what I would do is, in an automated workflow, use Relative Colorimetric Intent if you choose to repurpose incoming CMYK to your specific press profile. If the resulting color is not close enough to appearance expected, THEN convert the source Lab values to your press profile using Absolute Colorimetric Intent AND Relative Colorimetric Intent to see if which CMYK values produce a closer appearance to the original Lab values. If the AbsCol CMYk values are closer, use them. If not, use the CMYK values you got when using Relative Colorimetric Intent. Pretty much those are your only two choices. The only other thing you could do is look at mapping of out-of-gamut colors in the program that makes your press profile. The program that makes the profile is what controls the way that colors in the source that are out-of-gamut for the destination, are mapped into the destination (so when using ISOcoated, you get blue turning purple, where you don't if you use e.g. GRACoL2006_Coated1v2). Even that helps only to a point (since we know that we can only accurately reproduce less than half of PANTONE's colors in CMYK).

Now, as far as looking at printed Lab values in Photoshop, there's only one correct way to do it:
1. Make sure your rendering intent in color settings is set to Absolute Colorimetric Intent (which means that while it is set like this, you must use Edit > Convert to Profile to do any conversions, and NOT Mode > CMYK, because you'll want to control what rendering intent is used for conversion from Lab to CMYK).
2. Set your second color readout in Info palette to Lab Color.
3. Make sure the correct CMYK profile is assigned to the CMYK document before even attempting to read the Lab values from the Info palette.
3. When you read a patch of color, look at the Info palette to see the PRINTED Lab values.

When you're done with your testing, remember to set your color settings back to Relative Colorimetric Intent.

Don
 
Re: CMYK Values, SWOP, Gracol 2007 SWOP

>I was trying to correlate between creating a document in one color space; creating a document in another color space and converting the second document to the color space of the first document. My expectation was that for, let's say solid cyan, the Lab numbers from the first document would be the same in the second document once is was converted to the color space of the first document - but they weren't.

They shouldn't be though. When converting from USWebCoatedSWOP to Gracol2006_Coated1, the source is SWOP, so for a 100% cyan patch you have Lab values of 56, -38, -40 and upon converting to Gracol, those values are maintained to the best of the profiles ability ...
For example, converting from USWebCoatedSWOPv2 to Gracol2006_Coated1v2 using abs col. : 100C, 0M, 0Y, 0K -> 56, -38, -40 -> 94C, 3M, 9Y, 1K = 56, -37, -40

This is desirable if you want to maintain the original intent of the image (i.e. you want the cyan to look like SWOP cyan on a Gracol press). Obviously this can be less desirable to convert solid colorants, as they will be "contaminated" to reach the require Lab values. If you wanted the SWOP cyan patch to have the same colorimetric values as a Gracol cyan patch, you would assign the Gracol profile to the SWOP image, rather than convert (i.e. preserve numbers).
This is desirable to maintain pure solid colorants, but the original intent of the image is compromised.

For most raster images, CMYK-Lab-CMYK conversions will not largely detract from the quality of the image, but the luxury way would be to use a device-link profile where one could control the purity of the solids and black generation while optimizing the image for a close visual match.

>I discovered that the conversion option that is set in Color Settings overrides the conversion option one choses in the Convert to Profile dialog box - kinda of scary. Additionally it appears to affect the values in a new document too and documents that are already open - really scary. Am I right about this?

Changes to the rendering intent within the Color Settings will change the Lab values in the info palette (for all open documents), yes. But I am not seeing where this over-rides the convert to profile settings (unless your using an output profile identical to your working space profile...a null conversion, in which case changes to the rendering intent will not show any differences). The Proof Setup dialog will NOT change the Lab values, so I'm thinking this is what your refering to?

>My real question was - what are the Lab values for SWOP? Meaning when someone says their ink is SWOP compliant, what does that really mean?

"SWOP compliant inks" is kind of an ambiguous term...more correct is probably ISO compliant. SWOP inks have to be ISO 2846-1 compliant and result in the colorimetric targets /tolerances on a given paper as outlined in the specification.
 
Re: CMYK Values, SWOP, Gracol 2007 SWOP

Good point regarding the use of the absolute colorimetric intent. Typically only for proofing.
 
Re: CMYK Values, SWOP, Gracol 2007 SWOP

I didn't find any clear explanation as to how i1Share does the calculations but the most logical scenario for me is:
The PMS Lab values are converted to CMYK values using absolute colorimetric rendering intent.

+"My real question was - what are the Lab values for SWOP?"+

In my opinion SWOP was defined so tolerant that it is not really useful to comunicate color appearance. For that one must use Lab or even better, a Spectral measurement.
The interpretation I give to your question and to your artist request is:
The artist was happy in the past with the way the PMS colors were reproduced when printing SWOP. So if you have the SWOP CMYK numbers that approximate a PMS color you can get the Lab values by using for example the SWOP v2 ICC profile distributed with Adobe CS2 package.
Create a new CMYK document in Photoshop, assign the SWOP v2 profile, type the CMYK numbers and then convert the image to Lab.



Gabriel
 

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