I saw a question about how to get the right color on http://forum.adobe.com/webx/.3bc493d8
The question was concerning RGB, but I have a job in today (one of the rare ones that actually use Photoshop 4 Default CMYK ICC profile for everything, and really I just need to get with this customer and get us both on the same page as far as what profile to use throughout). This job also includes PANTONE TPX (Fashion and Home Textile) colors spec'd as Spot but given CMYK values. The Illustrator document was not tagged with a profile (so assuming the default SWOP profile), but the images were tagged with Photoshop 4 Default CMYK ICC profile. I couldn't get the Lab values in Illustrator since I don't have the PANTONE TPX library. What to do to make sure the colors are close to what the customer intended? Well, most of you know that I've said many times that I don't color manage incoming CMYK. But I also get most stuff in as SWOP or untagged.
Today I decided to see if I could get the Lab values for the TPX color(s) and then convert them to GRACoL2006_Coated1v2 to get my CMYK values. I couldn't find the Lab values for PANTONE TPX colors. So the next option was to assign the correct ICC profile to the CMYK values given for the PANTONE TPX colors in my Illustrator document. But which profile? SWOP or Photoshop 4 Default CMYK? Well, since the images were tagged with Photoshop 4 Default CMYK ICC profile, I assigned that ICC profile to the CMYK values of the TPX colors (it's just a guess because when you don't have the Lab values and ICC profiles aren't embedded, who really knows what it's supposed to look like).
After reading the post mentioned above, I decided to work with the Mac OS X ColorSync Utility's Calculator to get my "correct" CMYK values for GRACoL2006_Coated1v2. So I opened ColorSync Utility and went to the Calculator tab. There on the left I chose my color space I'm converting from: CMYK, chose the ICC profile I'm converting from: Photoshop 4 Default CMYK, and chose the Absolute Colorimetric Intent (to get printed values I'm converting from, which is the closest to the Lab values I'm gonna get it looks like since I can't get the Lab values of the PANTONE TPX colors - and I'll say this is not the best way to go about it).
In ColorSync Utility's Calculator, I saw that the max CMYK value that can be put in is 1, so I had to take the CMYK values given in Illustrator, move each value's decimal two places to the left and put in those values for my CMYK. For instance, I had CMYK values 37.25,21.57,17.25,3.92 in Illustrator and so I put in CMYK .3725,.2157,.1725,.0392 in ColorSync Utility Calculator.
In ColorSync Utility Calculator, on the right, I chose CMYK, GRACoL2006_Coated1v2, and Relative (because Absolute wasn't available, and although I wouldn't expect the intent chosen on the right side to actually make a difference in values given, it does make a difference for some reason). The CMYK values I got (what I need to get close to the correct color on the print) is CMYK .2553,.1601,.1378,.0001. That translates (after moving the decimal two places to the right) to CMYK 25.53,16.01,13.78,0.01.
I wanted to check the values, so on the right I changed the colorspace to Lab. The Lab values I was given were 76.7422,-0.9844,-5.5391. I then changed the colorspace on the left to Lab and the intent automatically changed to D50. I saw that the values on the left and right were identical (so now I'm converting from Lab since Lab is chosen on the left). On the right, I chose CMYK. Guess what happened? Automatically the intent on the left changed to Absolute Colorimetric Intent. The intent I've said for years is the best until recently when I've basically given up and saw that the only intent that would be used in Adobe InDesign or Quark to convert Lab (such as PANTONE Lab) to CMYK is Relative Colorimetric Intent. So the intent to get the best reproduction of a Lab color is not being used in the programs, and unless something changes, the best intent will still not be used for PANTONE Goe colors to get the closest reproduction for those colors. Although I've tested and at time had better reproduction using Relative Colorimetric or Perceptual, Absolute Colorimetric gives me the closest reproduction on the most colors in my testing. And Apple's ColorSync Utility Calculator defaults to this intent when Lab is chosen on the left and CMYK on the right. So let's see what CMYK values I get when using the above Lab values on the left, (the default) Absolute Colorimetric Intent, and on the right choosing CMYK, choosing GRACoL2006_Coated1v2 (and leaving Relative on the right although I wonder which is best to use and why changing it makes a difference). I get CMYK .2601,.1557,.1331,.0000. That translates (after moving the decimal two places to the right) to CMYK 26.01,15.57,13.31,0. Choosing Perceptual instead of Relative on the right makes the color 2-3% lighter in each separation. Still a long way from what I would have gotten on press printing the CMYK values I was given on GRACoL2006_Coated1v2. But which is correct? Really I don't know since I don't know the Lab values of the PANTONE TPX colors, and I can only work with what I have.
So we see that printing CMYK 37.25,21.57,17.25,3.92 on GRACoL2006_Coated1v2 is a lot different than printing 26.01,15.57,13.31,0 on GRACoL2006_Coated1v2, but unless we know the Lab values we're aiming for, it's all just an exercise. And since my customer tagged the images with a profile in Photoshop but didn't tag the Illustrator document, I can't say that the CMYK values I received were made for SWOP, Photoshop 4 Default CMYK, or what printing condition they were made for. But I can tell you this much: I will most likely use this ColorSync Utility Calculator for all my color matching in the future (if I can remember), because I like that Apple knows that Absolute Colorimetric Intent needs to be used to get the closest reproduction in CMYK when coming from RGB or Lab, and that's what Apple defaults to.
When we get the PANTONE Goe colors in our apps, how is Adobe and Quark (and Microsoft?) going to deal with them? Will they let Relative Colorimetric Intent be the default for all color conversions, or actually use different rendering intents depending on the subject to get the closest reproduction? Apple knows that in most cases in converting to CMYK, Absolute Colorimetric Intent is the best theoretically, so it uses it. Because we don't want paper in the source to become actual printing dots in the destination (to see this in ColorSync Utility Calculator, make left and right to both be CMYK, make values on the left CMYK 0,0,0,0, and on the left choose Absolute Colorimetric. You'll see that the values on the left don't stay 0, depending on what profile is chosen, and whether coming from RGB on left or Lab on left, values on right do not stay 0), Relative Colorimetric must be used for stuff that is already CMYK (if color managed at all), but is not the best to use when trying to get the closest reproduction on a color (but we must use it because it does put dots on paper where there should only be paper).
Even PANTONE said in the Goe White Paper that ICC color management had it's problems and that the vendors would have to come up with a way to get the closest reproduction of the Goe colors. What's going to be done to fix this situation and let us be able to as close as possible reproduce the Goe colors (or any colors as long as we know the Lab value)?
Don
The question was concerning RGB, but I have a job in today (one of the rare ones that actually use Photoshop 4 Default CMYK ICC profile for everything, and really I just need to get with this customer and get us both on the same page as far as what profile to use throughout). This job also includes PANTONE TPX (Fashion and Home Textile) colors spec'd as Spot but given CMYK values. The Illustrator document was not tagged with a profile (so assuming the default SWOP profile), but the images were tagged with Photoshop 4 Default CMYK ICC profile. I couldn't get the Lab values in Illustrator since I don't have the PANTONE TPX library. What to do to make sure the colors are close to what the customer intended? Well, most of you know that I've said many times that I don't color manage incoming CMYK. But I also get most stuff in as SWOP or untagged.
Today I decided to see if I could get the Lab values for the TPX color(s) and then convert them to GRACoL2006_Coated1v2 to get my CMYK values. I couldn't find the Lab values for PANTONE TPX colors. So the next option was to assign the correct ICC profile to the CMYK values given for the PANTONE TPX colors in my Illustrator document. But which profile? SWOP or Photoshop 4 Default CMYK? Well, since the images were tagged with Photoshop 4 Default CMYK ICC profile, I assigned that ICC profile to the CMYK values of the TPX colors (it's just a guess because when you don't have the Lab values and ICC profiles aren't embedded, who really knows what it's supposed to look like).
After reading the post mentioned above, I decided to work with the Mac OS X ColorSync Utility's Calculator to get my "correct" CMYK values for GRACoL2006_Coated1v2. So I opened ColorSync Utility and went to the Calculator tab. There on the left I chose my color space I'm converting from: CMYK, chose the ICC profile I'm converting from: Photoshop 4 Default CMYK, and chose the Absolute Colorimetric Intent (to get printed values I'm converting from, which is the closest to the Lab values I'm gonna get it looks like since I can't get the Lab values of the PANTONE TPX colors - and I'll say this is not the best way to go about it).
In ColorSync Utility's Calculator, I saw that the max CMYK value that can be put in is 1, so I had to take the CMYK values given in Illustrator, move each value's decimal two places to the left and put in those values for my CMYK. For instance, I had CMYK values 37.25,21.57,17.25,3.92 in Illustrator and so I put in CMYK .3725,.2157,.1725,.0392 in ColorSync Utility Calculator.
In ColorSync Utility Calculator, on the right, I chose CMYK, GRACoL2006_Coated1v2, and Relative (because Absolute wasn't available, and although I wouldn't expect the intent chosen on the right side to actually make a difference in values given, it does make a difference for some reason). The CMYK values I got (what I need to get close to the correct color on the print) is CMYK .2553,.1601,.1378,.0001. That translates (after moving the decimal two places to the right) to CMYK 25.53,16.01,13.78,0.01.
I wanted to check the values, so on the right I changed the colorspace to Lab. The Lab values I was given were 76.7422,-0.9844,-5.5391. I then changed the colorspace on the left to Lab and the intent automatically changed to D50. I saw that the values on the left and right were identical (so now I'm converting from Lab since Lab is chosen on the left). On the right, I chose CMYK. Guess what happened? Automatically the intent on the left changed to Absolute Colorimetric Intent. The intent I've said for years is the best until recently when I've basically given up and saw that the only intent that would be used in Adobe InDesign or Quark to convert Lab (such as PANTONE Lab) to CMYK is Relative Colorimetric Intent. So the intent to get the best reproduction of a Lab color is not being used in the programs, and unless something changes, the best intent will still not be used for PANTONE Goe colors to get the closest reproduction for those colors. Although I've tested and at time had better reproduction using Relative Colorimetric or Perceptual, Absolute Colorimetric gives me the closest reproduction on the most colors in my testing. And Apple's ColorSync Utility Calculator defaults to this intent when Lab is chosen on the left and CMYK on the right. So let's see what CMYK values I get when using the above Lab values on the left, (the default) Absolute Colorimetric Intent, and on the right choosing CMYK, choosing GRACoL2006_Coated1v2 (and leaving Relative on the right although I wonder which is best to use and why changing it makes a difference). I get CMYK .2601,.1557,.1331,.0000. That translates (after moving the decimal two places to the right) to CMYK 26.01,15.57,13.31,0. Choosing Perceptual instead of Relative on the right makes the color 2-3% lighter in each separation. Still a long way from what I would have gotten on press printing the CMYK values I was given on GRACoL2006_Coated1v2. But which is correct? Really I don't know since I don't know the Lab values of the PANTONE TPX colors, and I can only work with what I have.
So we see that printing CMYK 37.25,21.57,17.25,3.92 on GRACoL2006_Coated1v2 is a lot different than printing 26.01,15.57,13.31,0 on GRACoL2006_Coated1v2, but unless we know the Lab values we're aiming for, it's all just an exercise. And since my customer tagged the images with a profile in Photoshop but didn't tag the Illustrator document, I can't say that the CMYK values I received were made for SWOP, Photoshop 4 Default CMYK, or what printing condition they were made for. But I can tell you this much: I will most likely use this ColorSync Utility Calculator for all my color matching in the future (if I can remember), because I like that Apple knows that Absolute Colorimetric Intent needs to be used to get the closest reproduction in CMYK when coming from RGB or Lab, and that's what Apple defaults to.
When we get the PANTONE Goe colors in our apps, how is Adobe and Quark (and Microsoft?) going to deal with them? Will they let Relative Colorimetric Intent be the default for all color conversions, or actually use different rendering intents depending on the subject to get the closest reproduction? Apple knows that in most cases in converting to CMYK, Absolute Colorimetric Intent is the best theoretically, so it uses it. Because we don't want paper in the source to become actual printing dots in the destination (to see this in ColorSync Utility Calculator, make left and right to both be CMYK, make values on the left CMYK 0,0,0,0, and on the left choose Absolute Colorimetric. You'll see that the values on the left don't stay 0, depending on what profile is chosen, and whether coming from RGB on left or Lab on left, values on right do not stay 0), Relative Colorimetric must be used for stuff that is already CMYK (if color managed at all), but is not the best to use when trying to get the closest reproduction on a color (but we must use it because it does put dots on paper where there should only be paper).
Even PANTONE said in the Goe White Paper that ICC color management had it's problems and that the vendors would have to come up with a way to get the closest reproduction of the Goe colors. What's going to be done to fix this situation and let us be able to as close as possible reproduce the Goe colors (or any colors as long as we know the Lab value)?
Don
