Dan Margulis Applied Color Theory

Why Isn’t the Total Ink Limit Honored?


   Date: Mon, 6 Dec 2004 15:38:39 -0700
   From: Ron Kelly
Subject: CMYK conversions

Hi Group:

I have been making CMYK conversions from RGB colorspace for a long time now, and mostly with very good results.

Today, when I changed my custom CMYK setup to some new settings it occurred to me to ask some questions:

1. Whenever I specify , for example, 340 % total ink, I usually don't get it but instead I get about 330, or sometimes 315. Why?

2. I specify a black ink limit of 80 % with light GCR. Why does it always come in around 60%?

3. My neutral in the shadows I prefer to be 92-82-82-82. I often get 90-82-87-60., ie the yellow is a bit high. I typically leave this because it doesn't seem to cause any problems, and if I correct it with curves I often undo that correction because I like the result less.

FWIW I don't really have a problem with this situation because I always tweak my images and check the highs, lows, etc.. There's never been a file that's passed through untouched so it's no big deal to place the curves exactly where you want them. However, today it occurved to me: why aren't they already there in the first place?

Maybe photoshop's internal math engine has a little voodoo side that can't be totally suppressed, kinda like Ron Artest.

Just wondering,
Ron Kelly
___________________________________________________________________________

   Date: Mon, 06 Dec 2004 15:52:20 -0700
   From: Andrew Rodney
Subject: Re: CMYK conversions

On 12/6/04 3:38 PM, "Ron Kelly"  wrote:

1. Whenever I specify , for example, 340 % total ink, I usually don't
get it but instead I get about 330, or sometimes 315. Why?

Because the classic CMYK engine is broken.

2. I specify a black ink limit of 80 % with light GCR. Why does it
always come in around 60%?

Because the classic CMYK engine is broken.

Andrew Rodney
http://digitaldog.net/
___________________________________________________________________________

   Date: Mon, 6 Dec 2004 17:35:21 -0700
   From: Ron Kelly
Subject: Re: CMYK conversions

Andrew:

Don't just stand there; hand me the booster cables! You mean to tell me this thing miled out already?

RK
___________________________________________________________________________

   Date: Tue, 07 Dec 2004 02:51:13 -0000
   From: "Ernst Vegt"
Subject: Re: CMYK conversions

Ron, could it be that the RGB file does not measure 0,0,0 in the shadows? If it does, then you should be getting that, but then most of us doing images for print don't go that far as we risk loosing shadow detail.

By going with 10,s or 5,s in the deepest neutral shadows, we of course get less when we convert to CMYK for whatever printing condition, but that may not be such a bad thing, as long as saturated colours are as they should be, etc.

Regards,

Ernst Vegt
___________________________________________________________________________

   Date: Mon, 06 Dec 2004 18:51:17 -0700
   From: Andrew Rodney
Subject: Re: CMYK conversions

On 12/6/04 5:35 PM, "Ron Kelly" wrote:

Don't just stand there; hand me the booster cables! You mean to tell me
this thing miled out already?

Long time ago.

Andrew Rodney
http://digitaldog.net/
___________________________________________________________________________

   Date: Tue, 07 Dec 2004 02:41:33 -0500
   From: Dan Margulis
Subject: Re: CMYK conversions

Ron Kelly writes,
 
1. Whenever I specify , for example, 340 % total ink, I usually don't
get it but instead I get about 330, or sometimes 315. Why?
 
If you get 315 when your field says 340, it's because 340 is more than the maximum mandated by the type of black you chose. The "Ink Limit" field is not a command to *produce* a certain value, it is an absolute limit beyond which Photoshop may not go, but if the character of the separation precludes ever getting to that limit, the setting is never referenced.

The obvious example would be if you specify Maximum GCR, 100% Maximum black. You are then saying that you want the deepest shadow to be 0c0m0y100k. Under those circumstances the Ink Limit field is irrelevant. Similarly, if you pick Medium GCR, 100% Maximum black, then your effective total ink limit is around 260%, and any higher "Ink Limit" setting is irrelevant.

Departing from the specified shape of the separation by adding more CMY to the shadow is technically a bad idea. The only exception is when the job is being printed gravure. In that case, we use UCA to get the desired result.

 2. I specify a black ink limit of 80 % with light GCR. Why does it
 always come in around 60%?

Forgetting dot gain, which also has an effect, there are three major settings in Custom CMYK or any other sensible separation algorithm.

1) The desired shape of the black in relation to the CMY, aka the GCR setting.

2) The Total Ink setting.

3) The Maximum Black setting.

Unless you are determined to use UCA, which is ordinarily a bad idea, you can only get what you want in two of the above. The first is critical to reproduction, so it becomes a question of whether you forget #2 or #3. In making this decision, Photoshop goes for the one that produces the darker shadow. If the Ink Limit setting is low, that usually means that it will go for a full maximum black and not reach the ink limit. If the Ink Limit (as in your case) is rather high, Photoshop will let the CMY reach their natural values in the 90s, but not allow the black to reach its theoretical limit.
 
 3. My neutral in the shadows I prefer to be 92-82-82-82. I often get
 90-82-87-60., ie the yellow is a bit high. I typically leave this
 because it doesn't seem to cause any problems, and if I correct it with
 curves I often undo that correction because I like the result less.
 
If you dislike this you can fix it by editing the magenta ink to have less positive AB values, or by increasing its dot gain setting in the quartertones and decreasing it in the shadows. However, I don't think its worth it. I only teach that magenta and yellow should be equal in the shadows for convenience. Technically, the yellow should be a couple of points higher because it has less dot gain. Not quite so much higher as what you are reporting, but I don't think you have to worry.

Dan Margulis
___________________________________________________________________________

Adobe Photoshop training classes are taught in the US by Sterling Ledet & Associates, Inc.