[cairo] Spot colors (and CMYK)

[Chris Murphy]

On Feb 20, 2010, at 2:26 PM, ecir hana wrote:

> On Sat, Feb 20, 2010 at 9:17 PM, Chris Murphy <lists at colorremedies.com> wrote:
>> 
>> On Feb 20, 2010, at 4:19 AM, ecir hana wrote:
>> 
>> Both RGB and CMYK and spot values are meaningless without being tagged. You may specify CMYK values directly, rather than needing a CMS to produce them, however those CMYK values should be tagged with a profile so that their color meaning is usable downstream: for display, for proofing, and for preflight purposes to confirm the proper printing conditions are applied.
> 
> Ok, lets see what would happen with tagged and untagged workflows:
> 
> Tagged - I have CMYK SWOP 0, 0, 0, 100. If I print on coated paper in
> US, no big deal. If I print in Europe on uncoated paper, they just
> convert from SWOP to FOGRA (?). Now, how does my CMYK look like?
> Definitely not 0, 0, 0, 100 - there would be no zeros anymore and no
> 100, i.e. no solid, just 4 rasters - *this* is something I don't want.
> But then you say, that's because you didn't use device link - sure
> thing, but where can I get it?
> 
> Untagged - 0, 0, 0, 100. I print in US - 0, 0, 0, 100. I print in
> Europe - 0, 0, 0, 100.
> 
> What's so difficult to understand about this?

I understand the problem. I'm disagreeing with your solution. It is archaic, and obsolete. The way this is handled in PDF/X-3/4/5 is anything that would need to be repurposed is ICCBased, and items like black text are /devicecmyk or /devicegray *however* all device dependent "untagged" data is implicitly tagged with an output intent profile for the intended printing process. Since source=destination, there is no conversion for such objects when printing to press, only when displayed and proofed. Conversion is required, even for 100K only text, or it will not display or proof correctly.


> 
>> Metadata describing "black text" rather than "this generic object is C0M0Y0K100" is superior. Because how black text should be handled is different depending on the output process. For a printing press, it would need to be 100K only. But for a billboard, you'd print rich black because 100K is often a dark gray, and its perfect register makes rich black text feasible. Leaving devicedependent data floating around in the world is a pretty bad idea. Even for printing to known destinations, the PDF/X-1a model should be chosen, which is to tag all CMYK content.
> 
> 1. this 'metadata describing "black text"' is new to me, please, do
> you have more info? Is it in PDF specs?

It has been proposed. And some applications do treat black text in a unique matter internally, before they produce a PDF in order to ensure desired initial text handling.

> 
> 2. If they would print "my black" as CMYK 0, 0, 0, 100 first time, and
> as CMYK 40, 30, 30, 100 second time, I would ..... not like them.

If it looks correct and registers correctly, WTF do you care? This is a prepress and printing concern, it is not a design aesthetic concern.

> 
> 3. PDF/X-1a is the best thing since sliced bread. Really, if Cairo
> could output PDF/X-1a I would throw away Acrobat. PDF/X-1a has
> everything I need, it's in fact that good. (One tiny side note: I
> don't have the specs, as it is not freely available. But from what I
> read it does not require profiles for spot colors, just for its
> fallbacks - there is just one global tagged CMYK space, defined in
> print intend. If this is truly the case, then this is just what I
> want. If it is not, I apologize for the confusion and welcome any more
> precise description).
> 
> I, for one, don't understand what's so bad about untagged content - if
> you know what you are doing, it simplifies a few things.


Your black text, in PDF/x-1a is implicitly tagged with the output intent, while remaining /devicecmyk. That's how it proofs correctly instead of being as dark or light as black only on the inkjet printer or proofing system you're using, which invariably is not the same as a press.

Completely untagged content is simply a bad idea if you want to do anything other than print it.




Chris Murphy
Color Remedies (TM)
New York, NY
----------------------------------------------------------------------
Co-author "Real World Color Management, 2nd Ed"