[Openicc] Wrong color output of Xerox printer - Color Management problem?

[Michael Vrhel]

Hello Gerhard,

Ghostscript is already converting the postscript CIEDEF(G), CIEABC and CIEA 
based color spaces to equivalent ICC profiles without any issues for the 
Postscript quality logic CET tests.    Conversion of the CRD will not be a 
problem (just a simple matter of programming).

Understand that the the CRD is a mapping from CIEXYZ to device space.   The 
ICC profile that I create will also be a mapping from CIEXZY to device 
space.   Just as the CRD is mashed together with the source color space to 
create a mapping, the ICC profile that I create will be mashed together with 
the source ICC profiles to create a mapping which will be applied 
appropriately.

Kind Regards,

Michael

-----Original Message----- 
From: Gerhard Fuernkranz
Sent: Tuesday, November 20, 2012 10:43 AM
To: Michael Vrhel ; Open ICC Color Managment
Subject: Re: [Openicc] Wrong color output of Xerox printer - Color 
Management problem?


Thanks Michael!

Well, the m2300w print filter accepts PS or PDF input an delegates most of 
the rendering functionality (including halftoning and color management) to 
ghostscript. And since only traditional PostScript color management was 
available in gs 7.x and 8.x, a CRD was the only option to describe the 
printer color space...

I guess it won't be too difficult to convert a {source color space, CRD} 
pair to a device link.

However, it may not be generally possible to convert a CRD to a printer 
profile, because the color transformation defined by a CRD can depend on the 
white and black points of the individual _source_ color space which happens 
to be combined with the CRD. Furthermore, a CRD defines only the color 
transformation for a single rendering intent.

Best Regards,
Gerhard


Am 20.11.2012 18:21, schrieb Michael Vrhel:
> Gerhard,
>
> There is an open bug to address this 
> (http://bugs.ghostscript.com/show_bug.cgi?id=691749).  So few people do 
> CRDs these days that it has been a very low priority. You get the 
> distinction of being the first person to notice the issue.   Keeping gs 
> compliant with all aspects of the Postscript specification is important. 
> I will escalate this and hopefully it will be fixed prior to the next 
> release.
>
> Michael
> ________________________
> Michael Vrhel, Ph.D.
> Artifex Software, Inc.
> 7 Mt. Lassen Drive A-134
> San Rafael, CA 94903
> tel: +1 425 894 0585
> www.artifex.com
>
>
>
> -----Original Message----- From: Gerhard Fuernkranz
> Sent: Tuesday, November 20, 2012 3:18 AM
> To: Open ICC Color Managment
> Subject: Re: [Openicc] Wrong color output of Xerox printer - Color 
> Management problem?
>
>
> I also noticed color issues with the Minolta Magicolor 2300W, which seem 
> to be caused by the fact that the new ICC-color-managed gs 9.x does 
> obviously no longer support CRDs (CRDs are however a well-defined 
> PostScript functionality - i.e. the new gs is no longer fully PostScript 
> L2/L3 compliant wrt. this functionaily).
>
> Regards,
> Gerhard
>
>
> Am 20.11.2012 11:37, schrieb Till Kamppeter:
>> Hi,
>>
>> after a certain update of Ubuntu some color printers from Xerox do not 
>> print colors correctly any more, see
>>
>> https://bugs.launchpad.net/bugs/984424
>>
>> What changed during the development of Ubuntu is the color-managed 
>> rendering by Ghostscript and also the color management support in the 
>> surrounding CUPS filters.
>>
>> This can have negative effects on some color printers, due to their own 
>> way to do color management, requirement of printer-specific ICC when 
>> sending color-managed job, ...
>>
>> Can someone have a look and comment on the bug report? Please also feel 
>> free to ask the reporters of the bug for additional information, sample 
>> files, ...
>>
>> Thanks in advance.
>>
>>    Till
>> _______________________________________________
>> openicc mailing list
>> openicc at lists.freedesktop.org
>> http://lists.freedesktop.org/mailman/listinfo/openicc
>>
>
> _______________________________________________
> openicc mailing list
> openicc at lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/openicc
> _______________________________________________
> openicc mailing list
> openicc at lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/openicc
>