[argyllcms] Re: Devicelink Black Generation
- From: "Fabrizio Levati" <fabrizio.levati@xxxxxxxxxxxxxx>
- To: <argyllcms@xxxxxxxxxxxxx>
- Date: Wed, 16 Apr 2008 11:00:17 +0200
Hi Graeme,
first of all I would like to thank you for what you are doing with ArgyllCMS.
I agree with you pointing out that what I described seems to be cross purposed.
Maybe my mistake was using the term repurposing...
Anyway, to make things clearer, what I tried to describe is the typical
scenario that you will find in many printing plants. Usually the printer
receive files prepared for a specific printing process (for example webcoated)
and has to print a completely different kind of product (for example a low cost
advertising magazine on standard newsprint paper-press). The first answer to
this kind of needs is to use a devicelink profile to perform a repurposing.
This will assure the best results in term of colour accuracy but will lead to
different problems. For example, not preserving pure BLACK text will result in
mis-registration problems during the print run. For the primaries, the problem
is similar because the customer prefer a slightly different colour in a 0C 100M
100Y 0K text instead of seeing an halftoned text where it wasn't supposed to be.
Described with my poor English, this is what I was talking about.
Any thought or suggestion is welcome.
Fabrizio
----- Original Message -----
From: Graeme Gill
To: argyllcms@xxxxxxxxxxxxx
Sent: Tuesday, August 12, 2008 8:54 AM
Subject: [argyllcms] Re: Devicelink Black Generation
Fabrizio Levati wrote:
> I am trying to create a devicelink profile to convert from ISOwebcoated to
> ISOnewspaper (but could be any other kind of repurposing from one printing
> press/process to another) and I am not able to tweak the black generation
> without loosing the purity of the primaries.
It seems to me you are at cross purposes here. The whole point about
repurposing is to maintain the color, which if the two devices have
different colorants, means using different primary combination values
to represent corresponding colors, thereby resulting in non pure primaries.
> Besides the CMY purity, what I am trying to achieve is to create a
> devicelink with my own GCR and preserve the pure (100%) BLACK in the text.
Once again you seem to be at cross purposes. You can't impose a GCR and
also translate the incoming black ratio through to the output, since
these aims contradict each other.
> Considering that even the kt option result in a ³contaminated² black text,
> I am now wondering if and how this is possible.
The reason that K only in doesn't produce K only out with -kt
for the type of profiles you are using, is that color
accuracy is given a higher priority than black inking,
and the source profile has a lot blacker K only black
than the destination, so CMY is added to reduce
the error.
Note that collink has a -f option, that forces neutral input colors
to be K only output, but this probably doesn't do what you want,
since it will force CMY0 to K only too.
It's not too hard for me to add another option to collink
to force K only output for K only input. If I do this:
collink -v -qm -kt -ims -fk offset.icm newspaper.icm test.icm
I get this sort of result:
icclu test.icm
0.000000 0.000000 0.000000 0.000000 [CMYK] -> Lut -> 0.000000 0.000000
0.000000 0.008345 [CMYK]
1.000000 0.000000 0.000000 0.000000 [CMYK] -> Lut -> 1.000000 0.000000
0.000000 0.015225 [CMYK]
0.000000 1.000000 0.000000 0.000000 [CMYK] -> Lut -> 0.000000 1.000000
0.003097 0.000000 [CMYK]
0.000000 0.000000 1.000000 0.000000 [CMYK] -> Lut -> 0.000000 0.000000
1.000000 0.002731 [CMYK]
0.000000 0.000000 0.000000 1.000000 [CMYK] -> Lut -> 0.043310 0.053822
0.000000 1.000000 [CMYK]
1.000000 1.000000 1.000000 0.000000 [CMYK] -> Lut -> 0.674325 1.000000
0.000000 0.767079 [CMYK]
I'll include this new option in the V1.0.2 release.
Graeme Gill.
Other related posts:
