Kai-Uwe Behrmann wrote:
In the results, while a linear colour space like ISO coated (ECI) shows a
pure CMY black very dark and neutral, my physical proofer
(EpsonStylus3000) shows there a very bad cast. And with backpreservation
turned on this becomes a huge color shift in the ISO coated ->
EpsonStylus3000 transformation.
What are the experiences of others?
At one stage at Colorbus we played about with preserving black to
proofing inkjet output, and in some ways, it had a good effect
- original black only colors had robust neutrality, because they
had more black in them, and I think the illuminant independence
was better, since the ink makeup of the proof better matched the original,
meaning that there was a better spectral match between original and
proof.
After a while, we decided not take this approach though. The black level
in the inkjet had too many other effects to leave it up to the characteristics
of the input file to determine. If the black level is too low, the color
neutrality is not robust enough under different lighting, and the
bronzing effects of pigment ink are worse. Too high a level of black
and the black dots show up, and any banding in the black channel becomes
very prominent. We decided that in reality, for an inkjet, we had to control
the level of black in the output fairly carefully, to get a good
overall result.
(Of course if your "preserve black" isn't being done colorimetrically,
then the results might be quite poor in other ways. I implemented a
"straight through black" hack in the DiceNet/Colorbus device link
linker even earlier, but this was only used for business documents
to preserve black only text. You got some funny effects on images in
CMYK documents at times !)
Graeme Gill.
-------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc. Do you grep through log files
for problems? Stop! Download the new AJAX search engine that makes
searching your log files as easy as surfing the web. DOWNLOAD SPLUNK!
http://ads.osdn.com/?ad_id=7637&alloc_id=16865&op=click
_______________________________________________
Lcms-user mailing list
Lcms-user@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/lcms-user
