Colormanagement: Bugs, bugs, bugs ... :-(
Colormanagement: Bugs, bugs, bugs ... :-(
- Subject: Colormanagement: Bugs, bugs, bugs ... :-(
- From: Klaus Karcher <email@hidden>
- Date: Mon, 31 Jan 2011 12:41:31 +0100
Hi List,
I hope your weekend was better than mine. Mine was dreadful because I
stumbled from one bug to the next.
I've uploaded some test files and screencasts to document the issues,
therefore I just want to outline them briefly.
1.) Photoshop's Color Settings Panel has serious bugs. Sometimes the
panel neither reflects Photoshop's actual settings nor the
synchronization state correctly. As soon as one starts to play with
conversion options (e.g. toggles compensate for scene-referred profiles
and black point compensation or changes the rendering intent), Photoshop
behaves more or less erratic. It can happen that Photoshop ends up in a
state where neither the display nor the results are correct or where the
display is correct, but the Lab values in the info panel are wrong.
These nasty bugs should be fixed as soon as possible, the more so as
they date back at least until CS4.
2.) Some profiles seem to trigger Photoshop's
scene-referred-to-output-referred compensation by mistake.
3.) Even though the "compensate for scene-referred profiles" setting
seems to be part of the Color Settings File and hence should be
synchronized across all CS applications, it can cause inconsistent
results. Either InDesign and Acrobat simply ignore the setting or they
are not "bug-compatible" with Photoshop.
4.) The PDF/X-4 Preset in CS5 is buggy: The embedded profile is based on
FOGRA39, but the Output Condition Identifier is "FOGRA27".
5.) The profiles that trigger Photoshop's scene-referred profile
compensation also cause trouble in Apple's applications (Preview,
QuickView, ColorSync Utility). Depending on the white point of the
Monitor Profile the consequences can range from nearly invisible to
dramatic.
6.) The profiles in question are completely ICC v2 compliant (verified
with the ICC reference implementation "SampleICC-1.5.2"). The only
remarkable peculiarity is that their Media White Point slightly exceeds
Y=100.
7.) There is a rendering bug in ColorSync Utility 4.6.2 (Mac OS 10.6.6):
the 3D view shows wrong colors and transparency and does not resize
correctly.
Here are the screencasts and test files:
<http://www.screencast.com/t/I7icpG0r9>
<http://www.screencast.com/t/ETVRrCMM>
<http://www.screencast.com/t/5zJNzRdAbDB>
<http://digitalproof.info/colorsync-users/CS_testcase/test1.tif>
<http://digitalproof.info/colorsync-users/CS_testcase/test2.tif>
<http://digitalproof.info/colorsync-users/CS_testcase/test.pdf>
<http://digitalproof.info/colorsync-users/CS_testcase/p954n.L.spdb.boileddown.icc>
<http:///digitalproof.info/colorsync-users/CS_testcase/p954n.L.spdb.boileddown2.icc>
I hope that at least some of these bugs will get fixed soon. IMHO they
demonstrate clearly the disastrous situation of current Color Management
implementations: The workflows are extremely fragile, bugs lurk at every
corner and the chances to get accurate, predictable results with complex
workflows are rather small. Late binding does not even work reliably on
"Planet Adobe" -- I really don't want to know what else can go wrong
when other software (CMMs, RIPs, ColorServers, PDF workflow solutions,
printer drivers, ...) comes into play ...
Klaus Karcher
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Colorsync-users mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden