Re: Inadequate ColorSync support.
Re: Inadequate ColorSync support.
- Subject: Re: Inadequate ColorSync support.
- From: Roger Breton <email@hidden>
- Date: Tue, 18 Nov 2003 22:21:31 -0500
John,
Do you realize all the 'reverse-engineering' work you have to do to figure
out how to use the new Panther ColorSync technology at the print driver
level that Apple just implemented? You should not have to do this ugly work:
Apple should provide to the user community full documentation on how to use
these features if we ("the rest of us") are supposed to benefit from it.
Personnally, until I see a White Paper of some sort surface on Apple
ColorSync web site I am going to stay away from this "new" technology as
much as I can and continue living under the umbrella of Adobe color
architecture (which everyone know, here, does work and fully understands).
I am sure you have better things to do with your precious time.
Regards,
Roger Breton | Laval, Canada | email@hidden
http://pages.infinit.net/graxx
>
On Nov 18, 2003, at 4:22 PM, Pete Carter wrote:
>
>
> There is one exception and that is printing with
>
> preview from Photoshop with no colour adjustment in the printer driver.
>
>
My research on the issues suggests that even this is a maybe. As of
>
Jaguar at least, there is apparently no way to REALLY turn of colorsync
>
in the print driver because the colorsync happens in the pdf to raster
>
(and output profile registered by the printer) conversion which is
>
before the driver gets any of the data. The best you can do is have
>
the print driver and the application agree on the profile to use so
>
that colorsync doesn't actually do anything. This information comes
>
from @apple.com sources on the email@hidden mailing list.
>
>
Somewhere else (I forget exactly where) I read that even offering a "no
>
color adjustment" option in the driver user interface is (officially?)
>
discouraged since it only actually has that effect if the application
>
cooperates in the plan to simulate no color management. Some
>
applications just won't cooperate. The following rambling assumes the
>
above behaviors to reflect reality.
>
>
For example of an application that won't cooperate with "no color
>
adjustment", drop two versions of an image into TextEdit. Have each
>
encoded with a different profile. The PDF that gets injected into the
>
printing system has those two images with untouched raster data, each
>
tagged with the original profile. Obviously, at MOST only one of these
>
could possibly make it through to paper with "no color adjustment" with
>
the way things work now. I'd guess that we can take this as a model of
>
how an application ought to behave in Apple's Grand ColorSync
>
Integration Vision. I think there are a variety of sound arguments
>
that this a reasonable approach that wants accurate color with minimum
>
fuss--all the application has to do is maintain the association of the
>
data and profile and everything Just Works. If you use high level APIs
>
for image display, it is supposedly more or less automatic.
>
>
Another case: Photoshop. Near as I can tell, EVERYTHING that photoshop
>
puts into the printing system is tagged with the Generic RGB Profile,
>
regardless of the actual profile. Think about it for a moment and you
>
will realize that the ONLY way photoshop is going to be able to
>
correctly print an image that isn't already using the Generic RGB
>
Profile is to have photoshop convert to the output space and hope that
>
the printer registers "Generic RGB Profile" as the printer profile when
>
you select "no color adjustment". This certainly explains why my
>
"Same as source" printouts never looked right. An Adobe RGB file
>
printed with "Same as source" will have Adobe RGB data that the print
>
system will interpret as Generic RGB. (I'd really like to see if
>
Photoshop CS behaves differently.)
>
>
Most third party applications I've looked at behave more or less like
>
Photoshop...generating Generic RGB Profile tagged data regardless of
>
what the proper tag is. The result is more or less guaranteed bad
>
color.
>
>
Between these two examples, I find that TextEdit gives me a more
>
accurate print on my new Canon i960 than Photoshop does even if I use
>
Apple's CMM, so there is still some secret sauce in the driver at work.
>
Yesterday I discovered that OmniGraffle behaves like TextEdit with
>
respect to ICC tagged images.
>
>
> The
>
> really worrying thing is that nobody wants to hold their head up with a
>
> definitive answer why.
>
>
It looks to ME, a lowly amateur photographer trying to get a good
>
print, like Apple put together a Grand ColorSync Integration Vision but
>
didn't have sufficient participation of folks making printers, writing
>
drivers, writing applications that print, or people who print things.
>
It is painfully obvious that neither drivers, nor most third party
>
applications fit Apple's vision of how things should work.
>
>
It may be Apple's fault for not being inclusive enough in formulation
>
of the printing architecture, and Apple's fault for not highlighting
>
how color management in printing is DIFFERENT NOW and evangelizing
>
their Grand ColorSync Integration Vision to developers.
>
>
It may be third party's fault for not paying attention to the fact that
>
the color is all messed up in their printouts and investigating why.
>
>
-john
_______________________________________________
colorsync-users mailing list | email@hidden
Help/Unsubscribe/Archives:
http://www.lists.apple.com/mailman/listinfo/colorsync-users
Do not post admin requests to the list. They will be ignored.