Re: Color Management in Reverse?
Re: Color Management in Reverse?
- Subject: Re: Color Management in Reverse?
- From: Marc Levine <email@hidden>
- Date: Sat, 03 Jul 2004 10:50:36 -0400
>
Such as CMYK to PCS Source and then CMYK to PCS
>
> Destination
Let me try to clarify -
All profiles transformations (including those that happen in RIPs) use two
profiles which are commonly called source/destinstion or input/ouput. In the
BestRIP, they use the terms "reference" & "paper".
The first profile of any transform is essentially a color description of the
data you are feeding it. For example, when you select a "reference" profile
in Best, you are defining how the RIP will interpret the color values of the
image files that you intend to print. In Photoshop, when you assign a
profile, you are telling the system what the correlating color values are
for the RGBs or CMYKs that you have opened on the screen.
The second profile always describes how to convert the color values (from
the first part of the transform) into new device colorspace (RGB, CMYK,
etc). The second profile really contains most of the "trickery" as
converting Lab to a device space such as CMYK requires additional logic to
control the balance of black and CMY and has all sorts of implications
regarding the reproduction of detail, gradient smoothness, saturated color
rendering, hue compression, so on and so forth. That's why, when you select
"convert to profile" in Photoshop, the application actually has to do some
work. The same procedure happens in most RIPs when processing a file.
OK. Some RIPs (such as Imation, Fiery, Splash) are designed to use their own
"trickery" in the output profile. To do this, what manufacturers can do is
use the "forward" side of the output profile (usually the "inverse" in
used), extract the color correlation info, and REBUILD THEIR OWN OUTPUT
PROFILE. In such a case, a RIP will basically ignore the manufacturer
inverse table. That's why - when you make edits to that table - they will
not be "seen" by the RIP. However, if edits ARE made to the "forward" side
of the profile, they will be seen when the RIP rebuilds the forward data
into the new profile.
If that's not confusing enough, Engine's such as Imation's CFM expect color
to "round trip" through a profile. What this means is that, if the engine
sees an edit in the forward table, it assumes that the "negative" version of
the edit should go into the rebuilt profile. As a result, you would need to
make the opposite edit to the output profile forward that you want to affect
on the print.
Anytime you make as edit to a profile that doesn't get "seen" by a RIP, the
issue is that either the wrong table was edited (perceptual vs.
colorimetric), or the RIP is not using the part of the profile that you
think it is.
All profile transforms always use Device->PCS->Device logic. It's just that
sometime, the PCS->Device logic can be built using the RIP using the
"Device->PCS" part of the output profile. Confusing?
I would bet dollars to doughnuts (where the heck did that term come from?)
that, if you have any visibility issues with edits in the Best RIP, that you
are using Imation's version of the software.
Hope this helps.
Marc
--
Marc Levine
Sales Guy
Technical Guy
X-Rite Incorporated
Email email@hidden
www.xrite.com
_______________________________________________
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.