A slightly abstract way of thinking about this may help. Think in CIELab
terms. The program Roy wrote profiles only the L* axis - luminance. It
doesn't care about a or b channels and in fact thinks they don't exist. Of
course the printers produce non-zero a and b images - there aren't perfectly
neutral inks and paper has a hue anyway - and we can control these values a
number of ways, eg by our ink choices in QTR. A profiler such as PM 5 always
thinks in full colour terms and profiles the results generated from patches
of RGB or CMYK input in terms of both luminance and hue. Here we just have
luminance input and measure luminance/reflectance generated. It is hue
agnostic and we manage hue not with the profile but manually by ink
selection/sliders or whatever (or not at all in the case of, say, black
only).
The closest application to this, that I am aware of, is ColorshopX's
Grayscale Builder. However this program is a pain to use as it can't take
readings as input and the only way to change the shape of the profile tonal
response curve is via a transfer function which is not that intuitive.
Roy's program is a little wonder because it can read actual patches produced
from your particular printer with your particular ink and workflow.
So, once again, you are profiling luminance only ("contrast" shall we say)
and you are not profiling "colour" in the way PM5 or EyeOne Match etc do.
The profile allows you to use CM to manage the tonal compression from file
to print. Think of it as being half way between No Colour Management (Same
as Source) and full blown colour management in the colour printing sense.
Let's say you like the look of black only printing. One of the problems
with black only printing is that the greyscale produced by this workflow has
very uneven luminance change from blackest to whitest. You have to mess
with curves to get a progression that makes any sense and it is not
intuitive how the curves should look when looking at your original step
wedge. You could use Roy's program to simply profile the luminance axis of
the black only step wedge (no curves), ie to profile the luminance of the
print space. Then you could soft proof with this profile and convert your
image from your workspace to this print profile when printing. You will
likely like the tonal management out-of-the-box and if not then the
adjustments will be less severe and can be soft proofed as you do them.
> From: John Moody <moodymz3@yahoo.com>
> Reply-To: <DigitalBlackandWhiteThePrint@yahoogroups.com>
> Date: Tue, 12 Jul 2005 10:14:07 -0400
> To: DigitalBlackandWhiteThePrint
> <DigitalBlackandWhiteThePrint@yahoogroups.com>
> Subject: [Digital BW] QTR icc generation
>
> If someone already has i1 photo for profile creation does this new feature
> produce icc profiles any differently, other than being grayscale only?
>
> I thought I recalled Steve mention that this new feature would do something
> that even ProfileMaker 5 can¹t do. I don¹t quite understand what that would
> be.
>
> John
>Message
Re: [Digital BW] QTR icc generation
2005-07-12 by Steve Kale
Attachments
- No local attachments were found for this message.