Path: utzoo!utgpu!watmath!clyde!att!osu-cis!tut.cis.ohio-state.edu!husc6!bu-cs!mirror!frog!wang7!klm
From: klm@wang7.UUCP (klm)
Newsgroups: comp.sys.amiga
Subject: Re: Paint Jet Printers
Summary: I'm still here
Message-ID: <2629@wang7.UUCP>
Date: 4 Jan 89 03:36:00 GMT
References: <158@pande.UUCP>
Distribution: na
Organization: Wang Laboratories, Lowell, MA, USA
Lines: 66

In article <158@pande.UUCP>, richard@pande.UUCP (richard) writes:
> 
> Although, for all colour printers, it seems a correction has to be made.

Yup.
 
> If Kevin McBride (the guy in the cast) still had net.access, he would pop
> in here and point out that only Eikonix doesnt suffer from this syndrome.

NYAH NYAH NYAH NYAH!!!  (see below)

Actually, Richard, I'm flattered.  It's nice to know that somebody
remembers me.
 
> The conversion of absolute RGB, or screen colours, to printer colours
> is one that is left up to the user in the case of the Amiga system software,
> but is handled by the system software in the case of Eikonix. Hint, hint,
> guys.

Hey, I'm still here.  I'm just being uncharacteristically quiet.  I'm also
at a different address from where I used to be.  My current employer gives
me assignments that I ENJOY (for the most part.  Contrast that with DREADING
going to work every day.)  Also, no more cast.  My ankle is much better.
My left shoulder however,....

Eikonix (my former employer) does suffer greatly from this problem.  In fact,
we all suffered so greatly that very few of us are left.  We were sacrificed
so that the customer would have great f---king color. 

Sc-mb-gs. :-)  But, enough about corporate politics. :-)

What we have is a color science problem.  It is not trivial.  Yes, we did
it "correctly" at Eikonix, but we used 3 array processors and a PDP-11 to
acheive our results.  Not very practical for "affordable solutions" on low
cost equipment, eh?

If you're feeling really masochistic, pick up a copy of "Color Science:
Concepts and Methods, Quantitative Data and Formulae" Second Edition by
Gunter Wyszecki and W. S. Stiles.  Copyright 1982 John Wiley & Sons, Inc.

Keep in mind that all of this discussion about "Orange Reds" vs. "Red Reds"
is extremely subjective.  Color Science attempts to add a certain amount
of objectivity to the human perception of different wavelengths of light.
Red is Red and Blue is Blue simply because we say it is.  If that isn't
subjective, I don't know what is.

In summary, to get from the screen to the printer, you want to accurately
map an additive color space (RGB) to a subtractive color space (YMCK)
You need to know the characteristics of the phosphors in your CRT and you
need to know the characteristics of the ink AND paper in your printer.  In
a graphic arts shop, they even know the characteristics of the light source
they use to view proofs!  Finally, you need a whole hell of a lot of compute
horsepower to get from point A to point B.  IF you do it right, you should
be able to view soft copy and hardcopy side-by-side, under controlled
conditions, and not see any difference in color.

I'd show you part of the formula involved, but it's a trade secret. :-)
Seriously though, if you really want to know, read the book.

Anyway, I'm here, but I'm busy and have a flakey news feed.

-- 
Kevin McBride         Another  //     | A cure for AIDS has been found.
Wang Laboratories      Proud  //      | They've given it a Yamaha part
Desktop Image Systems     \\ // Amiga | number so now noone can get it.
...!ima!wang7!klm          \X/  owner |       - Motorcyclist Magazine