RE: Of colorful scepticism
RE: Of colorful scepticism
- Subject: RE: Of colorful scepticism
- From: Scott Olswold <email@hidden>
- Date: Mon, 1 Oct 2001 17:13:58 -0400
John,
Thank you.
Igor,
I think the point you are missing is that a profile of a digital camera is,
at its most simple, a description of how the CCD is behaving based on a
known target (whether that be IT.8 or something known that you've made up).
Now, a profile for a digital camera is a lot different than a profile for a
scanner...or is it? (Rhetorical question--sorry). The truth is, its not. Say
what you want about indoor lighting vs. outdoor, different ambient
conditions, etc. Those are things that photographers have been dealing with
for decades and they have learned to adapt to or mold the situation so that
they get a good picture. That means that transitioning from film-back to
digital-back to them means only one thing: different media. So how are the
two (scanners vs. cameras) similar? When I profile a scanner, the target is
reflected against a relatively known quantity and quality of light (the
scanning wand). When I profile a digital camera, the target is shot using a
relatively known quantity and quality of light because I profile in (as
close as I can get anyway) controlled conditions.
Now I know how my CCD (whether embedded in my scanner or on my camera)
interacts with my target.
My scanner is used to scan outdoor photography, indoor photography, still
life, action, high-key, low-key, all sorts of different types of pictures.
My camera is used indoors, outdoors, with motion, without motion, bright
light, low light, it takes all sorts of different types of pictures. Once I
get them into Photoshop and apply my profile, I know that the RGB data are
accurate.
I believe (in fact, am a victim of it--I'm B-G color deficient in my right
eye) your statement on the subjectivity of color between observers. But like
others (like John) have said, even though this is a statement in truth,
using it as a basis for no action is detrimental at best. The goal of color
management is predictable color output. If the observer notices no real
difference between steps A, B, and C then color management (and by
extension, our profiling) has done its job. It isn't my fault if he or she
doesn't see the apple printout as red if the real thing didn't start out for
them as red.
Scott Olswold
Senior Systems Support Engineer
Danka Office Imaging
MCSE, CNA 5, A+, Network+, Adobe Certified Expert (PageMaker and Photoshop)
For the latest drivers and utility downloads, visit
http://www.digitaldanka.com