Re: Artisan calibration/soft proof workflow
Re: Artisan calibration/soft proof workflow
- Subject: Re: Artisan calibration/soft proof workflow
- From: "Tony Riley" <email@hidden>
- Date: Mon, 5 Jan 2004 12:21:11 -0000
On 1/4/04, Will and Pam wrote
>
We are implementing new Sony Artisan monitors in an art museum scanning
>
project and have some questions about how to proceed. We've gleaned what
we
>
can from various sources, including posts to this list, books, on-line
>
resources, etc. and perhaps we've missed the answers to some of these
>
questions--if so, we apologize!
>
>
We're scanning primarily original photographs (black-and-white, cyanotype,
>
albumen prints, etc.), and 4x5 color transparencies depicting works of art
>
such as paintings, prints, drawings, etc. We are creating generic masters
>
for archival purposes but are also deriving files from the masters for
both
>
web delivery and print output. Our two output target devices so far are
>
Epson ink jet (1280) and, at a service bureau, a Durst Epsilon printer. We
>
have generic profiles for both.
>
>
Is there anything else not mentioned above that we ought to consider?
Will and Pam - There has been a large response to your query by members far
more expert than I, but surely a relevant fundamental issue not considered
(which I appreciate you didn't ask about) is the colour accuracy of your
scanning device with regard to copying artwork (paint pigment subjects). If
the original data worked with is not too accurate caused by the capture
device's response, reproduction problems are compounded from the start. I
have raised this point previously on the list and ended up with
diametrically opposed opinion from experts. Query has to cease when strong
expert statements are given without reference, which was the case last time!
Research done by the Munsell Color Science Laboratory at the National Museum
Washington compared digital reproduction from line scanner, 2D area array
scanner (camera), scanned 10x8 ektachrome and a specialist IBM copy system
which matched the ccd filter response to cone pigments. This 2000 report is
quite critical of museums and galleries methods of imaging artwork -
http://www.cis.rit.edu/research/mcsl/research/PDFs/Berns_art_digitize_lowres.pdf .
Line scan camera backs perform poorly in this comparison study, resulting
"in very poor color accuracy".
I dare say you are stuck with your system, but scientific approaches (e.g.
filter wheel with monochromatic camera, filter IR out of the light source)
have given remarkable accuracy, and don't have to be an expensive project in
a research lab. I am interested to know if you have found that some
paintings (artists pigments) reproduce more accurately than others with the
same system.
Finally I would like to agree with a most important point made by CD Tobie
in a later post -
>
From: email@hidden Date: Sun, 4 Jan 2004 17:52:08 EST
>
Subject: Re: Artisan calibration/soft proof workflow
>
Delta-Es mean suprisingly little in terms of user satisfaction with the
final
>
image. Custom profiles may make your results far more satisfying, even if
>
they don't decrease your delta-E range significantly.
There is now a lot of research on colour appearance and user acceptance of
accuracy, one of the strongest factors being the surround to the subject
being viewed. Even though I know otherwise, I still find myself tending to
assume that subjects have colour rather than producing the sensation myself!
Wouldn't it be nice if capture device manufacturers told us how
colour-accurate they were. It's the one specification they never mention.
Tony Riley
_______________________________________________
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.