Re: Scanmate 11000 / Color Quartet and ICC profiles
Re: Scanmate 11000 / Color Quartet and ICC profiles
- Subject: Re: Scanmate 11000 / Color Quartet and ICC profiles
- From: "Martin [apple account]" <email@hidden>
- Date: Wed, 16 May 2001 11:11:27 +0100
Josef
Could you re-cap on the version of CQ that you are using with your scanner?
Significant changes have been made recently -- the first versions of CQ 5
would not work properly with the updated ICC profiles that were supplied
with OS 9.1.
CQ 4.4.1 and previous versions did not use ColorSync for conversions from
RGB to Lab. CQ 5 uses the profiles specified in the ColorSync control panel.
It is also worth noting that CQ and the Scanview scanners were never
developed as wide gamut colour capture devices. It was anticipated that most
users were scanning for pre-press CMYK work and that a minority would be
working in 3 colour or 3 channel modes. So the software is a bit of a pain
for users like us that like to suck every bit of data out of a tranny or
negative...
Next you have to question why you need to use external profiling with your
scanner?
The internal IT8 calibration that is built into CQ is excellent and I
believe that it produces a "profile" that is much larger than a bog standard
ICC profile.
Surely the only reason to use a profile produced by another source would be
if it could produce an even larger look-up table?
We have produced scanner profiles using ColorBlind, but cannot see any
differences on our test scans -- visually or in the numbers.
I seem to recall that your main problem was a consistent red cast in
Photoshop that did not appear on your monitor in CQ.
We never get a red cast on tranny or print -- so perhaps your Photoshop
colour management is at fault here?
That is not to say that we never get red casts! CQ still has a bug when
using magnetic transparency mounts -- the images are always red when opened
in Photoshop. I also believe that there was an additional bug that would
cause a red cast if you didn't have virtual memory switched on on a
Macintosh.
We suffer the same problems as you when scanning negatives. The image on
screen is never as red as the result in Photoshop, no matter whether you
colour manage or not. There is also a very suspicious red-shift when you
view the curves on almost all of their supplied negative profiles - the red
curve is always separated from the green and blue.
We have to use a combination of colour numbers and intuition to get the
scans right -- which is a real pain in the arse.
The preview problem could be down to CQs internal workings. Normal positive
previews are Lab files, while negative previews are RGB files -- why this is
I don't know, but I found out by checking the preview files in Photoshop and
in BBEdit. When the data is cropped from the preview and inverted with the
curves set in the negative developer, the RGB data is then positive and it
is converted to Lab with the ColorSync generic RGB profile. They are then
treated the rest of the way through CQ as if they were positive images.
This doesn't seem very efficient to me -- and is the most likely cause of
the colour shift between approved screen image and the final data opened in
Photoshop.
Better stop my criticism now... might begin to sound like I don't love the
software and the scanners -- it's still the best.
--
Martin
Idea Digital Imaging Ltd -- the "image" specialists.
http://www.idea-digital.com