Olympus-OM
[Top] [All Lists]

[OM] Re: now increasing scanner range

Subject: [OM] Re: now increasing scanner range
From: Wayne S <om4t@xxxxxxxx>
Date: Thu, 14 Oct 2004 09:31:45 -0400
At 05:30 AM 10/14/2004, you wrote:

>This sounds like it could be exactly what I'm looking for, combined 
>with the help from others on Photoshop. I've never been happy with what 
>gets missed out by the scanning process - I see it on the slide, but it 
>drops out on the scan.
>
>I agree with Windsor on the Vuescan manual. Some of the most useful 
>stuff I've found out here.
>
>James

Depending on your scanner and whether you are using icc profiles
you can loose data in the shadows from the profiles themselves.
I'm not sure the it8 targets that are used to generate icc profiles
cover the dark regions well. And NikonScan has a tendency to drop
the shadows when you use color management with the scan.
My limited understanding of icc profiles is that they are like curves in
photoshop, the shape of which is determined by sampling various
colors/intensities and interpolating the RGB curves from that. It is
possible that the interpolation near the endpoints might be off,
especially if there are not enough samples in those regions -
highlights and shadows. (And I'm also not sure what the choice
of gamma does to the whole process). Since film has tails at
the endpoints, the profile is likely to be off more in those regions
when the exposure varies. You would probably need slightly
different profiles based on exposure, which is not really practical.

Nikonscan uses a proprietary form of icc profiles, so if you want to
use your own, you have to disable color management in Nikonscan.
Vuescan can generate profiles from a target, nice feature.

The best scans I get are from slides that have a centered histogram
with very little data near the endpoints. I have also noticed some
scans viewed in PS with histogram seems to indicate that there
is no data up in the highlights (or shadows), while if I search around
the image it is clear there *is* highlight and shadow data. The histogram
display does not show such data when the number of samples is
small, and it is easy to blow them out when you can't see them
in the histogram display.

The other suggestion is to use Astia if you plan on just scanning
the slides, where post scan you can adjust saturation, etc. 

Wayne - forever confused with color management


==============================================
List usage info:     http://www.zuikoholic.com
List nannies:        olympusadmin@xxxxxxxxxx
==============================================

<Prev in Thread] Current Thread [Next in Thread>
Sponsored by Tako
Impressum | Datenschutz