Re: Grain and film scanners

[Paul Salvaire <korop@xxxxxx>]

>From my experience with colour negs & CCD film scanners, the problem is
not really in the quality of the film, but in the quality of the scanner
illumination source. I.e.: don't change the film unless you change the
scanner :-)
B&W photogs will remember the old problem of condenser vs diffusion
light sources. Desktop CCD scanners use a hard (condensed, or  linear)
light source bringing the old "Callier effect" blues : distorted high
densities with coarse, mottled grain, due to light diffraction within
the film. This is one reason why CCD scanners have a hard time with
dense areas of slides, even when the manufacturer claims to scan Dmaxes
as thick as 3.4 or 3.5.
The Minolta QS 35 scanner seems to have a softer light source than most,
the Coolscans (especially the earlier models the hardest, apparently
because of their LED source.
Also, the big LS-4500 from Nikon has tobe watched. Grain blues with the
first releases, improved quite a bit since beginning '97. It's a good
scanner, especially for panoramic 35 mm and 6 x 12 cm shots, but I'd
still shy away from colour negs.
The only manufacturer I know of who really tried to tackle this problem
with CCD scanners is Agfa, with the Horizon (nope, not the Russian pan
camera) scanner, with a linear  aspherical condenser quite efficiently
softening the light beam. This scanner is brilliant, though expensive.
Also quite useful to scan panoramic film strips (like 35 mm x 10 in. !)
or 6 x 17 cm, because the resolution is adjusted to the film width, not
total area.

My advice : do not expect a super detailed grainless quality from a
desktop slide scanner. Get a sharp 5 x 7 " (or 8 x 10 ") enlargement
printed out in a lab, the scan the positive with a good flatbed scanner
(Agfa, Quato, etc.). Even a 4 x 6 " standard print will give better
results than neg.

Paul