RE: [Leica] Scanning

["Austin Franklin" <darkroom@ix.netcom.com>]

> > From: Austin Franklin [mailto:darkroom@ix.netcom.com]
> >
> > > The recent POP Photo scanner review *measured* the resolution of
> > > the LS-4000 at 60 lp/mm.  As I understand it, noise plays a major
> > > role in reducing the true resolution of a scanner down from the
> > > theoretical resolution (which would be 78 lp/mm for a 4000 SPI
> > > sensor).
> >
> > I'd have to think about that, but that does not make sense
> > (noise limiting
> > the resolution) to me.  Noise certainly does limit the
> > dynamic range (which
> > is on a per pixel basis), but I fail to see how it limits the
> > resolution,
> > which would be across pixels.  If the lense is not up to snuff, or the
> > scanner has vibration issues, I can see that limiting
> > resolution...but not
> > noise.
>
> I agree that electrical noise in the CCD would reduce dynamic
> range, not resolution.  Something is lowering the resolution,
> though (if Pop Photo is to be believed :-)  Could one culprit be
> the scanning lens itself?  After all, the light doesn't go
> straight from the LED to the sensor.  What about scattering in
> the film and emulsion itself?  And bloom, and vibration.  In all
> liklihood there are a large number of factors contributing to the
> degradation, just like there are with any other real-world process.

CCD red channels are typically "soft", because of smear, not bloom.  Smear
is like cross-talk, bloom is like clipping.  That is probably the reason.  I
wish they would test the channels independently...as well as some come out
with a scanner that scans B&W NOT in RGB...  My scanner scans B&W using a
single ND filter, but it's the only scanner I know of that has been designed
that way.