[filmscanners] Scanner Noise

["" <HMSDOC@aol.com>]

> However, the theory is clear and simple.
> > Noise is random; data is consistent. Therefore if you combine the
> > passes, and average the result, you should push down the noise level,
> > by the square root of the number of passes. That would cut the noise
> > about in half with 4 passes, to 1/4 with 16 passes.
> > Of course it will lengthen the scan times, but you won't want to do it
> > on most of your images, just those with especially important shadow
> > detail in slides, or highlight areas in negatives.>>>>
> >
> >
  I tried to repeat the scan in Vuescan using 8 passes.  The dark areas
  were now a nice smooth black with little noise.  On the Preview window
  in Vuescan the image appeared very pixelated around the edges leading
 me  to worry that perhaps the scans were not well lined up...but in
 photoshop there was none of the pixelation and the image looked good.
  In fact, I think it may have even looked sharper than the original
  single pass scan, though I did not do a careful comparison.  Would it
 be  possible for it to actually be sharper?

  In this particular image there was not that much shadow detail, but
 this  will obviously be very helpful for scans where there is.  Is there
 a way  to predict if a certain amount of noise (ie. noise visible at a
 certain  % enlargement in Photoshop) will show up in a print of a
 specific size?

  Howard
  HMSDOC@aol.com

----------------------------------------------------------------------------------------
Unsubscribe by mail to listserver@halftone.co.uk, with 'unsubscribe 
filmscanners'
or 'unsubscribe filmscanners_digest' (as appropriate) in the message title or 
body