Hi Chris,
I've spent the last
three hours intensely digging into this and have come to a few conclusions
that may or may not be somewhere near reality...
1. The main culprit in my images
looking like snail barf was the misuse of the USM... I was trying to cut it
too fine by putting the threshold at 2 or 3... I was trying to not have it
affect the noise but instead it made detail "blotchy" where it did sharpen...
so to do this right you have to select or mask and then use the USM at a
threshold of 0 so that everything is affected equally...
2. There is definitely a resolution
issue here... when I scan my negatives at 4000 dpi @ 16 bits I end up with a
file size of 120 MB! Averaging four 35 mm negatives like that is going
to be impressive data across the ENTIRE negative... but if you crop in tight
to a 10mm X 15mm sliver of it... things begin to blur out... the bottom
line... if you use the entire 35 mm negative it will blow a CCD shot out of
the water, but if you crop in to the same area as the CCD... the CCD
wins.
3. Many of the images that I did
for Rob were tiled together I believe, or he took very many frames and
carefully combined them for huge S/N gains... looks like that is what you have
to do to get a file that goes 16 X 20 print size.
Tony