Anyone interested in scanning LF films/plates on flatbed scanners, a
paper published in vol 68, issue 5 (most recent issue) of Journal of
the Society of Photographic Science and Technology of Japan is of
great interest. Nakatsuchi et al. used MTF test target and tested 4
flatbed scanners (2 Epsons, 2 Canons). They were interested in
selecting a scanner to archive historical silver gelatin plates.
One thing they found is that, the actual resolution of Epson GT-X700
and Canoscan 99x0F (where x is 0 or 5) are only half the value
actually set when scanning. That is, if you se the resolution to be
4800dpi, you'll get 2400dpi worth of information, but if you set the
scanner to 2400dpi, then you'll only get 1200dpi worth of
information. This is a waste of 4x storage space for the same
information. On the other hand, they found that Epson ES-10000G gave
actual resolution very close to the nominal scanning
resolution. ES-10000G is also the only model with focus adjusting
mechanism among the scanners mentioned here.
>From all these info, it is pretty essential to select a good scanner
with double the resolution you need, and always scan at least double
the resolution you need, and use a decent image manipulating software
to downsample the image, rather than letting the scanner or scanner
driver do this work. (However, this is pretty consistent with what
most people found by experience... nothing new here, except that
Nakatsuchi et al. did the right thing to obtain firm evidence.)
They decided to go with ES-10000G and they also determined the best
part of the scanning area in terms of resolution, made a plate holder,
and came up with a good focusing strategy.
They also say that none of the flatbeds is suitable for MF/35mm films.
Received on Mon Dec 19 17:22:52 2005
This archive was generated by hypermail 2.1.8 : 01/05/06-01:45:11 PM Z CST