[filmscanners] Re: IV ED dynamic range... DYNAMIC RANGE!

["Anthony Atkielski" <anthony@atkielski.com>]

Julian writes:

> (long explanation omitted)

Sounds reasonable to me.  Given this explanation, some other conclusions can
be drawn:

1. Unless a scanner and its software allows for multiple passes _with
different exposures_ for a single scan, only the dynamic range is important,
unless you plan to do multiple scans of an image and combine them in
Photoshop (which in turn pretty much implies that the film must not move in
the scanner between scans, for proper alignment).

2. If the scanner uses autoexposure and if the density range of the film
image falls within the dynamic range of the scanner, a single scan with
autoexposure is sufficient to obtain all useful detail, since the scanner
will slide the exposure to fit the density range of the image into the
dynamic range.

3. If the film density range for a given image exceeds the dynamic range of
the scanner, you either live with some cutoff of shadow or highlight, or you
do multiple scans with different, manually-set exposures and try to merge
them in Photoshop (probably not worth the trouble for most scans).

Next question:  Can I reliably tell from the histogram whether or not
everything was captured in one scan?  If the dynamic range falls off
smoothly towards its limits, it might be hard to distinguish any chopping of
the image in the histogram.  On the other hand, if it falls off abruptly, it
should be easy enough to see evidence of this in the histogram, either as
data that suddenly stops at one end, or as data that continues or bunches up
right against the extremes (although this must be distinguished from the
actual maximum and minimum densities that might occur on film, i.e., clear
film produces an abrupt cutoff on high-key images, which may show in the
histogram if the scanner can go further).


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