RE: filmscanners: Re: Dynamic range

["Shough, Dean" <dean.shough@lmco.com>]

> What is the dynamic range figure - i.e.3.2, 3.4 or whatever - a
> measurement of?  Or maybe I should ask, what is the unit of
> measurement?
>

Two different answers:

1) The units are "specs" and it is a measurement of how far the manufacture
is willing to push them.

2) There are no units for dynamic range - it the log base ten of the ratio
between the lightest and darkest material that can be measured.  These days
it tends to be nothing more than the range of the A-D converter, with each
bit adding 0.3 to the dynamic range.  Thus a 12 bit system is said to have a
dynamic range of 3.6.

As far as I know there is no standard on how to decide what is the darkest
material that a scanner can measure.  As a test I placed a series of neutral
density filters (no filter, ND 1, ND 2, ND 3, ND 4, and ND 5) in my Microtek
8700 scanner.  I was clearly able to distinguish between the ND 4 and the
ND5 filters, but only because each was a large uniform area that I could
average over.  The noise level was larger than the difference, but by
looking at a large number of pixels I could tell the two patches apart.  I
had enough peculiarities in this test that I want to repeat it to see if
something was wrong.