Hmm... The last disclaimer in the challenge is interesting: actually, the
reason I work with 16 bits is that every time you apply a curve (which I do
quite extensively in B&W images) you stretch a range of values and compress
another: with 16 bits you have a better rounding of the newly generated
numbers, and thus get smoothest values to start with when you apply further
changes. Now, it's absolutely reasonable that you'll never see any
difference with usual tweakings in photographic images, yet the fact that
Dan reserves himself the right to convert the 16 bits image to 8 and then
back to 16, although only in extreme cases, proves that *working* with 16
bits may be worthy, no matter how much information you start with.
Anyway, I'll experiment a bit (no pun intended) myself: working with 8 bits
may have the extra advantage that I'll need not buy a new PC when I get a
4000dpi scanner...
Alex
-----Original Message-----
From: Maris V. Lidaka, Sr. [mailto:mlidaka@ameritech.net]
Sent: sabato 14 luglio 2001 19.24
To: filmscanners@halftone.co.uk
Subject: Re: filmscanners: Semi OT: 16-bits [was Which Buggy Software?]
<snip>
Meanwhile, I am attaching his public challenge, as posted on his Color
Theory group , which asserts that
no photographic image can be corrected better in 16-bit color than it can in
8-bit color.
Maris
