Re: 16 bits = 15 bits in Photoshop?
Re: 16 bits = 15 bits in Photoshop?
- Subject: Re: 16 bits = 15 bits in Photoshop?
- From: Tim Vitale <email@hidden>
- Date: Thu, 14 Apr 2005 20:05:37 -0700
I capture historic negatives that can have up to a density of 3.9D, but
average about 3.5Dmax for the denser negatives. Many older negative are
similar to today's negatives, with only a 1.6-2.2 Dmax.
The latter modest tonal range negatives still requires at least "a
12-bit capture", because 8-bit RAW, Gamma 1.0, capture has a usable
maximum of 1.5 Dmax, or a range of 5-stops. When 8-bit G1.0(RAW) data
is displayed at G2.2, the maximum usable density is 2.7D. Most films
produce a better tonal range than 8-bit RAW, 0.2-1.5 D. Most films have
an input tonal range (exposure) of 7-8 stops, or density range of
2.0-2.5D. When the 8-bit "RAW capture" is converted to 8-bit G2.2 space
for display, the range increases to 0.2 to 2.7D, much more respectable.
12-bit RAW capture has a 2.7 Dmax, RGB=8. When RGB=8 (2.7D at G1.0) is
converted to G2.2 for display, it has a 5.3D density.
Capture on a true 16-bit scanner allows 13-stops, or a 3.9 Dmax. The
last 0.1D step, after 3.8 D (RGB=10) to 3.9D (RGB=8), has only two RGB
values, RGB 9 & 8; one value for data and one for noise. OK, maybe
16-bit can't really do a full 13-stops, but that is the buzz.
I was taken aback to be *reminded* that the PS-CS implementation of "16
bits" was still 15 bits(+1). I understand that this quirk is to allow
faster operation of older platforms, but, I'm not happy about the lack
of truthfulness in such highend imaging software.
On the optimistic side of this issue: I plotted a 3.9Dmax, RGB=8,
G1.0(RAW) at G2.2 for both 15- & 16-bit display. In 16-bit G2.2 space,
3.9D is RGB=1106. In 15-bit G2.2 space, 3.9D is RGB=553. Plenty of
room to move around, but not as advertised all over the software,
documentation and literature. Many users are not number freaks and
don't care, but I expected better. Certainly they are answerable to the
folks that know what the numbers should be, even if its only an asterisk
or a pop-up message somewhere.
I'll have to scan a Stouffer 4.05 D step target (with 0.1 D steps) at
16-bits; correct the step values to the appropriate G2.2 slope in
Picture Window Pro; then determine if PS-CS is giving true 15-bit RGB
values all the way across the scale. I think I remember someone telling
me a few years ago there was a disconnect in the middle of the scale.
...unless someone else has confirmed this already?
Tim Vitale
Paper, Photographs and
Electronic Media Conservator
Digital Imaging and Facsimiles
Film Migration to Digital
Preservation and Imaging Consulting
Preservation Associates
1500 Park Ave
Suite 132
Emeryville, CA 94608
510-594-8277
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Colorsync-users mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden