re: color neg drum scanning
re: color neg drum scanning
- Subject: re: color neg drum scanning
- From: Rick McCleary <email@hidden>
- Date: Thu, 23 May 2002 09:32:54 -0400
Scanning negs with unknown colors, such as a sunset or
through a microscope, pose their own challenges, and the final color
balance
often is up to the scanning operator to second guess the client.
We've had the best luck training color printing technicians, who are
accustomed to color corrections, to do scanning on negative film.
Creating successful color negative scans seem to pivot on being able to
control for several factors:
1) The moving target presented by different orange masks from film to
film,
2) Inconsistencies in the orange mask within one film type due to
processing variances,
3) Characteristic curves that are radically different than those of
transparency film (particularly at each end),
and perhaps most importantly -
4) Unknown colors
Three of these factors (1, 2, 4) relate directly to neutral gray
balance. And the fourth (3) relates to black point/white point. Since
a negative is an abstraction of a scene (as opposed to a literal
rendition - as is a tranparency), we have an impossible situation if the
person running the scanner is not the same person who shot the picture.
The scanner operator can not view a negative and know what the intention
of the photographer was. Where to set BP/WP? What to key on for
neutral gray? There are interpretations to be made that the scanner
operator is not equipped to make.
And so, here we are - mired in a world full of spectacularly
sophisticated instruments being operated by intelligent humans - and
still we can't get good drum scans from color negatives (unless the
photographer happens to own the Tango himself - not likely!)
I think the problem is not one of specific settings, targets, etc. But
rather a more general issue of approach and expectation. The tradition
in the scanning business (service bureaus, printers) has been to supply
a scan with optimal tonal range and neutral-balance color. That is an
unrealistic, impossible standard when scanning color negs for the
reasons stated above. I would suggest a different standard for color
negs scans (i.e., high-end, fine-art quality scans). And that standard
would not be to capture optimal tonal range / neutral-color balance.
Rather, it would be to capture (in a 16-bit scan) ALL THE INFORMATION IN
THE NEG regardless of optimal tonal range and neutral color balance.
The scan, when viewed raw in Photoshop would, of course, look like dog
droppings - flat and unsaturated. But if all the information is there,
the image would be correctable with a few curves moves. And it works -
I've done it. The challenge is to get the scanner operator to do that
raw capture (with no USM!) and leave the interpretation to the
photographer.
I had on off-line exchange with Bruce Fraser about this, and he
suggested the following (... and Bruce, if you're reading this, tell me
if I've got it right):
"tell the scanner operator to do as little as possible, and just grab
the raw high-bit data from neg-to-pos conversion. That means not setting
black and white points, not doing anything to gray balance, etc. There
should probably be at least 20 levels of headroom at the shadow end and
about 30 or more at the highlight end"
Perhaps high-end service bureaus running Tango's could offer color neg
scans in this way. If a customer comes in and asks for a color neg scan
"flat and unsaturated", that would convey the message.
_______________________________________________
colorsync-users mailing list | email@hidden
Help/Unsubscribe/Archives:
http://www.lists.apple.com/mailman/listinfo/colorsync-users
Do not post admin requests to the list. They will be ignored.