[Leica] Scanning B&W vs color on a Nikon LS2000

["Mark E Davison" <dmark8@qwest.net>]

Summary: On my Nikon LS-2000 there are visible advantages to scanning B&W
prints vs. negatives when using Tri-X, which has visible grain. For color
negative materials the advantage goes the other way. The root difference is
that the LS-2000 film scanner at 2700 dpi cannot resolve the film grains,
leading to grain aliasing, which reduces the apparent micro-contrast and
smoothness of tones. The grain aliasing is visible after sharpening as a
"salt and pepper" appearance, as opposed the the fairly uniform grain
patterns in the original print. In a recent experiment, scanning a 9 x 6
inch traditional print on
my Epson 1680 at 600dpi, the grain pattern was resolved satisfactorily.

Question: has anyone tried scanning B&W negs on a 4000dpi scanner, with a
film that produces visible grain? Can the scanner resolve the grain well
enough to prevent grain aliasing?

Caveat: this difference is important in practice only if you are scanning in
order to create prints, or for publication at sizes where the grain pattern
will be just visible. For images destined for the web the differences are so
subtle as to be practically invisible.

Further comments:

See

http://www.photo.net/photodb/folder.tcl?folder_id=132210

for examples of this experiment. This particular example does not
demonstrate the "salt and pepper" appearance very well, but you can see the
reduced micro-contrast. You can also see some geometric scanner aliasing on
the edge of the glasses frame. Choose the "large" size images on photo.net.
I will search my archives for images displaying the "salt and pepper"
aliasing.

Scanning B&W negative material on the LS-2000 is not straightforward.
Besides turning off the ICE defect removal, I also tell the scanner I am
scanning a color slide in RGB at 12 bits per channel, and then invert in
Photoshop. This seems to increase the optical dynamic range that the LS-2000
will cope with. I convert to a black and white image either by using the
Image > Adjust > Channer mixer, or converting to 8 bits per channel, then
changing the mode to LAB, throwing away A and B, and converting back to
black and white.

My scanner (or perhaps the driver software--I'm still at firmware 1.12 and
Nikon Scan 2.2) has a glitch that forces me to turn the scanner off and on
again after I change settings. What a pain.

I get the same "salt and pepper" grain aliasing with Neopan 1600.

The advantage of scanning prints is not large enough for me to install a
dark-room. I am waiting to buy a 4000dpi film scanner first.

For the print in question (a vintage 1970's shot of my wife in her youth)
untrained observers actually preferred an inkjet print made from the
negative scan to either the original or and inkjet print of the print scan.
This is because the loss of micro-contrast made her skin look better!

It's amazing how good the inkjet print from the scanned print is! It some
ways it even looks better, because the old glossy print is a little faded,
and the new print is on archival matte paper.

I have seen Ralph Gibson's inkjet prints from flat-bed scans and they are
quite beautiful. He loves the look of B&W grain (he uses Fuji Neopan 1600),
and it comes through very well in the flat-bed scans. The pictures on his
site, of course, cannot do justice to the inkjet prints, as the texture of
the grain is invisible on the low resolution screen images.

I am getting much better B&W prints from my Epson 1270 since I calibrated it
using Monaco EZcolor. Now I get a very neutral gray-scale. It works best if
I convert the image to RGB just before printing, and set the print dialog
box in PhotoShop to convert to the calibrated space in PhotoShop itself
(Print > Space: > "choose the .icm file generated for the printer by the
calibration process"). The printer driver is set to do no color correction
at all (EPSON 1270 Stylus Photo 1270 Properties > Mode: "Custom", Advanced >
Color Management: "No Color Adjustment".)

Anthony A's bland assertion that prints always clip the dynamic range of
negatives is, as usual, both right and wrong. It is right for most automated
color prints that I get from drug stores. It is wrong for carefully done B&W
prints, such as you might do yourself or get done by a custom lab. There is
a vast literature on
creating B&W prints where either dynamic range clipping does not occur, or
the clipping is under the conscious control of the photographer. That is
what the zone system is all about.

Note: if you are creating prints to scan, you may want to create a print
that doesn't appear to completely fill out the exposure range of the
paper--this way
you can avoid unintentional clipping, and make the final judgement as to
which shadows to let go black after you have scanned.

Mark Davison