Re: [Leica] Re: Digital Leica and reality

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

Frank Filippone writes:
>
>You are correct that in the imperfect world we live in, the lens, paper,
> etc. all invent things ( we call them aberrations ) that were not in the
> original image. They are somewhat uncontrolled and unrepeatable.


I think there are some valid points being missed in these "debates".

1) In traditional photochemical photography, the aerial image which the lens
forms on the film modulates a photochemical reaction which has an intrinsic
stochastic nature. One way to think of this is that the aerial image is
approximated by stochastic half-toning. When people talk about how much
"information" there is in the film image they mean one of two very different
things:

a) how much reliable information is there about the original aerial image
that the lens formed or
b) how much data must be collected to convincingly reproduce the grain
pattern as well.

Both questions are valid, and they have different answers. Considering the
two cases makes it clear that photo-chemical image capture is not an
intrinsically "pixel" oriented process.

In some photographic aesthetic traditions the grain pattern is extremely
important to the final look of  the photograph, so asking question b) is not
just idle Luddite speculation.  The question is: can I reproduce traditional
black and white grain-aesthetics by scanning film and printing to an
inkjet-printer?

I have not been able to do this with my LS-2000 and Epson 1270, and I
suspect that the 2700dpi limit on the Nikon is the problem.

Inspite of Anthony A's theoretical calculations, I have seen beautfiul B&W
prints (by Ralph Gibson) with wonderfully reproduced grain which were done
by scanning traditional B&W prints on flatbed scanners, and then printing on
Epson printers using piezography drivers. Mark Rabiner has confirmed that he
is doing this also. This bee flies!

2) Both traditional photochemical prints and inkjet prints use forms of
half-toning: very fine-grained photochemical half-toning in the former case,
and various forms of half-toning (usually quasi-stochastic and not
dithering) in the case of the latter. It is notoriously difficult to reason
about half-toning systems, as they are designed to fool the human eye rather
than to submit to simple-minded analyses. Look at some good Epson 1270-1280
prints under a loupe to get an idea of what I mean.

3)   I always thought that the purpose of photography was to make images
that communicated visual messages and were sometimes beautiful, rather than
to live up to the ultimate resolution capabilities of any given lens.
Besides, high-quality lenses have important properties other than
high-resolution, such as: high contrast at lower resolutions, freedom from
flare, freedom from excessive geometric distortion. I believe you will
notice these properties at the lower resolutions required by anti-aliasing
to 6MB chips. You certainly can see all these properties in 2700 dpi scans.

4) If you do service photography for non-profit groups (churches, schools
and the like) you will quickly see that there are times when digital capture
offers compelling advantages, chiefly in labor costs.

Examples:

take 200 head shots for the church directory and deliver them in digital
form for incorporation in a published member roster

photograph 100s of objects for incorporation in an auction catalog (also to
be published electronically)

Note that in both cases you are photographing for publication, where digital
technology has completely won. It is not just the web which requires digital
(or digitized) images.

Also the images in question are not going to be timeless art. You will never
make them into large prints for gallery display, and there is no incentive
to archive them for posterity.

5) Digital capture can also be a big benefit in teaching photography, as I
have done at my daughter's elementary school. Here the immediate feedback is
quite important, as are the lower processing costs.

6) Note that digital photographs taken in black and white mode can be
astonishingly beautiful, as you get around the color interpolation problem.
My experience here was using an Olympus C3030 on a tripod doing still-lifes,
working the children of my daughter's class. The prints were 6 x 9 inches,
done on an Epson 1270.

7) Even if you are willing to restrict yourself to the smaller print sizes
which typical single-shot digital capture limits you to, there are
weaknesses in the current (pro-sumer) cameras. I find I cannot use the C3030
for candid shots of people in naturally-lit interiors--the combined focus
and shutter lag is simply too long. It is also difficult to work the manual
controls--imagine having to set your aperture by nudging a button 1/3 of a
stop at a time! The old analog dials have a real advantage there.


I still shoot artsy stuff and candids on film cameras, but I use the digital
for a lot of service work. I'm also working on a proposal for a crafts book,
and I shoot film for the illustrations, because my digital camera doesn't
produce very good-looking specular highlights in color (it has the notorious
purple-fringe problem).

It will be interesting to see if the Leica/Matsushita offerings get around
any of the current digital limitations.