Archived posting to the Leica Users Group, 1998/04/24

[Author Prev] [Author Next] [Thread Prev] [Thread Next] [Author Index] [Topic Index] [Home] [Search]

Subject: [Leica] Leica: MTF question.
From: "Dan Post" <dwpost@email.msn.com>
Date: Fri, 24 Apr 1998 11:12:41 -0400

I hope this doesn't start a flame war, but the recent discussion of
Modulation Transfer Function, aside from the concommitant copyright
discussion has me a bit confused.
In Neblette's book on photography, he holds that the MTF is a 'system'
measurement and includes not just the camera and lens, but the film, and
processing, at the very least. This would mean that for each film/developer
combination, there would be a different MTF chart, even with the same lens
and camera combination. I got the impression that the MTF was of limited use
for judging a camera/lens. It rather gave a generalized idea of the
performance, that taken with other data, the MTF is merely a component of a
method for evaluating optical systems.
I could see that if the tester standardized his film and development, say
with Hurter and Driffield's formulas for sensitometric evaluation, and
tested many camera and lens combinations, that you could make judgements for
that 'population' of cameras, at that particular time, and for under that
particular film andeveloper combination.
If one goes on to make prints, then MTF data should include testing the
camera/lens combinations with the particular enlarger lens/paper/development
data... I can understand the concept of multivariant analysis, but it seems
that pretty soon, you end up picking the fly spots out of the pepper, and
really not getting anywhere.
I can tell you that when I started to use a Leica, I had to hustle to get a
lens on my enlarger that could do justice to the negatives. I use Vivitar
VHE lenses- they were the best I could get, and I am sure that if I got a
Leitz lens, I could even see more improvement!
Do they now use lasers for MTF? What is the process today? Is it able to be
standardized so that a lab on one continent can reproduce the results of a
lab elsewhere?
TIA,
Dan'l
dwpost@msn.com