[Leica] optical testing

[Erwin Puts <imxputs@knoware.nl>]

Ask any optical designer about the possibilities and intricacies of 
comparing lenses, they will wring their hands in utter distress and 
propose that we should eat lunch first. And after that event they 
conveniently forget that I ever asked the question. When pressing the 
matter (we Dutch are quite stubborn, remember that we in our Golden 
Century challenged the British and became one of the first world 
powers, being defeated by demography rather than by strategy) they 
will tell you that comparing lenses is a highly evolved matter that 
has to be studied extensively before making any firm statements.
They will point out that there are so many error-sources involved in 
any setup of a test, that there are so many aspects to be considered 
and so many parameters of comparison to be taken care of, that any 
attempt to compare lenses in a comprehensive matter is not lightly 
undertaken. And if some setup might be devised, it is questionable 
what valid conclusions may be drawn from the results.
This extremely cautious perspective may be compared to the many 
attempts of users of photographic lenses and the photographic press 
to construct a test that gives the simple answers we are all waiting 
for. We are very eager to know if lens A is better than lens B and if 
marque A is better than marque B.
And here we go: if we devise numerical tests like SQF and MTF we will 
protest and say that these tests do not take into account the real 
objects we wish to record, that is 3D reality versus 2D test pattern. 
And if some report give higher figures than we assume or expect we 
declare the test wrong. Then we will note that the real test is the 
acceptance by the user and we quote more or less famous persons who 
like or swear by a certain lens. And in the  last resort when we are 
with our backs to the wall we will note that the Human Eye should be 
the final arbiter: do you see IT or do you not see IT.
There is a remarkable unwillingness to accept the bottom line.
The truth is simply this:  every person has an opinion about the 
characteristics of an image produced by the photographic lens onto a 
silver halide suspension  and we all know that this opinion is 
intertwined with a cultural heritage and a psychological mechanism. 
So intimately in fact that we are not able to look at the image from 
the engineering perspective of recording information as a visual 
pattern of density differences. If we study optics we are interested 
in how these patterns are recorded and transformed and when we 
evaluate lenses we wish to know how the recorded patterns differ from 
the ideal ones and we introduce part of visual theory to explain why 
and when some pattern is regarded and interpreted as an accurate 
representation of the reality. (which is a visual phenomenon in 
itself).
The fascinating relation between these two worlds: the optical one 
and the visual one is a source of amazement and wonder and not in the 
least of study.
The real and fundamental question is this: why and in what way do the 
density patterns as recorded by a Leica optical system differ from 
the ones recorded by lenses from marque X and if we are able to note 
these differences in a reliable and consistent way, given IDENTICAL 
circumstances that is a rigorous ceteris paribus clause!!, why and 
how are differences in visual impression and interpretation (if any) 
to be explained and assessed.
The upshot? While I am writing this I am listening to Bob Dylon: The 
bootleg series. The differences between the engineering capabilities 
of recording sound and the capabilities of the human ear to evoke 
emotions and feelings are as great as is the interrelation between 
these two. So the world of optical engineering  when recording light 
signals is closely related to and in the same time different from the 
visual signal processing of the brain. Testing a lens and assessing 
any image produced by the optical/chemical interaction of recording 
light energy in silver halide is a task that demands delicacy of 
procedure and  a firm grasp of both the technical and perceptual 
issues involved. The hesitation of the optical engineer and optical 
designer stands in stark contrast to the ease with which lenses are 
compared and evaluated in the photographic world.
I do not talk about any person who uses any lens professionally and 
is or is not satisfied with the results or his/her clients are or are 
not. That is a different matter: here the position is simply this: 
tool A is not suitable for its intended task. Pity the tools. If a 
different set of tools does satisfy all requirements then glory to 
set of tools B. The inference that tool A is not as good as tool B 
would be too rash, when stated without reference to the goal.
So we need to describe our goals and yardsticks in measurable or at 
least objectified dimensions, when discussing lens performance 
and/or make a careful distinction between the two worlds outlined 
above.
Bernard Shaw was right, was he not?


Erwin