Re: M3 and Wide Angle

[imxputs@knoware.nl (Erwin Puts/imX)]

>I found Erwin Puts' posting on his own test results with the 35mm and 50mm
>Summicrons.  That his results do not jive with mine -- nor, for the 50mm,
>with the Japanese test results I posted some weeks ago -- merely proves the
>validity of each of us working with our own lenses for our own purposes.
>One caveat, though:  early lenses, especially, vary a good deal and testing
>ought not be based upon one exemplar.
>
>My experiences are based on the ownership and use, over the past several
>years, of two first-generation 35mm's, two current-production 35mm's, one
>35mm Summilux, two collapsible 50mm's, two rigid 50mms', one NF 50mm, three
>intermediate 50mm's, and two current 50mm's, as well as current version of
>the 50mm Summilux.
>
>I have retained for my own use the 35mm Summilux, one collapsible 50mm, the
>rigid 50mm NF, and the 50mm Summilux.
>
>The Japanese tests, unhappily, were flawed, as were Mr Puts' tests, by being
>performed upon only one exemplar of each version of the 50mm Summicron.
>
>Marc
>msmall@roanoke.infi.net  FAX:  +540/343-7315
>Cha robh bas fir gun ghras fir!

Hello Marc,
How do you know that I only tested one exemplar of each Summicron. I didn't
say a word about that. To be precise, I tested several exemplars and having
access to the Leica quality assurance standards I could easily figure out
which lenses were typical of the performance as the factory intended and of
course I only used the lenses which were well within the factory-tolerance.
My tests are done with strictly controlled testparameters and on the same
testtargets (2- and 3- dimensional).
As far I know, it is valid to base conclusions on lenses which are within
the factory specs to evaluate the typical performance of a given optical
product. I know that two Summicrons will slightly differ in their
performance due to the inevitable production tolerances, but I also know
(and have tested) that the differences within a Summicron generation are
much smaller than between Summicron generations.
In former days the editor of the distinguished British Journal of
Photography told us that a meaningful optical test could only be based upon
a statistical sample of 5 or more specimens of the same lens. But he
referred to japanese lenses when the quality assurance was lower than it is
now. He also explicitly stated that the need for such an approach was not
necessary for Leitz or Zeiss optics because of the much better production
quality.
Erwin Puts