Re: [Leica] Subjective lens impressions: Erwin's comments

[Erwin Puts <imxputs@knoware.nl>]

I wonder how much of the difference between Erwins and
>Alfreds statements could also arise from different samples of the same
>lens? I would be very interested to know if Erwin has ever done optical
>bench comparisons between identical lenses. Can such variations ever be
>greater than the design differences? I guess (hope) variations are less on
>current than older lenses? I have always believed that part of Leica's high
>costs are due to covering higher rejection rates to maintain tighter
>consistency between samples. Is there any factual data on this?
>Sorry for so many questions. If this is a well worn thread could someone
>please summarise, or point me to some web addresses to catch up.
>Mike

Yes, I am in the happy position to be able to compare several lenses of the
same  series and even to compare lenses from different periods.
Remark one: comparing identical lenses from the same series: any lens I
have measured is slightly different from anyother sample.I have not yet
found lenses that are absolutely identical. In some the contrast is s
little higher, or the resolution a bit better, or the offcenter aberations
are a bit different or differently distributed. etc. When you rotate the
lens sometimes the performance also differs. My rule is to find out the
worst position of the lens and record the measurements of the worst
situation. Some older lenses have the noted difference between best
resoluation plane and best contrast plane. In that situation I look for the
extremes and excercise some intelligence to balance the conclusions.
You are right: current lenses have less sample for sample deviation than
older ones. But when doing the Summicron evaluation I used two samples of
every version, from different production years, and they quite closely gave
the same kind of performance. Design differences invariably have a much
greater impact than production differences.
There are no facts from the factory about rejection rates. I once put five
new Summicrons out of the box on the bench and noted some variation in
performance (see above). Leica'has pointed out that they have minimum
standards of performance and that every lens at least should have this
minimum level. All the samples had this minimum, but some were slightly
above. Do you see it in practice: I doubt it.
Erwin