At 03:23 1/8/01, C.H. Ling wrote:
The SC 5xx I have is an exceptional one, which is the best 50/1.4 (in
resolution) I have ever seen.
One point I'm not quite sure, seems that starting from 5xx,xxx the
construction of 50/1.4 was not changed (except coating), the different
in performance might be came from the manufacturing, as the production
quantity of 50/1.4 was very high, tooling or fixture might wear and
changed a few times during the whole manufacturing history.
C.H.Ling
I will second C.H. Ling's observation about manufacturing variation and
tooling wear. I know it all too well with what I do professionally in
electronics manufacturing.
I will add that performance variation can also occur in used lenses with
their varying condition and generally unknown provenance. A lens may have
been damaged and repaired to excellent cosmetics but not to original
optical specifications. It is a risk the buyer of any used lens takes,
unless the complete provenance of the lens is known, which is quite
rare. Additional variation among a range of user observations about
performance can also be caused by the fact they are used on different
camera bodies. There _will_ be some variation in focusing accuracy (flange
to film plane distance versus flange to mirror to focus screen distance)
which _will_ confound lens performance. What makes Gary Reese's data
valuable is generally consistent use of the same set of camera bodies in
his testing . . . which is why I'm glad he has posted which body was used
for each test.
While my 50/1.4 MC with 789,### S/N is second to my 50/1.2, it is a very
close second, and both are much better than the old 50/1.8 F.Zuiko I bought
in 1980. I'm not that surprised that C.H. Ling has a 500,### S/N that
performs better than all the other 50/1.4's he has. All the other sources
for possible variation in performance can dominate theoretical differences
inherent in the designs when you compare specific lenses.
The evidence shows some measurable probability if a specific 50/1.4 "Zuiko"
has a S/N >1,100,###, it will be a better lens than a specific "MC" with a
S/N <1,000,###, or a specific "G.Zuiko" but it is definitely _not_ a 100%
guarantee that it will be. I believe this is all one can legitimately
conclude from Gary Reese's testing: statistical significance between the
mean performance of entire populations of different versions of the 50/1.4
Zuiko.
-- John
< This message was delivered via the Olympus Mailing List >
< For questions, mailto:owner-olympus@xxxxxxxxxxxxxxx >
< Web Page: http://Zuiko.sls.bc.ca/swright/olympuslist.html >
|