Olympus-OM
[Top] [All Lists]

[OM] Re-interpreting PHOTODO MTFs--long!

Subject: [OM] Re-interpreting PHOTODO MTFs--long!
From: ALEXSCIFI@xxxxxxx
Date: Thu, 27 Jan 2000 19:18:53 EST
Hi Folks,

    As an owner of the 100F2.0 OM, I too was disappointed (Initially) in the 
test results. However, I've thought about it a bit and have reached a 
conclusion that Olympus and Miatani may have DELIBERATELY eschewed a pure 
numbers approach (highest MTFs or resolution) in favor of something more 
subtle. Canon themselves say in their EOS lens book ("EF LENS WORK II", page 
200): 

          ". . . that the closer the characteristics of the M (Meridional)
           and S (Sagittal) curves are, the more natural the back-
           ground blur."

In other words, how well do the solid lines (S) overlap the dashed lines (M) 
at the same frequencies. "Background blur" by the way, is Canon speak for 
"bokeh" in their American literature. If ones skips around and looks at 
various MTFs  of 80-135 teles for the better brands, you'll observe some of 
the following:

          *    The OM 100F2.0 has one of the best behaved set of MTFs 
                wide open, in terms of M matching S, general smoothness
                from center to edge (left to right), and flatness of the 
curves.
                Even over it's higher scoring sibling the OM 90F2.0.

         *     In fact wide open, the OM 100F2.0 blows away a number of 
                others--the higher scoring Nikon 105F1.8 for example. And the
                most important aperture (and most used) is F2.0!

         *     PHOTODO's test were undoubtedly at large distances. It
               is entirely conceivable that the OM 100 (and the 90 without
               a doubt) optimum distance might be much closer. A
               similar observation has been made about one of the Pentax
               85F1.4 lenses--it performed better up close than far. This 
makes
               sense if you think in terms of optimizing performance for
               bokeh.

         *     My own personal observation based on some 16x24 enlargements
               and innumerable slides, is that the OM 100F2.0 is something
               "special"!
               
         *     Also consider the overkill construction of the OM 100F2.0 and 
               90F2.0, particularly the "9" diaphragm blades when most others
               are content with 8.

          *    The newly designed and introduced Minolta 85F1.4 lens,
                displays very similar behavior as the Olympus 100F2.0 
                in terms of curve characteristics and even the final score
                of 3.9!  Minolta is known to be among those that put an
                immense premium on "bokeh", and some of their other
                lenses are among the highest scorers in their class--the 100
                macro and 28F2.0 for instance. So it's not as if they don't
                know what they're doing. 
 
          *     The 90F2.0 macro's own curves are very good in terms of 
                 the above characteristics. Given the fact that it's the only 
F2.0
                 macro ever made (AFAIN), it's 4.2 can be considered an
                 excellent. Most of the others--Leica, Canon, etc., are F2.8s.

          *      Meanwhile the 35-80F2.8 zoom holds it's own at F8.0 
                 against some of the best--Canon and Zeiss efforts while
                 being faster than most (F2.8) and more compact than most
                 (62mm filter thread as opposed to the 82"!" mm filter threads
                 on a number of Zeiss zooms, Canon was generally 72mm).

    So to use an analogy from audio, numbers are nice, but the proof of the 
pudding is in the listening! And regretfully, the West (US and Europe) have 
lagged some 10-20 years behind Japanese aesthetic trends in audio (tubes, 
single-ended designs, horns, etc.). The same appears to be true in 
photography. Only Mike Johnson's Phototechnique makes a regular mention of 
"bokeh". I have yet to see it taken into account in any other American or 
English publication.

Alex


< 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 >


<Prev in Thread] Current Thread [Next in Thread>
Sponsored by Tako
Impressum | Datenschutz