I have been reading digests for a while now, but have never posted. However,
the recent spot metering thread has sparked my interest. A worm of doubt has
been planted into my mind...
So I checked my OM-4T and OM-2S and to my horror I noticed the same kind of
discrepancy. I have been taking both spot and CW readings for my pictures and
have never seen this problem in practice (a 1.7EV underexposure would
definitely show on the slides). What was I doing wrong? What was I missing?
Then it hit me. There are four exposure "readings" available on the camera:
1. average indication (using the fresnel mirror, value not stored)
2. CW exposure (off the film, actual light read durng the exposure)
3. spot indication (using fresnel mirror, stored into memory)
4. spot exposure (using memory only, light is not measured during exposure)
What matters is "2" and "4". I promptly checked that (with a piece of film in
the camera). Bingo! "2" and "4" agree to within 1/3 EV. "3" also agrees with
"4" (hardly surprising because 4 is based on the stored value of "3", no light
is measured during the spot exposure). "1" is obviously wrong, depending on the
lens in use. But "1" serves only as a guide, a rough indication. Now all is
well and I can go back to taking picures.
Some further tests (nothing rigorous, more for the sake of curiosity) revealed
that the difference between "1" and other readings depends on the focal length
of the lens, as well as the maximum speed. My selection of lenses is somewhat
limited, so I could draw limited conclusions. Slow short lenses seem to cause
largest discrepancy, fast and/or long lenses are on the opposite end of the
scale. For 50/1.8 the readings are exactly the same (this is the lens normally
used during light meter adjustments).
I am really guessing now about the cause of this behaviour. One possibility:
during CW indication the fresnel mirror is used to emulate the off the film
light reading. In fact the light path is shorter and for this reason the mirror
is not flat (fresnel lens on it). The ability of the mirror to emulate the OTF
reading may depend on the size (angle) of the light cone exiting the lens, and
this angle is narrowest with the slow short lenses (small exit pupil). An why
does the spot metering not depend on this angle? Spot angle is much narrower
than the light exiting any practical lens (spot meter only "sees" the centre of
the light cone). Perhaps?
As I already said, this is just a guess. A better explanation may exist.
However (at least in my case) the practical implications of the "problem" are
of no real consequence for the exposure accuracy and I think it is not worth
losing any sleep over this.
Rod
< 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 >
|