As good as the 100mm f2.8 may be, is it as good as, say, a 100mm F2 (9
blades, a 90mm F2 (9 blades) or a 135mm F2.8 (8 blades)?
I certainly can't answer this question, not having used a 100mm f2.8. I
am not implying that the 100mm f2.8 is not very good, bokeh or otherwise,
but it is always possible that other lenses might be even better so I do
not think the good performance of the 100mm f2.8 precludes there being
nicer bokeh and performance in other lenses so one shouldn't conclude that
the blades theory is a myth or not.
I think this blades thing on this list started with me. So you know
who to blame.
Having noticed that different lenses I have produced noticeably different
results in terms of out of focus areas (bokeh) I wondered why. Then I
noticed that my lenses did not all have the same number of aperture blades
and I wondered why that would be and also noticed that there seemed to be
a correlation between what I considered pleasing out of focus rendition
and the number of aperture blades a lens had. It seemed to be that the
more blades a lens had the more I liked it's bokeh.
So then I tried to think of what physical mechanism there might be to
explain this relationship. So after staring at a few photographs for a
while I came up with this theory.
Out - of focus highlights often can be seen to have an angularity which
corresponds to the shape of the aperture - hexagonal, octagonal etc. I
believe that this shaping caused by the aperture can not just be confined
to highlights but that it must affect all out of focus areas, it's just
more obvious with highlights.
The more aperture blades a lens has the more nearly circular it is and so
the less angular, more nearly circular, will appear any out of focus
highlights - and all other out of focus areas.
Heres the tricky bit to explain. - Now if you can imagine that all out of
focus elements in a photo have been rendered by a hexagonal aperture,
then one effect will be that all elements have been vaguely shaped as
hexagons - I don't mean that the whole out of focus area will appear as a
myriad of distinct hexagons, because in most circumstances it wont as
there are so many of them (infinite?) and they are all very blurred.
However, all these hexagons, though individually indistinct will have a
cumulative effect that the brain picks up on because there will be a
myriad of lines which are all parallel to each other. I probably haven't
explained this very well.
I believe the impressionist painter Seurat painted with distinct dots of
colour which up close look like dots but when viewed from a distance an
image formes as the brain integrates the distinct elements to form a
recognizable image. I think if the dots of paints were triangles - all
perfectly aligned so that all equivalent sides were parallel to each other
then one stepped back - an image would still form but it would seem to have
a different quality to it than one composed of round dots.
It's just a theory.
Giles
sje@xxxxxxxxxxxxxxxxxx wrote:
> Which brings me to another question:
> As this lens is so good why do people suggest that 8 or 9 blades are
> preferred? The Zuiko 100 f/2.8 also has only 6 blades but also has good
> performance (including bokeh). Is the "more blades=better" a myth? There is
> likely to be some improvement but I suspect it is not as important as some
> people maintain. Those of you with 9-blade lenses please don't feel too bad
> about it ;-)
>
< 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 >
|