sje@xxxxxxxxxxxxxxxxxx wrote
> The 'distortion' under discussion was, I thought, a product of focal length
> - - fewer mm=more exaggeration of perspective. So why would one 24mm have
> more than another? I'm confused.
Doris Fang's correction to my post explained this. There are three
separate effects here, all called "distortion":
i. "False" perspective due to the wide-angle lens. What matters
is the ratio of viewing distance of the finished photo to subject
distance when the shot is made. A wide angle lens alters this from 1.
Most viewers no longer object to this. The number who do rises with
more extreme wide angles. More people accept long lens images
than wide ones. A very few people reject non-normal focal lengths
completely, even some famous artists such as Cartier-Bresson and
Jacques Tati.
ii. "False" perspective due to tilting. When viewing the scene
in real life, the eye scans the tall building and the brain
integrates it into a single image which apparently doesn't
tilt for some people (it does in my mind's eye). Some people want
to "correct" for this with a shift lens or view camera. Whichever
way you go you will disappoint someone.
iii. What is sometimes called "linear distortion", the bending of
lines. That's from the bodgy lens. Straight lines which don't pass
directly through the centre of the image are curved or (worse?)
wavy. This is usually worse at the edges, worse with wider
lenses, and worse with cheaper lenses.
Joey mentioned the image "looking funny", which if he is
the average person probably means (ii). He also asked if it would
be worse at the edge, which is only (iii). Either way, he may
be better off with a tamer wide (and, as Doris suggested, a good
book on the subject).
I don't think (iii) is acceptable, but some people do, perhaps the
same ones who don't mind grain. I think (i) and (ii) are almost
always acceptable, but some people don't.
< 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 >
|