Mailinglist Archives:
Infrared
Panorama
Photo-3D
Tech-3D
Sell-3D
MF3D
|
|
Notice |
This mailinglist archive is frozen since May 2001, i.e. it will stay online but will not be updated.
|
|
P3D Normal lenses and other myths...
- From: fj834@xxxxxxxxxxxxxxxxxxxxx (Dr. George A. Themelis)
- Subject: P3D Normal lenses and other myths...
- Date: Tue, 30 Dec 1997 16:28:28 -0500 (EST)
Mike K. asks if there is a technical reason that the "normal" focal length is
equal to the diagonal of the image frame.
This is one of the unsolved mysteries of photography for me. I believe
that this distinction is arbitrary. The explanation usually given is that
this focal length matches the field of view of human vision which is
approximately 50 degrees.
What do they mean that the field of view of human vision being 50 degrees?
We see a sharp image at the center that fades gradually as we go to the
edges of the field of view, well beyond the 50 degrees.
Nobody seems to take into account the *viewing* distance in these
discussions. If a negative or slide recorded with a 50 mm lens is viewed
from 50 mm distance then the scene is reproduced as the eye sees it. Same
if the image recorded with a 135 mm lens is viewed from 135 mm distance.
It will be smaller but accurate in proportions.
The well-known "compression" effects of long focal length lenses are a
result of viewing the image from a distance closer that the "ortho" viewing
distance. (Yes, there is ortho in 2-d too!) The picture will look normal
if viewed from further away.
So, the question is, how do we view things? I once did an experiment. I
took images of a wide variety of sizes, from a stamp to a painting, and
asked different people to view them from the distance most comfortable to
them. I recorded the diameter of the image and the viewing distance and
calculated the ratio. As I recall, there is a correlation (the larger the
image the further away we stand to enjoy it), a bias towards longer viewing
distances (we tend to stand further than the diagonal of the image), and a
non-linear correlation (I forget the trend of this non-linearity).
For these thoughts and observations I conclude that the definition of a
normal lens as matching the diagonal of the image is an arbitrary but
practical one since we normally view images from distances close to the
diagonal, which is of course related to our field of view (the distance we
stay away from an image to comprehend it and enjoy it is related to our
field of view).
As an extension, I believe that photographers are overusing/abusing short
focal length lenses. Most P&S cameras come with 35 mm lenses. Our desire
to capture more make us less careful in what we capture and how we do it.
An 80 mm lens in a 35 mm camera would make us more selective on what we
decide to include in our pictures and make the images more pleasant and
"accurate" given our longer viewing distance bias.
Similarly for stereo, we tend to view images further than the "ortho"
position either with viewers or in projection. I would love to have a
full-frame RBT camera with 70-80 mm fixed lenses. That would work well in
projection too.
-- George Themelis
------------------------------
|