P3D Re: Optimum Sharpness

["KENNETH LUKER" <KLUKER@xxxxxxxxxxxxxxxxxxxxxxx>]

John Toeppen responded to my query, but unfortunately posted it to 
Sell-3d.  I've taken the liberty to copy his response to this list, 
where it belonged anyway, because it seems so thorough and well 
expressed.  KL

Ken Luker asked:

"Can someone explain how
one can see the difference between un-sharpness due to wide aperture
(loss of depth of field) and un-sharpness due to TOO SMALL an
aperture which has caused diffraction?  How do these differ in
appearance, and how can one determine the optimum f-stop? "

Spot size due to diffraction is driven by the formula:

2.44 x wavelength x f#

example:

2.44 x 400nm x f22 = 10736 nm = 10.7 microns  (blue light)

example 2:

2.44 x 600nm x f22 = about 15 microns  (red light)

Since 50 line pairs per millimeter equals 100 dots per mm this means
that a 10 micron to 15 micron focus spot size will form in the focused
image.  Since 50 lp/mm is about what you have with Ektachrome 50, I
think that we are talking a marginal loss here.

Most photographic taking lenses of this vintage achieve their best
performance at f5.6  Spherical abberation dominates below this point
and  curvature of field is also a problem.

Depth of field is usually the limiting factor of the perception of
sharpness.  While a blurred background or foreground may be a good way
to direct attention to the subject for composition it is not always 
best
for stereo.  Sometimes the sharply focused flower in the foreground
works best with the distant oaks also in focus.  The large parallax
difference allows both to be seen alone, producing a multisubject
photo.  Sharpness through creates a sense of a sharp image, blurr spot
size be damned!  Compositional choice, I could argue the other way 
too.

I do have a collection of stereo slides of wall size resolution/test
targets taken at David White in the early 50s.  The only perceptable
difference was that all of f11 and f16 had all of the targets (center
and diagonal targets, both radial and tangental, and chroma lines as
well). A little loss of brightness in the corners at and above f16. 
Even
a bad lens can make a good image at f16.

Depth of field is related to the size of a cone of light if it is
intercepted by the film either before or after it comes to focus.  The
f16 spots are thin rays with long larger waists.  The low f#s have a
more hourglass profile along their length so their "not yet focused
spots" are large.  This spot size is a simple trigometric function (=
what is the cone diameter if sliced before focus).

What does it mean - shoot sunney 16?  Yes, in general.  But these are
just reference points.  Composition, light, and control are almost as
important as experiment and general understanding.  Kind of makes one
wonder why people like the f2.8 lens except for low light situations
requiring faster action.

For a more detailed understanding than my "depth of focus and field
commentary, I would seek out a book written by Arthur Cox called
"Photographic Optics."  While long out of print this can generally be
found in any university quality library ($80 OPC if you can find it).

John Toeppen

http://home.pacbell.net/toeppen/
_______________________________________________________________
Kenneth Luker
Marriott Library Systems and Technical Services
KLUKER@xxxxxxxxxxxxxxxx


------------------------------