I am curious about diffraction's role in photographic imaging, if any.
That's my background, with some physics from an engineering
curriculum. SO I hope to understand replies, just lack enough
education to my question myself. I'm not in the sci.optics filed, so I
realize there is some inherent ignorance in my questions.

OK, sorry.

It was a long-winded question. I was kind of looking for an answer
like:

A) I oversimplified a complex subject.
B) Yeah, I nailed it.
C) I misunderstood and made too many assumptions.

It's not for a class, by the way. I couldn't sleep the other day and
that's what I was thinking about. I figured the best kind of Dr. to
talk to about that problem would be here. Born & Wolfe was one of the
books I was reading a year or so ago. I got the big picture but not al
lthe details.

Really all I wanted to know (maybe I can make it fit category A B or
C), were the following, which might better be asked in a photo forum
but for all the old wive's tales, urban legends & krap that gets
passed around there. :O)

1) Am I correct in insisting that pinhole photography works for all
practical purposes solely BECAUSE of diffraction?

2) If Q1 is true, is diffraction fundamental to image formation with a
lens or is it reduced to merely an undesirable side effect to be
controlled like aberrations (I believe diffraction is not an
aberration).

I hope that's more acceptable. If not, I'll keep reading here & there.

Murray

If you in Ann Arbor and want to talk optics over a beer I'm around.  I
think there is also an active OSA group here too but I haven't
attended their meetings.

Cheers,
Philip

Lastly, regarding the fact that a simple lens and a pinhole aperture
both invert an image, is either diffraction or the Fourier transform
the common factor causing the inversion. (I'm not entirely comfortable
with this question, as the phenomenon existed before the 'name'
Fourier transform became available...and thus may only be a model or
explanation for the phenomenon.

Yeah, I'm starting to think whether I'm being narrow minded in
thinking diffraction is responsible for imaging. Geometry, ray
bending,etc alone seemed too simple :O(

Yet diffraction seems to be responsible for more than just limiting
resolution. Maybe it's just wishful thinking on my part!

In article <47DF2DDF.3090...@electrooptical.net>,
 Phil Hobbs <pcdhSpamMeSensel...@electrooptical.net> wrote:
snip
For paraxial rays, you can calculate most diffraction effects using
GEOMETRICAL OPTICS! This is the method made popular by Kogelnik and Li
using ABCD matrices. This technique was developed to analyze laser
resonators. The method is presented in many books about lasers. AE
Siegman's (who post here) book Lasers is one of them. This technique is
insufficient to describe marginally stable resonators.

snip
Of course it isn't only in optical resonators that this is true. What

murrayatuptowngallery@yahoo.com wrote:
Yeah, I'm starting to think whether I'm being narrow minded in
thinking diffraction is responsible for imaging. Geometry, ray
bending,etc alone seemed too simple :O(

Yet diffraction seems to be responsible for more than just limiting
resolution. Maybe it's just wishful thinking on my part!

Wave propagation (and cleverly designed lenses and mirrors that modify it) is
what is responsible for imaging.
Diffraction is really just a name we give to those phenomena in which wave
propagation is noticeably different from the predictions of the ray model.

For most purposes, ray optics is better than good enough, and it's very
simple and (reasonably) intuitive. Ray optics doesn't take into account the
limitations of imaging caused by the finite wavelength of light or by
polarization, and when that matters, we fix it up afterwards using
diffraction theory and other hand methods.

Cheers,

Phil Hobbs

Of course it isn't only in optical resonators that this is true. What
most optical designers like me find extraordinary is that for an
optical system which is "useful" (in other words, not too badly
aberrated), the aberrations and hence the diffraction performance, can
be estimated through ray tracing. In other words, starting with the
assumption that light only travels in straight lines, you can come up
with a very good approximation to the distribution of intensity in the
diffraction pattern in the image plane, which is based on the fact
that it doesn't! And because it is the image plane that you are most
interested in most of the time, the fact that these approximations
fall apart in the intervening spaces is irrelevant. What is
particularly stunning about this is that it made the design of
excellent lenses possible long before we had today's computing power.
The earliest calculations of diffraction patterns were carried out
when rays were traced using 7-figure tables.

Diffraction theory (of imaging) is not merely a method of determining
the part of the image that is different from geometrical imaging. The
usual approach is in the form of a diffraction integral that gives the
field in a plane. If it is a geometrical image plane, you will get the
geometrical image PLUS all the artifacts attributed to diffraction. For
a wavelength approaching zero, you merely get the geometric image.

For paraxial rays, you can calculate most diffraction effects using
GEOMETRICAL OPTICS! This is the method made popular by Kogelnik and Li
using ABCD matrices. This technique was developed to analyze laser
resonators. The method is presented in many books about lasers. AE
Siegman's (who post here) book Lasers is one of them. This technique is
insufficient to describe marginally stable resonators.

This is a valuable technique to study because variations of it can be
used to cover many fields other than resonators. Electric circuits and
optical thin-films are just two of them. See Louis Pipes in the Condon
and Odishaw Handbook of Physics.

Of course it isn't only in optical resonators that this is true. What
most optical designers like me find extraordinary is that for an
optical system which is "useful" (in other words, not too badly
aberrated), the aberrations and hence the diffraction performance, can
be estimated through ray tracing.