Comment below.
At 12:54 AM +0000 12/3/02, olympus-digest wrote:
>Date: Mon, 02 Dec 2002 18:33:16 -0500
>From: "John A. Lind" <jlind@xxxxxxxxxxxx>
>Subject: Re: [OM] Mind Bender (intermediate focal length?)
>
>At 10:54 11/28/02, Joe Gwinn wrote:
>
> >I was thinking also of the lenses used on view cameras, where the field of
> >view (that is, image circle) vastly exceeds the film size. Ditto, shift
> >lenses for 35mm. So, I was looking for an algorithm that included only
> >data that was true of all lenses, regardless of design, which leaves only
> >focal length.
>
>Cannot be done. A practical lens design has an "acceptance angle"
>independent of focal length which is the limiting factor for the FOV of the
>lens' _image_circle_. You are correct that technical camera lenses
>normally have an image circle diameter significantly larger than the film
>diagonal to allow for the adjustments moving the circle around across the
>film plane. It also means the 80mm Mamiya Sekor acceptance angle for an
>M645 must be much larger than the 85mm Zuiko acceptance angle for an OM.
True enough. That in a nutshell is why I focused on focal length.
>However, when one speaks (or writes) of a system FOV, the limiting factor
>is the film dimension, if it's properly designed with the lens' image
>circle being larger (even if slightly more) than the film dimension. Even
>in 35mm small format and medium format systems, the image circle is usually
>a little larger than the film gate, especially for shorter
>lenses. Reason? It mitigates cos^4 falloff in the corners by placing the
>bulk of this falloff outside the film gate. The tradeoff is the extra
>light that doesn't end up in the photograph potentially bouncing around
>inside the lens and reducing contrast, or worse yet causing aperture flare.
I think we are talking past each other. Everything you say above is true, but
it butresses my argument, which isn't likely the intent.
In your posting on magnification, you start with the formula for paraxial rays
through thick lenses, 1/f = 1/u + 1/v, where u is the distances from object to
front node, v is the distance from back node to film, and f is the lens focal
length. This is what I had in mind as well. I would claim that the
"intermediate" focal length is that focal length whose magnification (computed
according to your analysis) is the geometric mean of the magnification of the
larger and smaller focal lengths whose intermediate is sought. This will be
close or equal to the geometric ratio of the focal lengths. (I don't have time
to work the math out right now, but perhaps in a few days.)
Joe Gwinn
< 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 >
|