[MF3D.FORUM:1169] Re: Camera design

[Richard Rylander <rlrylander@xxxxxxx>]

Tom Hubin wrote:

> There are ways to tilt the film and make the magnification constant. I
> think one way is to image 1:1. But that is not very useful to most
> folks.
>
> The other is to use an afocal imaging system. That produces a constant
> magnification, independent of object distance. A tilted object still
> creates a tilted image but the magnification is constant. This still
> meets the Scheimpflug condition.

I think you  mean "telecentric", not "afocal".  Afocal systems are literally
"without focal length", like astronomical telescopes with infinite
conjugates.  Telecentric systems have the entrance pupil and/or exit pupil
located at infinity to produce constant magnification for varying subject
distances - extremely useful in metrology applications.  A telecentric
system has a definite plane of focus normal to the optic axis of the lens -
satisfying the Scheimpflug condition still requires that the subject plane,
lens plane and image plane all intersect in a single line.

> Why all this discussion of tilts? I must have ignored the original. Is
> this to avoid using a pantoscopic lens or decentering the taking lens to
> image a tall building without having the top of the building image
> smaller than the bottom of the building?

The objective is to avoid the need for extremely small apertures to produce
sharp images of tilted subject planes, not to straighten out tall
buildings.  In stereo photography, perspective effects are fine, even
desirable, but the main goal is to keep everything in the image as sharp as
possible.   Tilting the lens would be the best way to do this since it
avoids the diffraction and motion blurring of long shutter speeds needed
with small apertures.  The problem is that large tilts need lenses with
large image circles.  I was just mentioning that you could get the same
sharp image of an oblique plane without requiring a large image circle by
tilting the film instead.  The drawback with tilted film is that it does
alter the perspective slightly, but in most cases this perspective change
would not be noticeable or objectionable.

Richard Rylander