Olympus-OM
[Top] [All Lists]

[OM] Re: [lens] Re: Shift lenses

Subject: [OM] Re: [lens] Re: Shift lenses
From: "John A. Lind" <jlind@xxxxxxxxxxxx>
Date: Tue, 06 Jul 2004 04:47:21 -0500
tOM,
Yes . . the FOV (in degrees) does shrink as the film plane is shifted 
off-center from the lens.  I was still struggling with a pile of equations 
working it geometrically using the principle that the three interior angles 
of a triangle always add to 180 degrees (acute or obtuse).  Then I realized 
it could be shown via constructive geometry and all became very clear . . . 
that my brain had run so fast the first time that I had actually visualized 
it, but only for a very fleeting moment before leaping so quickly to its 
conclusion that the path that got me there was lost.

I will describe it with an example.  Draw an isosceles triangle with a 
height of arbitrary length "b" and a base with length of three times some 
arbitrary length "a" (this trisects the base from the get-go).  Draw two 
lines from the apex to the two points that trisect the base.  We now have 
three triangles:  one isosceles in the middle bounded by two identical 
obtuse triangles on each side . . . all having a horizontal base with 
length "a".  The length of the leg of this center isosceles equals the 
squareroot(.25*a^2 + b^2) [Thank You Pythagoras!].

Now copy the length of one of the legs of the smaller center isosceles 
triangle and mark its length down the leg of the larger one we first drew 
starting at the apex.  It will fall short of reaching the base.  Connect 
that end point with the nearest trisection point on the base of the large 
triangle (distance "a" from the end of the base).  We now have another 
isosceles triangle with the same leg length of the center one.  It should 
be obvious now that this new isosceles triangle's base is shorter than the 
base of of the one with length "a" in the middle.  Why?  If you look past 
its base, there is a length "a" at an angle to it that must be, by 
necessity, longer!  Because it has the same length of leg with a shorter 
base, its apex angle *must* be less.  Thus, the apex angles of the two 
obtuse triangles with base length "a" are smaller than the apex angle of 
the isosceles in the middle with length "a" base.  By intuitive logic this 
special case can be generalized.

Quo Erat Demonstrandum:
The Field of View shrinks as the film gate of fixed dimensions is shifted 
off-axis in the image circle.

An additional note.  If you think about it, the FOV as defined by the fixed 
film gate dimensions must shrink because its center is being moved farther 
from the apex of the image cone.

Thankyou tOM for challenging me on this . . . seriously . . . it forced me 
to walk through it in detail to demonstrate it with certainty.  I make too 
many quick and dirty mental analyses and it frightens me at times that I 
haven't thought it through thoroughly enough.

-- John Lind
[who must now explain the sudden TP shortage and stubby Crayolas to the 
rest of the household.]

At 02:29 AM 7/6/04, tOM wrote:

>Ackshully, I think I was dead wrong. The angle viewed is less at the
>outside of the image circle than at the middle, for a given negative
>width.
>
>tOM


==============================================
List usage info:     http://www.zuikoholic.com
List nannies:        olympusadmin@xxxxxxxxxx
==============================================

<Prev in Thread] Current Thread [Next in Thread>
Sponsored by Tako
Impressum | Datenschutz