How to use maofd 
(the maximum allowable on-film deviation) program.

     This program calculates many factors which are useful in 
making macro and hyper stereo pairs, but is not limited to only 
these stereo applications. The program assumes that the two film 
planes (for the two images which make the stereo pair) are coplanar.

     First let me say that one of the primary purposes of the 
program is to calculate that stereobase which will cause the 
maximum allowable deviation to appear in the stereo pair.  This 
does _not_ mean that you _must_use_ this calculated stereobase.  
You may use anything up to this amount of stereobase but nothing 
in excess of it according to the conjecture.  (The conjecture 
seems to be working out very well - pairs taken at maximum 
stereobase are eminently viewable.)  

     This spreadsheet assumes you are using the more usual sort of 
lens, the sort which you focus by moving the whole lens closer to 
or farther from the film.  If you use a front cell focussing lens, 
the effective or operating focal length of the lens will not be 
calculated properly.  In most cases, this creates such a tiny 
error that there won't be a problem.

     There are many columns hidden to make the spreadsheet fit on 
your screen.  You are encouraged to open them up if you want to do 
something special in the way of data entry.  Where possible, inputs
have been moved to column E.  I will give those entry points in
parentheses in the text below.

     The first three entries, which go into columns A (or E8), B, 
and C, are fairly obvious I hope.  They are the focal length of 
the camera's lens(es), the nearest point in the scene you wish to 
photograph, and the farthest point in the scene you wish to 
photograph.  Typically you might be using an SLR on a slide bar 
and that SLR might have a 50 mm lens on it and the scene you are 
photographing might have its closest object 250 mm (~10") 
distant and its farthest object 500 mm (~20") distant.  So you 
would enter, 50, 250, and 500.

     The first output (column D) tells you the best distance 
setting on your lens barrel so that you get the nearest point in 
the scene and the farthest point in the scene equally into focus.  
For 50, 250, and 500, the best distance is 333 mm.

     The second output (column E) tells you how far out the lens 
has been focussed.  Using the numbers from above, you see the lens 
operating focal length is 58.8 meaning that the 50 mm lens had to 
be turned (focussed) out 8.8 mm to give you a proper focus.  

     The next entry is column F (E9) and this is where you enter 
the maximum allowable on-film deviation.  Typically, an audience 
will be able to tolerate an on-film deviation of up to 1/30th of 
the focal length of the lens if they are viewing from the proper 
distance.  If you're using a 50 mm lens, you would then enter 
1.67 mm here.
     The output you get is in column G and it is the stereobase 
that will create the maximum allowable on-film deviation.  To 
reiterate, you don't _have_ to use a stereobase this large; you 
just shouldn't exceed it.  As an example, suppose you have the 50 
mm lens and you've entered your 1.67 mm and you have a near point 
of 500 mm and a far point of 600 mm.  Column G tells you that you 
can use a stereobase up to 91 mm.  Well that's a hyper shot and you 
may not want to have a hyper because it results in "Lilliputism".  
So you may instead choose to hold the stereobase down to 65 mm.  
This is where you start to see the beauty of this maofd approach: 
if you had used the 1 in 30 rule, you would have mistakenly 
limited yourself to a stereobase not exceeding 500/30 or 16.7 mm!

     Column H is where you enter the f/number you wish to use.  
We'll see later that we can also get the program to calculate a 
reasonable f/number for you. 
     The output you get from the f/number input is the effective 
f/number, column J.  As you recall, we had to focus the lens out 
so we need a focus correction factor on the f/number.  The 
corrected f/number is what you would use to do your strobe or 
light meter calculations.  You can go ahead and peek at column I 
but it probably won't be very useful - it's the linear diameter of 
the iris.

     Next come the outputs of columns K and L and this gets 
interesting.  As you stop down a lens, fuzziness due to an object 
being out of focus becomes less of a problem but fuzziness due to 
diffraction becomes more of a problem.  (The first fuzziness is 
due to "geometric optics" and the second one to "physical 
optics".)  If you adjust the f/number (change the entry in column 
H) so that the fuzziness due to one cause equals the fuzziness due 
to the other cause, you have will reach a pretty good happy 
medium.  The resolution values are given in minutes of arc for the 
angle subtended at the eye.  In other words, it tells you how 
fuzzy it's going to look when you view it.  For reference, under 
reasonable conditions a good eye should be able to see 1 minute of 
arc.  So if the fuzziness result is 5 minutes of arc, you're going 
to see some fuzziness under decent seeing conditions.

     The next input is resolution of the film and lens, column M (E10).  
This is where you enter what your camera will actually do with the
same film in it as you normally use to shoot stereo pairs.  Try to 
be realistic here.  A lens may do 100 lpmm under laboratory conditions 
but what does your lens/film combination do under field conditions?
     Column N just shows you what the resolution you entered turns 
out to be in angular terms.

     The next output, column O, gives the summation of fuzziness 
due to all causes: geometric, physical, and film/lens resolution.  

     You can use the spreadsheet's goal-seeking function to set 
column P to zero by changing column H.  This will set the 
fuzziness due to geometric optics equal to the fuzziness due to 
physical optics for an approximate best result.  The goal-seeking 
function will be selecting the best f/number reading on your lens 
barrel for you.  You may be surprised at how fuzzy a tabletop 
scene is if you take the stereo pair from very close.  However, 
remember that if you are half as far away, the scene is not twice 
as fuzzy, so there _is_ a net gain in resolution of object detail.

That's the end of the first section.  However, this spreadsheet 
will calculate more values for you.  

     If you have a stereo camera, you can enter the spacing of its 
lenses in column R (E11) and the spacing of its film gates in column 
S (E12).  These are center to center measurements.  
     Column T will output the current distance to your built-in 
stereo window using these inputs and the current operating focal 
length of the lenses (from column E).  If you fool with this, you'll 
see that changing the focus on the usual stereo camera doesn't 
move its stereo window very much.

     As we know, when we take a close up shot with a stereo 
camera, we will be closer than the built in stereo window and so 
we'll have to use a narrow mask to bring in the window.  One good 
question is how narrow the mask aperture will have to be.  If you 
enter the distance from the left side of your left film gate to 
the right side of your right film gate in column U (E13), you will 
see what your chip width is in column W.  Just for fun, column V 
tells you what you _wish_ your film gate spacing were (in column S) 
(aka E12) so that your stereo window would be where the nearest point 
in your scene is.  It's a sort of measure of how badly you're 
misapplying your camera as built.

That's the end of the second section.  Next is something specifically
for folks who use slide bars.

When you use a slide bar, you generally are using a mono camera.  
Mono cameras have the lens centered over the film, not offset as 
in a stereo camera.  So you have to throw away some film width to 
create the stereo window.  This section calculates how much film 
you will have left if you set the stereo window at the same 
distance as the nearest object in your scene.  To make it work, 
all you have to do is enter the width of your camera's film gate  
in column Y (E14).  Typically this will be 35 to 36 mm in an SLR 
and 54 to 57 mm in a medium format camera.