[tech-3d] Re: Stereo Base Calc. with $20.00 Handheld

["Michael K. Davis" <zilch0@xxxxxxxxxxxx>]

Hi Allan,

>Date: Sun, 19 Nov 2000 14:19:00 +1100
>From: "Allan Griffin" <agriffin@xxxxxxxxxx>
>Subject: Re: Stereo Base Calc. with $20.. Handheld
>
>Reference posting by Mike Davis and further discussion on the tech-3d List
>ref his recommended calculator and calculation formula:
>
>Hi Mike and All,
>
>Mike has cleared up some points that were not clear to me in his piece.
>Thanks Mike.  Mike says that his advices about calculating the stereo base
>are entirely irrelevant if you can't vary the separation.  It seems to
>follow then that those of us who use standard stereo cameras, including
>custom RBT and the like for work in the mid-range, are not included in the
>group to which the posting was directed.  Even in the field of twin-cameras
>used for "wide-base" work, where the stereo base can be set according to any
>value or formula one wishes, I find that the question of MORE or LESS stereo
>"bite" is quite subjective.

You're so right!  And it seems the longer you work with stereography, the
more depth you crave.  I'm certain that I am undergoing a "depth
desensitization" as I spend more time with shooting stereo - I'm developing
a "tolerance."  I told a friend of mine recently that lately, I have gotten
so hooked on just staring into my viewer for long periods, that pretty
soon, I'll have to get a T-shirt made that reads, "Just say NO to Stereo!"
Heee-heeeeee!


>Steve Spicer (a good friend of John Bercovitz and myself - and whose website
>you quote) has expressed the opinion on several occasions that my hyper
>stereos are more to his taste when I use LESS relative separation (flatter).
>However, Boris Starosta says today that he finds that my stuff somewhat on
>the FLAT side (it appears that he would prefer MORE relative separation or
>stereo bite).  Since I am not the one proposing the theory in question, I
>don't feel moved to make any of my stereo pairs more available on the web
>than are at present available.

It is SO subjective that demonstrations of method are almost hopeless, but
one should be able to demonstrate a method's ability to deliver
consistency, even if that look is not found pleasing by some individuals
(I'm speaking of methods for twin-camera base calculations again...)

>About the only other observation I feel like throwing into the ring is my
>belief that, in the mid-field or central half or so of the bell-curve,
>stereo is more appealing when the apparent scale approximates life as we
>know it.  It is for this reason, I believe, that the use of a base close to
>that of our natural eye separation, produces the best psychological response
>across the board.  Standard stereo cameras with no variable base handle this
>problem very well indeed.  

Great point you're making here.  I can't get my twin cameras any closer
than 165mm.  With the 65mm lenses I use, that causes me to hit my head
against the 100% of MAOFD ceiling (an on-film deviation equal to 65mm/30)
when my Near point is still 13.63 feet away (with Far at Infinity).
Speaking of subjective preferences, as I wrote earlier, I prefer 80% of
MAOFD, but I can't achive this until my Near subject moves out to 17 feet
(again, with Far at Infintity.)  So, yes, I dearly miss the ability to
shove my lenses closer together!

>Most people who leap into the twin camera arena
>where the minimum base is usually over twice that of the human being,
>produce stereos featuring "little" people until they learn that the distance
>to the "near" point needs to be considerably increased unless an appropriate
>increase in the local length of the lenses used is made (usually to 135mm).

Well said!  That's exactly what I suffer - even at 80% MAOFD, achievable
with Near at 17 feet and Far at Infinity, my 165mm minimum base made my dad
say recently, while looking at a pair I shot of a family walking ahead of
me on a trail, "Is that a family of midgets?"  So, even with my limited
experience shooting 3D, I have already hit this wall!  It goes away, as you
say above, when the Near is far enough away (I've not experimented with
longer lenses.)

>
>I apologise to Mike for even suggesting that I would like to see sample
>stereo pairs produced with one or more of the cameras, rigs or other that he
>or John Bercovitz would use when applying the theories espoused in his piece
>which he says is based on the premiss that you have an infinitely variable
>stereo base available to you in the real world.  If not, I think he is
>saying:  "This is not for you".
>
>In conclusion, can someone on List try and tell me in layman's language
>whether calculations made with the unit Mike recommends are for actual use
>and not simply a theoretical exercise?  

Well, I'll give it a go by just saying that all I did was migrate the
formula Bercovitz uses onto the handheld calculator, tossing in a
compensation factor to accomodate viewer FL mismatch to camera FL.  So, it
is VERY practical within the limits of your equipment and resources.

>If so, how and on what kind of
>camera gear is the infinitely variable base obtained over the full range
>from very close to very far?

Using a single camera on a slide bar one REALLY CAN achieve small bases, by
shooting, moving, and shooting again.  With one or two cameras, you can
make your base as large as you like.  In aerial photogrammetry (spelling?),
enormous bases are used to deliver the depth desired (of distant terrain.)

I feel like you probably know all this, so I'm wondering if I
misinterpreted the question.  Still, I like your point about the advantage
of fixed-separation cameras in the mid-field.  I seldom have the chance to
shoot single-camera (for smaller bases than 165mm) because my subjects are
seldom static.  Any movement between first and second exposure and it's toast!

Mike




-------------------------- eGroups Sponsor -------------------------~-~>
Create your business web site your way now at Bigstep.com.
It's the fast, easy way to get online, to promote your business,
and to sell your products and services. Try Bigstep.com now.
http://click.egroups.com/1/9183/3/_/520353/_/974701324/
---------------------------------------------------------------------_->