Re: Computer Compositing

[T3D Bill Stratemeyer <wwstrat@xxxxxxxx>]

john bercovitz wrote:
> Bill Stratemeyer writes:
> > If I had a DXF file of a tera fold knot
> > I could set up a test with it as this object might not give the
> > viewers much preconceived shape information?

> That's different from a trefoil knot?  If it's the same as a trefoil,
 
  Well, that's what my spellchecker called it (tera fold).  I'm sure the 
object I am thinking of is a terafoil.  It was from my memory of a recent
post, subject "Gordonian Knot".
  The real question, is this a suitable object for this kind of test?
After thinking about it I have decided to do a new stereo pair with
several primitive geometrical shapes and see what shows up and take it from
there.
 I was having difficulty deciding what type of objects to use, because I'm
not sure what kind of distortion we will see.  I downloaded John B's gif 
depicting the expected results from what the geometry tell us will happen.
And I agree that an object might squash but I don't understand why we
should see any vertical or horizontal size change. 
   Changing deviation without changing the actual image size on print
or parallax tells me that the only thing that has changed is the
object's image position on the retinas so changing only the angle of view
for that object and affecting only depth as compared to the rest of the
scene.  This doesn't happen in the real word so it is a challenge to
think this out first then see if the results match my opinion of what
are the most powerful 3D Que.'s.
   
  I will render and modify some images over the weekend to share with 
anyone interested.

 Thanks for your indulgence and advice.

Bill Strat...


------------------------------