Jlw wrote, and in English I might add, adn that's no boolean sheet,

>Very interesting Robert.  Have you encountered any good stuff on
>perspective projection.  Unless I FIND something someday, looks as
>though I may have to reinvent the wheel.  As I may have mentioned
>at one time, my longest standing, unfinished program is an architectural
>CADD Program that I first started in Fortran 77 on my 512k Mac.  I've
>made VERY slow progress with it, BUT SOMEDAY!  Particularly with FB^3s
>potential.
>
>I haven't had time to review Robert P's 3d Demo in detail, but it MUST
>contain some "perspective" concepts.  Heh, Robert?
>
>TIA,
>
>Joe Wilkins

http://freespace.virgin.net/hugo.elias/ contains what he claims is a
perspective correct method on mapping any random picture onto a rectangle
in perspective. Basically his outline uses iterations of vanishing points
until a single pixel is derived, then move onto the next. Like a lot of
micro rectangles.

I have read that what is known as "affine" texture mapping of triangles is
speedy, but not accurate, and the for true perspective correct texture
mapping, one needs to use a quadrilateral, though I don't believe this is
necessarily true as the multitude of 3D apps would suggest otherwise.

Mr Purves demo is nice, absolutely, but in the presence of such competence,
I prefer to throw on the waders, and derive my code from the manure of
which I know.

That is, write my own. I don't understand anybody else's! :)  If I can
grasp a principle, then I don't need the C code to get a going, but if I
don't then I port in danger.

What relevance this has, I know not.

As to Derek's suggestion of OPEN GL, I hope that people  do not forget
QuickDraw 3D, which is a much better API in my opinion, for the higher
level things. Very soon I will have a demo, if someone doesn't beat me to
it, of Lambert (flat) , Gouraud , and Phong shading.  Not in QD3D, just in
FB 3.

Purves can do Ray Tracing. :)  Thanks in advance, R.

Robert