Geez I need help.
After spending all yesterday writing a program to create tristrips, I
realised that was about 1000 times more work than needed for what I want to
do.
Now, I can rotate a triangle - specifically the points of a triangle - in
povray so that all points lie in the plane z=0. I did this with Margus
Ramst's v_reorient macro. No worries there. But I need a pascal routine to
do the same thing so I can print the triangles.

Problem: how the hell do I code the built-in vaxis_rotate function in
pascal?
I know it's simple - I did it several times last year for another project.
But I just can not get it to work today, even after reading the matrix
tutorial. I'm confusing myself between rotating an object, and rotating a
position vector, around an arbitrary axis. I just can not get this, and I
did linear algebra at Uni several years ago. Depression's not helping, I
guess :-(

So, I now couldn't give a stuff about the theory. I just need some
pseudocode, or pascal/c source to do this...please?
And yes, I have the pov source, just can't follow multi-function c code at
the moment.

When you can write a program to create tristrips in a couple of hours, and
then have trouble rotating a few points around a couple of straight lines,
you know you're sick in the head <attempted grin>.

Dave.


"jarbee" <d_j### [at] yahoocom> wrote in message
news:3bae4270@news.povray.org...
> I've been thinking it possible to create paper models of some of my
meshes,
> by printing out the individual triangles on paper and glueing them
together.
> It would probably be easier to do this with triangle strips rather than
> individual triangles, but that's a patience issue :-)
>
> How would it be possible to 'unwrap' a triangle mesh, so that the position
> of each triangle could be known, and a plan created to allow the paper
> triangles to be stuck together in the correct positions?
>
> I have seen triangle to strip converters etc, but I can't work out how to
do
> this properly. And I can't seem to find any tutorials or algorithms on
this?
> But would it be similar to how Steve Cox's UVMapper works?
>
>

Post a reply to this message