...and I'd like to hear some comments, if anyone actually
has use for it. Never sure about that last part... :-)

Before people begin shouting: the macros are meant to
do some basic stuff with an array of two dimensions, filled
with vertices. It doesn't generate that one, but builds upon
it, by generating the required triangles, calculating the normals,
do some little modifying (hence the name).

So, basically you take your rectangular grid of vertices,
and the macros work with it.

Anyways, even if no one has use for it, enjoy!

Tim

PS: The link is in my signature...
--
Tim Nikias
Homepage: http://www.digitaltwilight.de/no_lights/index.html
Email: Tim### [at] gmxde

Post a reply to this message

"Tim Nikias" <tim### [at] gmxde> wrote in message
news:3e24b7e5$1@news.povray.org...
> ...and I'd like to hear some comments, if anyone actually
> has use for it. Never sure about that last part... :-)
>

Very cool stuff. The docs are quite nice. My suggestions for a future
release would be optimized quad splitting and a few examples of array
creation for those who might not immediately grasp the concept.

One very small problem is that your macros write out unneeded duplicate
vertices for cylinder and torus shaped meshes. Not a big deal, but it adds
just a small amount to the parse time.

 -Shay

Post a reply to this message

> One very small problem is that your macros write out unneeded duplicate
> vertices for cylinder and torus shaped meshes. Not a big deal, but it adds
> just a small amount to the parse time.

That's a side-effect I can't easily handle. If that information had to be
inserted somewhere into the array, it wouldn't be as easy of use as it is.
IMHO
its a minor drawback, but I was thinking about implementing some macro which
will just join a side (for a cylinder, for example), so that one does not
need to
have the vertices twice.
And, as mentioned somewhere in the docs, the idea was to have something very
easy to visualize without a graphic interface, and at some point I want to
implement
UV-Mapping. It's easier to imagine if you've got your rectangular grid and
work
with that, at least for me.

> Very cool stuff. The docs are quite nice. My suggestions for a future
> release would be optimized quad splitting and a few examples of array
> creation for those who might not immediately grasp the concept.

Thanks.
I understood the part of "grasping the concept", but that quad-splitting...
I've
implemented it so that the shorter cross-section will be used for the
triangulation,
instead of an always lowerleft-upperright (or reverse) splitting. What's
wrong
with that?

--
Tim Nikias
Homepage: http://www.digitaltwilight.de/no_lights/index.html
Email: Tim### [at] gmxde

Post a reply to this message

"Tim Nikias" <tim### [at] gmxde> wrote in message
news:3e258e36$1@news.povray.org...

: Tim Nikias:
:   That's a side-effect I can't easily handle. If that
:   information had to be inserted somewhere into the
:   array, it wouldn't be as easy of use as it is.

You wouldn't need the information in the array, but you would need a switch
perameter in each macro. I agree with you that this may very well be less
intuitive for the user.

: Tim Nikias:
:   but that quad-splitting...
:   I've implemented it so that the shorter cross-section
:   will be used for the triangulation, instead of an always
:   lowerleft-upperright (or reverse) splitting. What's
:   wrong with that?

Sorry, missed it. A slightly better way of doing it might be to select the
dividing line which is closest to a point representing the average of the
four corners. I doubt it would make a noticeable difference, however.

 -Shay

Post a reply to this message

Shay wrote:
> Sorry, missed it. A slightly better way of doing it might be to select the
> dividing line which is closest to a point representing the average of the
> four corners. I doubt it would make a noticeable difference, however.

For that matter, I could easily just create four instead of two triangles
and
use the average of the four corners as a vertex for each of the four
triangles. This way of looking for closest connection I found somewhere
on the net, and when using smooth_triangles, most issues with this quad-
splitting are solved anyways.

If you've got any other suggestions, I'm open for almost anything,
and you know where to find me...

Regards,
Tim

--
Tim Nikias
Homepage: http://www.digitaltwilight.de/no_lights/index.html
Email: Tim### [at] gmxde

Post a reply to this message