data interpolation given a 2d parameterization (terrain skinning) is nothing at all like recovering a general 3d surface from scattered points. the 3d version is a solved problem, but it's not done simply (look into the siggraph proceedings from early-to-mid nineties for a succession of methods, if you need scaring).

a reasonable hack for molecular rendering is to have your vertex generator output a pmesh for each atom surface as a sphere, and let the rendering engine's hidden surface removal solve the intersection problem for you at render time, if you've got the graphics horsepower. you could winnow out a few of the faces by doing inside/outside testing for all vertices of a face, discarding any having all vertices inside, but the render engine's going to do that for you anyway

-- w

Will Welch
e [EMAIL PROTECTED]
c (415) 794-9172
t (415) 546-9172
a 650 Second Street #203
  San Francisco, CA 94107

On Apr 19, 2005, at 3:36 AM, Miguel wrote:

Yes, I think it is difficult to extract triangles from vertices ubt it is
possible (if I understand correctly the problem).

Yes, is possible, but the solution has eluded me.

Are all the vertices in the same plane ?

No

That is, do they form a polygon ?
If so, my idea would be to use a recursive methode :
- find 3 consecutive vertices that make a triangle included in the polygon
(there is always a set of vertices that matches this condition). Not only
the angle must be convex, but there is also a check to see if the segment
between vertices 1 and 3 intersects an other segment.
- remove the vertice 2 from the polygon to get a polygon with one vertice
less and reapply this method.

Yes.

Sounds similar to Delauney Triangulation. See, for example:

 http://astronomy.swin.edu.au/~pbourke/terrain/triangulate/


Miguel



-------------------------------------------------------
This SF.Net email is sponsored by: New Crystal Reports XI.
Version 11 adds new functionality designed to reduce time involved in
creating, integrating, and deploying reporting solutions. Free runtime info,
new features, or free trial, at: http://www.businessobjects.com/devxi/728
_______________________________________________
Jmol-developers mailing list
Jmol-developers@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/jmol-developers



------------------------------------------------------- This SF.Net email is sponsored by: New Crystal Reports XI. Version 11 adds new functionality designed to reduce time involved in creating, integrating, and deploying reporting solutions. Free runtime info, new features, or free trial, at: http://www.businessobjects.com/devxi/728 _______________________________________________ Jmol-developers mailing list Jmol-developers@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/jmol-developers

Reply via email to