Re: Delaunay Triangulation

Hi Desiree,

I think you are seeing the extra triangles that VisAD's
display code generates to connect all the triangles into
one long polytriangle strip.  These connecting triangles
have zero area and are invisible with filled triangles.

An alternative is to generate triangle edge lines directly
from the Delaunay topology, as in the main() method of, about lines 1013-1126.  Note this code is
generating the lines as a UnionSet of Gridded3DSets with
manifold dimension = 1, whereas you probably want to generate
the lines in a LineArray.

Bill Hibbard, SSEC, 1225 W. Dayton St., Madison, WI  53706
hibbard@xxxxxxxxxxxxxxxxx  608-263-4427  fax: 608-263-6738

On Mon, 9 Oct 2000, Desiree Hilbring wrote:

> Hi Bill and all others,
> I hope you remember me,
> I am using the delaunay triangulation of VisAd for constructing a
> Java3D shape3D and displaying this in my own application.
> My goal is to display a terrain-surface.
> The user selects a range of 3D-Points which I want to display as a
> surface.
> Unfortunately the 3D-object I get, does not look the way, I expected.
> There is a triangle edge in my surface, which is shared by 3 triangles.
> That is not what I want, I am looking for a smooth surface, with only 2
> triangles per edge. So my questions are:
> 1. Did I use VisAd properly (I attached my class)
> 2. I get my 3D-Shape directly form Visad, then I take the GeometryArray
> from this shape. After this I change y and z because my
> axes-representation is different. Then I put this Geometry in a new
> Shape3D and displays it in my own Display-class GIS3DView. Is it possible
> that this might have an influence in the wrong appearance.
> 3. Or is it a bug?
> I attached my TerrainLayer class and two screenshots, where you can see
> the funny triangle in polygon_line mode.
> Thanks for your help in advance.
> Desiree
> o------------------------------------------------------------------------o
> | Desiree Hilbring    Institut fuer Photogrammetrie und Fernerkundung  | 
> |                     Universitaet Karlsruhe, Germany                  |
> |                                                                        |
> |                     email: hilbring@xxxxxxxxxxxxxxxxxxxx             |
> |                             # 0721 6083676                                  
>  |
> o------------------------------------------------------------------------o