Paul F Brewster (p.f.brewster++at++larc.nasa.gov)
Tue, 6 Jul 1999 16:10:53 -0400 (EDT)
> Paul F Brewster wrote:
> >
> > Hello, I'm not exactly a Performer expert but I do know a little bit about
> > graphics... I have a couple questions that I was hoping somebody could
> > help me out with...
> >
> > I'm trying to do some finite element work, so I want to show a fairly
> > detailed model with a finite element mesh on top of it. What I am doing is
> > just using perfly to view an inventor file that has the object data in it.
> > The main problem I have is what is the best way to draw a finite element
> > grid on top of an object?
> >
>
> Perhaps you could try a glPolygonMode call in a draw callback and draw
> your geometry twice, this will require lot's of state overrides to
> ensure the lines are white or whatever. I'd just add the geometry in
> question to the scene graph twice with callbacks around the second
> instance. Use polygon offset or stencil to ensure the lines 'win' the
> depth test.
How is this different than just drawing it in wireframe mode? I think
Performer converts quads to triangles when it loads the inventor file. Is
this true, and is there a way I can turn that off?
>
> > I can do it as a collection of quads and then draw them as a wireframe,
> > but the quads get divided into triangles, which gives me extra lines on my
> > mesh, that's bad. Is there a way to turn that off?
> >
> > I can draw them as individual lines. Problem, there are so many lines that
> > I can't get very good interactive rates. It takes me twice as long to draw
> > the line mesh as it does to draw the object. Since it is a finite element
> > mesh I don't really want to do any decimation type stuff, I need it that
> > detailed. Is there a fast way to draw lines? Treat them as degenerate
> > triangles maybe?
> >
> > Another problem, while loading a collection of lines in Inventor format
> > into perfly I run into a problem. Some lines come out missing. I at first
> > thought it was being blocked by the solid model behind it, but that is not
> > the case. This happens with just the line set. What happened to the
> > missing lines? I have heard it is a culling problem, but I am not sure how
> > this applies... Can anyone help?
>
> Could be a loader bug but it's difficult to tell. You sure the .iv file
> was kosher?
Yeah, the .iv file is fine. I've checked it out several times...
This archive was generated by hypermail 2.0b2 on Tue Jul 06 1999 - 13:11:00 PDT