Robert, I could do either. I would like to try both approaches, but would like to hear people's experience with this.
My gut says that the mesh would look nicer, but that gl_points would be easier. I need to keep a somewhat reasonable framerate. Don't know how bad points are compared to polys. For example, with an average tri-strip length of 3, I could represent 5 points. Multiply that by 100000, and I wonder what would be better on an average reasonable card. Won't get to fool with the hardware and run tests until next week. In middle of move now. Assuming that the rates are comparable, I wonder what looks nicer, and if there are good algorithms around for shrink-wrap meshing. Chris -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Robert Osfield Sent: Tuesday, October 17, 2006 3:43 AM To: osg users Subject: Re: [osg-users] Optimal rendering of point cloud data. Hi Chris, Are you just rendering points, or do you need to mesh the points? Robert. On 10/16/06, Dorosky, Christopher G <[EMAIL PROTECTED]> wrote: > > > Hi, > > A simple redirect to a paper or web site will probably do for this. > > Does anybody know of a good method for rendering non-terrain separate > point clouds? > > I was thinking of two different approaches. > > 1. Fast gl_point system. > 2. shrink-wrapping algorithm to render a mesh of the outside surface. > I don't have a good algorithm for this yet. > > > Can anybody point me to some papers / examples? > > Thanks, > > Chris > > _______________________________________________ > osg-users mailing list > [email protected] > http://openscenegraph.net/mailman/listinfo/osg-users > http://www.openscenegraph.org/ > > _______________________________________________ osg-users mailing list [email protected] http://openscenegraph.net/mailman/listinfo/osg-users http://www.openscenegraph.org/ _______________________________________________ osg-users mailing list [email protected] http://openscenegraph.net/mailman/listinfo/osg-users http://www.openscenegraph.org/
