when creating a display list that is going to be compiled, do I have to worry about what commands I use to put the data in to the display lists?
what I mean is, is there any performance difference between using something like a vertex array, or anything else, to using to just declaring the indiviual vertices by calling glVertex for each vertex.
Could somebody please let me knwo the most optimal way to fill a display list.
thanks
I think that when you use a display list, the normal methods of speading rendering up don’t apply. For example, I think that vertex arrays don’t add to performance. Anyone know if things like using triangle strips vs. triangles matters?
Yes, triangle strips and fans increase performance because you are not resubmitting the same vertex data over and over again for each triangle.
I know they do out of display lists, but I thought that display lists, since they were compiled (pre-proscessed) it would make those improvements almost nill…
The data format might be optimized, and as the method of transfering data over to the gpu/driver is taken care of, then it dont matter weather you use glBegin,glEnd / glDrawElements / VAR or what.
Primitive type is probably NOT optimized. So when using a display list, it’s still better to use Tri strips than normals triangles.
Nutty