VAO slower than not using it

ileben · 2009-10-16T04:26:13

I've installed the new driver from nVidia's site (191.07). My graphic card is 9800GTX. I am using the VAO functions through the GL_ARB_vertex_array_object extension. In the scene that I'm rendering there is a total of around 160,000 triangles, split into many objects (meshes) resulting in around 800 VBOs. All I did now is wrap the calls to set up buffer bindings and data pointers to cache them into VAOs and the result is a frame rate drop from 80fps to 55fps. I did check and make sure that the VAO setup bit only happens once per each mesh, so from frame one onwards, only the glBindVertexArray() function is being called. Here is a snippet from my code: if (!meshVAOInit) { glGenVertexArrays( 1, &meshVAO ); glBindVertexArray( meshVAO ); bindBuffers(); bindFormat( shader, format ); meshVAOInit = true; } else glBindVertexArray( meshVAO ); //bindBuffers(); //bindFormat( shader, format ); material->begin(); //Walk material index groups for (UintSize g=0; g<mesh->groups.size(); ++g) { //Render current group TriMesh::IndexGroup &grp = mesh->groups[ g ]; renderGroup( grp ); } material->end(); //unbindFormat( shader, format ); //unbindBuffers(); glBindVertexArray( 0 ); Has anyone got experience with using VAOs on similar hardware? After an hour of googling I still haven't found any information on the performance issues with VAOs.

Graphics and GPU Programming Programming

Started by ileben October 13, 2009 10:24 PM

7 comments, last by modus 14 years, 7 months ago

Yann L

1,806

October 15, 2009 04:43 PM

Quote:Original post by samoth
No experience here, but a quick Google brought me to the OpenGL forums where there's a thread with pretty much everyone agreeing that VAO is 100% the same speed-wise as not using it, making it a total waste of time using.

I don't really agree with this. While not stellar, I have experienced a performance increase using them, of about 10% in CPU limited cases. But this may very well depend on the exact usage scenario. In my case, I was sourcing a large amount of vertex attribute streams, so that the old style per-frame call of multiple glVertexAttribPointer would show a significant impact. I had about 1.2m faces visible in the frame, distributed over about 500 VBOs.

And of course you won't notice any impact if you're not CPU limited. If, for example, you're fragment shader or memory bandwidth limited (which is often the case), then VAOs speedup will be about zero.

modus

148

October 16, 2009 04:26 AM

Yep, and besides the general idea is to expose an opportunity for an optimization, not to mention a potentially significant user convenience.

Look at FBOs - same thing. You could create single FBO and bind all targets to it or create one for each set of targets and bind the whole enchilada on demand. Better? Worse? Dunno. Test and decide for yourself. (D3D has no notion of a Framebuffer object, at least at the API level; but I suppose GL API revolutions happen infrequently enough to warrant the long view where possible; even if, as an outsider, you need a telescope to fully appreciate it sometimes ;-)).

VAO slower than not using it

This topic is closed to new replies.

Popular Topics

Recommended Tutorials

VAO slower than not using it

This topic is closed to new replies.

Popular Topics

Recommended Tutorials

Reticulating splines