Well, i have a program written using vertex arrays and i’d like to upgrade it to use compiled vertex arrays, because i read thy’re faster than normal vertex arrays.

So, afterinitialising this extension, the only thing I added to the display routine was:

glLockArrayEXT(bla, bla)
glDrawArrays(bla,bla,bla) <-I had this before

But the frame rate is the same with or whitout compiled vertex arays, how are compiled_vertex_arrays used???


Compiled vertex arrays can only be a win if
you re-use the same vertex data more than
once. Even so, if transform is not your
bottleneck (i e you have a GeForce, Radeon,
or better) then compiling probably won’t
give you much anyway.

No, compiled vertex arrays can be a win even if transform is not a bottleneck. They can also help if copying data is a bottleneck.

Expect to see a much-improved CVA implementation in a future driver version.

  • Matt