18.5.2 Using Vertex Arrays

Vertex arrays are available in OpenGL 1.1. They offer the following benefits:

• The OpenGL implementation can take advantage of uniform data formats.

• The glInterleavedArrays() call lets you specify packed vertex data easily. Packed vertex formats are typically faster for OpenGL to process.

• The glDrawArrays() call reduces subroutine call overhead.

• The glDrawElements() call reduces subroutine call overhead and also reduces per-vertex calculations because vertices may be reused. Be aware that using indexed vertices may introduce other problems with cache misses if the access pattern corresponding to the indexes is irregular enough. Indexed arrays are often most useful with implementations which perform the vertex processing on the CPU and may tend to degrade the performance of systems which have fast geometry processing in the accelerator if they become bottlenecked by the memory subsystem.

• Use the EXT_compiled_vertex_array extension if it is available. This extension allows you to lock down the portions of the arrays that you are using. This way the OpenGL implementation can DMA the arrays to the graphics adapter or reuse per-vertex calculations for vertices that are shared by adjacent primitives.

If you use glBegin() and glEnd() instead of glDrawArrays() or glDrawElements() calls, put as many vertices as possible between the glBegin() and the glEnd() calls.