triangle strip cache size

OpenGL OpenGL: Advanced Coding
21

Interesting.

There is another factor that should be considered.
Calling DrawElements with a few indices vs Calling DrawElements with many indices (degen triangles used)

The number of cash misses is potentially the same for both.

CVA is no longer modern so don’t expect anything.
VBO is the only choice in modern GL.
On occasion you may see a boost with immediate mode.

22

jwatte, good precision. But what are you calling IHV ? And the ‘pre/post T&L cache’ you used is confusing me: Are they the same things ?

Also note that the cache size you specify is the “actual” cache size, not the “effective”
cache size you may have heard about.

Finally, I didn’t have heard about the “effective” cache size.