Optimising the way mesh data is sent to the GPU.

include/SurfacePatchRenderable.h

@@ -10,6 +10,9 @@
 
 namespace Ogre
 {
+	//IDEA - If profiling identifies this class as a bottleneck, we could implement a memory pooling system.
+	//All buffers could be powers of two, and we get the smallest one which is big enough for our needs.
+	//See http://www.ogre3d.org/wiki/index.php/DynamicGrowingBuffers
 	class SurfacePatchRenderable : public SimpleRenderable
 	{
 	public: