Regarding vertex processors
I have a small question regarding the way in which information is sent to the GPU. For eg: I do the following:
a) set MVP matrices, draw an object(i.e send everything to the shader A)
b) Change MVP matrices, draw another object(i.e send everything to the shader B)
Finally, Do a SwapBuffers()
Now, lets say in (a), the object has 10 vertices and in (b) there are 10 vertices. And also, lets say there are 10 vertex processors. How exactly is the decision of sending vertices to different vertex processors taken? What are generally the ideas beind telling a vertex processor to execute a vertex shader?
Can anyone please provide some links as to where i can read more about it?
Let's not do that. It is hardware specific detail and not something you need to know.
Originally Posted by myk45
PS. it is a safe bet that no relevant hardware even has any such specialization - making your question rather meaningless (ie. what do you mean by "vertex processors" and on what specific hardware).
OpenGL doesn't specify any of this - it will be entirely hardware-dependent, and you can bet that whatever decision - say - NVIDIA make will have a good chance of being completely different to the one AMD make (and vice-versa).