09-13-2005, 09:07 AM
I have a question with regards to display lists and optimizations.
First, my code uses display lists. Yet, I'm tempted to use an array style implementation. eg. glInterleavedArrays with glDrawArrays.
Its my understanding that glBegin/glEnd pairs are to be avoided in immediate mode. So, I compile that into a display list. And things run fast.
So, they say I can edit the array. But I figure I can do that with a vertex program using the display list anyways.
I understand the display list puts things into an optimal format and onto the video card. Where the array implementation leaves the data in client space.
But the question is does using the array methods prove to be faster than my display lists? Is my display list going to have overhead of calling glVertex etc.
On more funky question, I assume I cannot put glInterleavedArrays with glDrawArrays in a display list?
Thanks for you patience.. :)
First, my code uses display lists. Yet, I'm tempted to use an array style implementation. eg. glInterleavedArrays with glDrawArrays.
Its my understanding that glBegin/glEnd pairs are to be avoided in immediate mode. So, I compile that into a display list. And things run fast.
So, they say I can edit the array. But I figure I can do that with a vertex program using the display list anyways.
I understand the display list puts things into an optimal format and onto the video card. Where the array implementation leaves the data in client space.
But the question is does using the array methods prove to be faster than my display lists? Is my display list going to have overhead of calling glVertex etc.
On more funky question, I assume I cannot put glInterleavedArrays with glDrawArrays in a display list?
Thanks for you patience.. :)