I am using GL_COMPILE... what is the difference between GL_COMPILE and GL_COMPILE_AND_EXECUTE?
-menasius
"Quitters never win...
winners never quit...
but those who never win and never quit, are idiots"
Display Lists in Opengl, the anti-optimization at high poly counts?
Author
also it doesnt make sense that it would loose so much for simply compiling one list and displaying it once as opposed to just placing the glBegin...glEnd block in as is. If nothing else the methods should yield the same performance, thats what I gathered from the documentation on the web and in books about them.
-menasius
"Quitters never win...
winners never quit...
but those who never win and never quit, are idiots"
-menasius
"Quitters never win...
winners never quit...
but those who never win and never quit, are idiots"
Hi menasius,
The difference between GL_COMPILE and GL_COMPILE_AND_EXECUTE is that with compile & execute, OpenGL commands are executed as they are encountered (i.e. it will draw to the screen, or whatever) as well as compiled (GL_COMPILE just compiles them to a display list).
I find your situation pretty damn weird to say the least. As you seem to know, all a display list does is capture all commands called in it, along with vertex data values etc at that moment in time, and store them in an efficient manner for fast output. BTW, it is completely valid to perform tranformations within a display list, and can even be more efficient than doing them outside, as only the resultant matrix of transformations is saved in the list (i.e. it doesn''t do the calculations each time it is called).
Methinks I''ll be going home to check this out with my terrain engine (which uses display lists) on my TNT2 now... just as soon as I fix my installation of Win2k
-------------
squirrels are a remarkable source of protein...
The difference between GL_COMPILE and GL_COMPILE_AND_EXECUTE is that with compile & execute, OpenGL commands are executed as they are encountered (i.e. it will draw to the screen, or whatever) as well as compiled (GL_COMPILE just compiles them to a display list).
I find your situation pretty damn weird to say the least. As you seem to know, all a display list does is capture all commands called in it, along with vertex data values etc at that moment in time, and store them in an efficient manner for fast output. BTW, it is completely valid to perform tranformations within a display list, and can even be more efficient than doing them outside, as only the resultant matrix of transformations is saved in the list (i.e. it doesn''t do the calculations each time it is called).
Methinks I''ll be going home to check this out with my terrain engine (which uses display lists) on my TNT2 now... just as soon as I fix my installation of Win2k
-------------
squirrels are a remarkable source of protein...
actually, i meant transformations with variables in them. that's what i should say, with passing variables through. doesn't that slow it down?
supposing you had glvertex(xvar, yvar, zvar). this wouldn't be fast in a display list. right?
sorry if this has nothing to do with your code, menasius, since you don't even have transformations, and less likely any variables passed through.
(oh, and if i'm totally wrong about the variable-passing thing totally, i'd like to know, cuz then, i'd use it in my game. i have the perfect application for it.)
a2k
Edited by - a2k on 4/10/00 1:40:55 AM
supposing you had glvertex(xvar, yvar, zvar). this wouldn't be fast in a display list. right?
sorry if this has nothing to do with your code, menasius, since you don't even have transformations, and less likely any variables passed through.
(oh, and if i'm totally wrong about the variable-passing thing totally, i'd like to know, cuz then, i'd use it in my game. i have the perfect application for it.)
a2k
Edited by - a2k on 4/10/00 1:40:55 AM
This topic is closed to new replies.
Advertisement
Popular Topics
Advertisement
