🔒 Locked

Quick Question on Carmack's OpenGL code

Started by Chrysaor Jul 12, 2004 at 1:41 AM 7 replies 1.4k views

Original Post

Jul 12, 2004 at 1:41 AM

Hey, I was looking at carmack's quake 2 and quake 3 code, and i noticed that he puts a q infront of all opengl functions. for example: qglBegin(), qglEnd(), etc... I was wondering why he does this. I know that in qgl_win.c (in the win32 folder for the quake 2 source code), he gets the proc addresses for all those functions, so i know why they work, but i dont get why he does this. Any ideas?

Ainokea

Jul 12, 2004 at 1:50 AM

q for quake?

Luctus

Jul 12, 2004 at 2:09 AM

I believe the q-abstraction layer was put there so the engine could handle more than one kind of renderer (OpenGL, Glide, software..), just change the function pointers and you're set to use another renderer.

exa_einstein

Jul 13, 2004 at 3:09 AM

why he does it? maybe cross-platform purposes? (is there some qgl_unix.c?) or do-it-yourself dogma?

_the_phantom_

Jul 13, 2004 at 11:06 AM

at the time Quake2 was written there was no garenteed OpenGL support on any system, so it needed an easy fall back to software path, using function pointers is the simplest way todo this and afaik the only way todo it with C.

Sunray

Jul 13, 2004 at 2:22 PM

Because he cares about compile times. Including windows.h/gl.h in every file leads to huge comple times. Q2 compiles on 10 seconds on my machine.

Promit

Jul 13, 2004 at 2:27 PM

You'll also notice that implementations for all of those functions exist in the software renderer. (They might not be named exactly the same, but that's where the function pointers go.)

Topic Locked

This topic has been locked by a moderator. New replies are not allowed.