I’m completely against compiling my maps in vertexlight, because there is no single Q3 player out there running on a Voodoo Banshee card in 2003… Having no lightmap is faster, but how can you compare 320FPS and 125FPS. (Those numbers coming from my timedemo results on an nVidia GeForce 1 DDR @16bit color.) 90FPS and up is just fine.
The average video card for a Quake 3 player is between GeForce 3 Ti500 and GeForce 4 4400 (And their ATi equivalents.) Performance is no longer an issue, and people shouldn’t need to use Vertexlight when playing.
Now compare these images:
Which would give the player a cheating advantage? 1 or 2? Definately 2. (vertex and high picmip) You can’t even see the player in shot 1.
I just wanted to get your full writeup on old vertex light, so I have a little backup from the horse’s mouth.
It’s fullbright, it’s no better than gl_modulate in Quake 2.
Wallhacks let you see players behind walls. Vertexlight lets you see players behind shadows. They’re nearly the same cheat, when you think about it.