Originally posted by Hanners
But what would be the point? The whole idea of a Doom III benchmark is going to be to show performance of boards in Doom III - If the timedemo is doing things differently to the game then it's a waste of time.
A timedemo can just crap out as many fps as possible following some fixed cam-path? Like that flyby thing from ut2003, gamelogic becomes a non issue in that case. The CPU isn't really that important in this case, since it is either fast enough to push 60 ticks per second or it's to slow ( which will give you slowdown as in bullettime