Originally posted by Lonestar
You could consider 3Dmark03 as a game. Actually it is a game, the only difference is that you donít get to control the character. A higher score in 3Dmark means overall better performance. By overall I mean that itís a general estimation how good your system (not only your GP) is. Most games rarely use state of the art rendering techniques which means they get could get rather high score in a particular game but maybe not with 3DMark.
Choose between a GP, which gives you more points in 3DMark and one generates a much lower score, what would you choose? No one here could possibly go for a lower score.
Graphic is graphic, no matter if you control the viewpoint or not. Nvidia is whining!
3dmark is not
a game by the definition that you cannot play
look at Max Payne the game. look at the Max Payne scene in 3dmark2001. do benchmark results between the two corroborate each other? the answer is no. that's as close as 3dmark gets to a game, and even in that case, it fails miserably at predicting actual game performance.
, you keep saying nvidia's argument is all about PS1.4 vs PS1.1.
what do you have to say to my earlier post?
isn't that the point? i'm not going to judge the validity of nvidia's statements, but they are saying that the benchmark game tests are not indicative of future games or future game performance.
it sounds like they were complaining about the implementation of the real time shadows; saying that it is poorly done and that future games will not have code like that.
are you saying that nvidia is lying and that the lighting algorithms used in 3dmark03 ARE efficient, that this type of algorithm WILL be used in future games? all your posts do in fact say that 3dmark03 IS a good FUTURE indicator of game performance. yet Doom3's performance is MUCH better than the "Doom3" game test in 3dmark03.