What has become almost commonplace here at nV News is the use of Fraps to
measure the performance of the card under various games. While the results can
never be a true apples-to-apples comparison, the results can give you an idea of
what to expect from the card in terms of real-world performance given the
dynamic and unscripted nature of gaming.
Halo
Bungie's Best-Selling Xbox title has finally been ported over to the
platform for which it was first intended. Bungie rewrote some rather huge
chunks of code in order to have the game run on a variety of systems. As such,
various Pixel Shader paths were implemented in order to provide gamers with any
type of video card to enjoy Halo to its fullest. We'll be examining the
performance of the various Pixel Shader versions by making use of the
game's built in
timedemo feature. Note that the timedemo performance results are
not indicative of in-game results as the vast majority of the timedemo is in the
in-engine cinematic format. Each test was run three times with the average of
the three runs being posted. Sound was enabled for all tests. Performance
between runs barely swayed more than 1%.
Halo Timedemo Results
AA and AF were not enabled during testing as Gearbox's lead rendering
engineer, Shawn Green recommends against it. The eVGA e-GeForce FX 5950 Ultra
doesn't seem to gain too much performance by dropping pixel shader versions.
This has to do with the fact that the game is capable of providing the GeForce
FX architecture with partial-precision hints to help boost it's PS 2.0
performance. There isn't any reason to not make use of PS 2.0 on the 5950 Ultra.
Max Payne 2
Max Payne 2. With so much hoopla surrounding the launch of its predecessor,
Max Payne 2 surprised more than a few gamers by appearing on the shelves of
their local game retailer. Max Payne 2 remains focused on the single player
market by provided a fairly involved story all while pushes some of the
prettiest triangles seen yet.
Max Payne 2 Benchmark Settings
Max Payne 2 Benchmark Results
I just picked the game up yesterday to include it in the article. As such,
I've only played for a few minutes. Benchmarking consisted of killing the first
two enemies in the first level (Part 1, Level 1) and then following the same
path with the various settings. Max Payne 2 is highly configurable and is
capable of controlling the AA and AF features of the card. Therefore, I tested
the game with the AA and AF sliders set in application mode. AA and AF only
begin to take their toll on performance at 1600x1200. The game remains extremely
fluid at that speed and I believe that it will be the setting that I will
continue to use to complete the game.
F1 Challenge '99-'02
F1 Racing. Hardcore F1 are a crazy lot. Believe me, I've been there on three
separate occasions. The passion these people have for the sports runs in their
blood. EA's F1 Challenge '99-'02 has found a permanent spot on my hard drive and
it just doesn't want to let go. I thought it would be cool to see the 5950's
performance in this game.
F1 Challenge Benchmark Results
Enabling AA is a godsend in this game. It eliminates all the banding
typically found on billboards or along the track the car advances. There's no
other way to play this game. A 3-lap race in Austria where I started last and
climbed up to first within one and a half laps was used as a timedemo of sorts.
I tried to be as consistent as possible when passing, and I believe the results
reflect that. 1600x1200 with all the eye-candy enabled is simply marvelous. 6X
and 8X FSAA nudged the performance just under my minimum threshold for
acceptable performance in a crowd.
