 |
Unreal Tournament 2004 Demo Even though I own the full game, I decided to use the demo for benchmarking. This is because SantaDuck’s CTF benchmark is working well in the demo, but not so in the full game. To work around the single CPU bug, I have the sound turned off on both machines. Besides that, all settings are set as high as the demo allows it. It seems like the PowerBook’s performance is limited by the CPU speed, since increasing the resolution doesn’t slow it down. The MDD on the other hand suffers more from increased resolutions, so the Radeon 9000 Pro doesn’t handle UT2k4 as good as the 9700 Mobility. Jedi Knight II This game is highly CPU bound, and changing the resolution doesn’t affect performance at all. Thus I have chosen to use this game to compare the CPU speed and comparing FSAA speeds. All settings are set to high and the timedemo is the jk2ffa timedemo hidden within the game. This is the test where the dual 867 and the single 1.33 heads out for real. Even if JKII crashes if you try to enable SMP, it still gains some speed advantage from the second CPU. When turning off the second processor with Apple’s CHUD tools, you can see that the 1.33 GHz processor in the PowerBook plays the game about 31% faster than the 867 MHz processor in the MDD. It’s not quite the 53% that it would be in a perfect world, but it’s not that far from it. As well as with the other games, we can see that the 9000 Pro is the first card that gets enough when we increase the FSAA quality. The 9700 doesn’t even break a sweat when increasing the quality.
|
|
