Quantcast
BROWSE ARTICLES BY CATEGORY
ATI Radeon X800 Pro 256 MB
 
Author:
Editor:
Sponsor:
Published:
Brian Kristensen
Kurtis
ATI
Sep. 8, 2004
Need for Speed Underground

Need for Speed Underground is the only game we benchmark with that does not support any type of recording or playback features. Since the game is popular and has plenty of eye-catching special effects, we still wanted to benchmark with it. To do this, we played three one lap races on the Olympic Square map in a stock Toyota Celica. FRAPS was used to gather the average FPS throughout each race and the results were averaged for each test. The average FPS recorded for each race does not tend deviate more than 5%.

Because we were actually playing the game and not a recorded demo, there was much more than just graphics performance being analyzed. Besides the graphics, the AI for the cars and traffic, as well as the physics engine stress the CPU and memory as well. Just like the Far Cry benchmarking, this does a better job at evaluating the game's overall performance of the entire system, but this leads to bottlenecks and the performance of the graphics card alone is much harder to see. Even with Anti-Aliasing and Anisotropic Filtering enabled, the difference in performance between the cards will be limited. Another limiting factor is the fact that NFSU only allows for a maximum resolution of 1280 x 1024.

Need for Speed Underground
(Show All Graphs)
No AA / No AF 4x AA / 8x AF
ATI Radeon X800 Pro
XFX GeForce 6800 GT
PowerColor Radeon 9800 XT
Leadtek WinFast A380 Ultra
37.81
38.54
37.36
38.14
0
(FPS)
56
 
 
No AA / No AF 4x AA / 8x AF
ATI Radeon X800 Pro
XFX GeForce 6800 GT
PowerColor Radeon 9800 XT
Leadtek WinFast A380 Ultra
36.46
38.46
35.67
36.71
0
(FPS)
56
 
 

All four cards perform so closely here that a comparison is not really necessary. Graphics is obviously not a bottleneck for NFSU. Even at 1024 x 768 with AA and AF disabled (not graphed) the average FPS doesn't exceed 40 FPS. Every card at 1280 x 1024 with 4x AA and 8x AF allows for a good experience.

 
<< Previous
Page 8 of 13
Next >>
Page 1: Introduction
Page 2: The Card
Page 3: The Card (cont.) / The Bundle
Page 4: Test Setup
Page 5: Benchmarks: Doom 3
Page 6: Benchmarks: Half-Life 2
Page 7: Benchmarks: Far Cry
Page 8: Benchmarks: Need for Speed Underground
Page 9: Benchmarks: Splinter Cell
Page 10: Benchmarks: Call of Duty
Page 11: Image Quality
Page 12: Overclocking
Page 13: Conclusion

4 User Comments
1 - Posted by Rich on September 9, 2004 - 8:45 am

I am curious why you chose not to install SP2 for the testing. Any idea if there is a performance difference between SP1a and SP2 for XP?

2 - Posted by Brian on September 9, 2004 - 11:16 am

Testing of the cards began prior to the release of SP2. After the upcoming review of the XFX 6800 GT, the test system will have SP2 installed.

3 - Posted by Rich on September 10, 2004 - 8:42 am

Makes sense, I would be a bit interested to find out if there are any performance differences between SP1a and SP2.

4 - Posted by Brian on September 10, 2004 - 8:49 am

http://www.hardcoreware.net/reviews/review-237-1.h...

SP1 and SP2 seem to constantly switch places in different performance tests. Overall SP2 is slightly faster but definitely not noticeable unless you are comparing bench #s.

Add Comment

To add a comment without being a member, you may omit the password field, but you must enter your name (or nickname) along with your comment. * Denotes required fields.

Username: *


Password: (optional)
(Remember my login information: )

Comment: *


What is 1+2?: *