Please register or login. There are 0 registered and 1515 anonymous users currently online. Current bandwidth usage: 326.30 kbit/s December 11 - 04:04pm EST 
Hardware Analysis
      
Forums Product Prices
  Contents 
 
 

  Latest Topics 
 

More >>
 

    
 
 

  GeForce FX, just special FX? 
  Jan 30, 2003, 09:30am EST 
 

Benchmark results, 3Dmark2001se


By: Sander Sassen


Naturally we did not have all four cards at our disposal so we used PowerStrip to clock the GeForce FX 5800 Ultra down to GeForce FX 5800 clockspeeds, which meant lowering both the core and memory frequency to 400MHz. The Radeon 9700 Pro was given the same treatment and operated at Radeon 9700 speeds, so we could also measure the Radeon 9700' performance. From the 3Dmark2001SE graph we can clearly see that the GeForce FX 5800 Ultra takes the performance crown and is nearly 5% faster than the Radeon 9700 Pro. The gap between the GeForce FX 5800 and the Radeon 9700 is smaller though and measures just 3.3% with the GeForce FX having the advantage.

Naturally both cards had some reserves left in terms of performance by increasing the clockspeed of both the core and the memory. The Radeon 9700 Pro could easily be clocked up to 380MHz core and 345MHz memory which yielded scores higher than those of the GeForce FX running at default clockspeeds. But naturally the GeForce FX could also be coached into running a little faster and we managed about 535MHz core and 545MHz memory clockspeed reliably, which yielded the highest scores overall. But since were comparing cards running at stock speeds here, and not cards running overclocked, this was just an exercise on our part.

We did discover however that if the GeForce FXs core was kept at stock speeds and the memory was clocked up, it performed better in comparison to the Radeon 9700 Pro with the same increase in memory clockspeed. This does indicate that the GeForce FX core is hungry for more bandwidth, whereas the Radeon 9700 Pro isnt so much. The theoretical bandwidth of the Radeon 9700 Pro is larger per definition as it uses a 256-bit memory bus, whereas the GeForce FX uses a 128-bit one, although running at a higher clockspeed. Well see in the other benchmarks if this indeed becomes an issue at higher resolutions where bandwidth starts to play an important role.



1. Introduction
2. Features and benefits
3. Benchmarks and more
4. Benchmark results, 3Dmark2001se
5. Benchmark results, UT2003 Standard
6. Benchmark results, UT2003 AA and AF
7. Conclusion

Discuss This Article (12 Comments) - If you have any questions, comments or suggestions about the article and/or its contents please leave your comments here and we'll do our best to address any concerns.


Rate This Product - If you have first hand experience with this product and would like to share your experience with others please leave your comments here.

 

    
 
 

  Related Articles 
 
 

  Newsletter 
 
A weekly newsletter featuring an editorial and a roundup of the latest articles, news and other interesting topics.

Please enter your email address below and click Subscribe.