Corsair PC2-6400C3 Meets AMD’s AM2 ProcessorsWed, May 24, 2006 - 12:00 AM
Final Thoughts and Conclusions
It is obvious from our testing that the memory controller on AMD’s AM2 processors is critical in system performance when it comes to a number of applications. For this article we were trying to get the highest gaming scores and our efforts proved to be very successful. Starting out with PC2-5400 memory running CL4 timings our base score was 134.0 fps. After moving up to PC2-6400 memory with CL3 timings and increasing our memory capacity from 2GB to 4GB the score increased to 159.1 fps — a 25.1 fps increase or 18.7%.
The next time you build a system keep this in mind. I always hear people saying “buy what’s cheap, because memory all comes from the same place in China/Taiwan”. That is hardly the case. Enthusiast memory companies like Corsair spend a lot of time and money finding the right combination of IC’s and that is the reason their memory costs more than other companies. They then have to bin the IC’s and modules, which increases costs and they have to do something with the fallout that they already purchased. If you are a gamer/enthusiast you know that a ~19% boost in gaming performance from doing nothing more than tweaking the memory is significant.
So there you have it… If you want the ultimate gaming system you should stick with at least 800MHz memory running CL 3 timings. Having 4GB of memory increases the bandwidth throughput over the same system running just two modules. Right now the Corsair TWIN2X2048-6400C3 memory kit that we used in the review costs a hefty ($453), which means to run 4GB of this memory will set you back a cool $906. If this is out of your price range the Corsair TWIN2X2048-6400C4 memory kit that we used for all of the CL4 testing runs $270 and offers a better bang for the buck when it comes to price versus performance.
Editor’s Note: I know that I didn’t run a ton of gaming benchmarks, but did also run F.E.A.R., Far Cry, Quake 4 and other titles that showed similar performance increase across the board. I used DOOM 3 as the single gaming benchmark in order to get the article out the door, and move on to the next article. Would you rather see 15 games with roughly the same results or another interesting article! Cheers