|
AMD
System |
|
CPU |
AMD Athlon64 4000+ Processor |
|
CD/DVD |
NEC 3540A |
|
Motherboard |
DFI LanParty NF4 SLI-DR |
|
Memory |
2 x 1GB OCZ PC4000EB, Corsair XMS4000 |
|
Hard Drive |
3x Seagate Baracuda V 120GB |
|
Video Card |
BFG 7800GTX |
|
Sound Card |
|
|
NIC |
|
|
All testing was done with the OCZ and Corsair at their tightest possible timings, and optimal voltages for maximum overclocking. The system was powered by an OCZ PowerStream 600W PSU. Sandra scores are averaged.
Performance @ 400mhz |
|
|
|
Sandra unbufered |
Sandra buffered |
| Corsair XMS4000 2x1GB |
|
|
|
|
|
|
|
|
|
|
|
Starting off at 400mhz, the OCZ and Corsair were pretty similar, with the OCZ PC4000EB having a slight edge in timings, giving it slightly better Sandra scores.
Maximum performance |
|
|
|
Sandra unbufered |
Sandra buffered |
| Corsair XMS4000 2x1GB |
|
|
|
|
|
|
|
|
|
|
|
Pushed as far as they were willing to go, the differences between the OCZ and Corsair were much larger. The Corsair maxed out at 556mhz, and to get there, it required a hefty voltage increase, and 3-4-4-8 timings. I was only able to push the system FSB 3mhz farther with the OCZ, for a top speed of 561mhz. However, the OCZ required just 2.7V, and only needed a reduction in Cas latency to 3 in order to reach maximum speed. This major timings advantage over the Corsair gave it a much higher score in the Sandra unbuffered test, which is much closer to real world applications than the buffered test.
