Test Setup
We attempted to keep a consistent CPU speed through testing, however due to limitations in chipset straps between CPU front side bus and memory bus when overclocking, a closest fit was applied where necessary.
Memory Speeds Tested
Kingston HyperX KHX9600D2K2/2G- PC2-800MHz (2.3V, lowest latency, 3-3-3-8-18-2T) 2.93GHz/266MHz FSB;
- PC2-1200MHz (2.325V, rated timings, 5-5-5-15-31-2T) 2.93GHz/266MHz FSB;
- PC2-1232MHz (2.4V, maximum overclock, 5-5-5-15-32-2T) 2.964GHz/270MHz FSB;
Corsair CM2X1024-9136C5D - PC2-800MHz (lowest latency, 3-3-3-6-14-4tWR-2T) 2.93GHz/266MHz FSB;
- PC2-1142MHz (rated timings, 5-5-5-15-30-5tWR-2T) 2.93GHz/266MHz FSB;
- PC2-1200MHz (maximum overclock, 5-5-5-15-33-6tWR-2T) 2.93GHz/266MHz FSB;
Corsair CM2X1024-10000C5D - PC2-800MHz (lowest latency, 3-3-3-6-10-3tWR-2T) 2.93GHz/266MHz FSB;
- PC2-1250MHz (rated timings, 5-5-5-18-15-6tWR-2T) 2.981GHz/271MHz FSB;
Intel Core 2 Extreme X6800 (operating at 2.93GHz - 11x266); 2x Kingston HyperX PC2-9600 KHX9600D2K2/2G, Inno3D nForce 680i SLI, Nvidia Geforce 7900GTX; Seagate 7200.9 200GB 7200RPM SATA 3Gbps hard disk drive; OCZ GameXStream 700W power supply unit; Windows XP Professional Service Pack 2; DirectX 9.0c; Nvidia Forceware 94.24 WHQL; Nvidia nForce 680i SLI 9.53 WHQL;
Memory Bandwidth
-
Kingston HyperX KHX9600D2K2/2G (800MHz)
-
Corsair CM2X1024-9136C5D (800MHz)
-
Corsair CM2X1024-10000C5D (800MHz)
-
Corsair CM2X1024-9136C5D (1142MHz)
-
Kingston HyperX KHX9600D2K2/2G (1200MHz)
-
Corsair CM2X1024-9136C5D (1200MHz)
-
Kingston HyperX KHX9600D2K2/2G (1232MHz)
-
Corsair CM2X1024-10000C5D (1250MHz)
-
5274
-
5027
-
5133
-
5152
-
4953
-
5179
-
4995
-
5291
0
1000
2000
3000
4000
5000
MB/s
-
Kingston HyperX KHX9600D2K2/2G (800MHz)
-
Corsair CM2X1024-9136C5D (800MHz)
-
Corsair CM2X1024-10000C5D (800MHz)
-
Corsair CM2X1024-9136C5D (1142MHz)
-
Kingston HyperX KHX9600D2K2/2G (1200MHz)
-
Corsair CM2X1024-9136C5D (1200MHz)
-
Kingston HyperX KHX9600D2K2/2G (1232MHz)
-
Corsair CM2X1024-10000C5D (1250MHz)
-
7854.0
-
7574.0
-
7639.0
-
7182.0
-
7236.0
-
7208.0
-
7342.0
-
7405.0
0
1000
2000
3000
4000
5000
6000
7000
8000
MB/s
-
Kingston HyperX KHX9600D2K2/2G (800MHz)
-
Corsair CM2X1024-9136C5D (800MHz)
-
Corsair CM2X1024-10000C5D (800MHz)
-
Corsair CM2X1024-9136C5D (1142MHz)
-
Kingston HyperX KHX9600D2K2/2G (1200MHz)
-
Corsair CM2X1024-9136C5D (1200MHz)
-
Kingston HyperX KHX9600D2K2/2G (1232MHz)
-
Corsair CM2X1024-10000C5D (1250MHz)
-
2585.0
-
2294.0
-
2206.0
-
2355.0
-
2466.0
-
2474.0
-
2489.0
-
2497.0
0
500
1000
1500
2000
2500
MB/s
The Kingston's performance at 800MHz low latency is superb. We managed to find timings that really set this stuff on fire (quite literally, they ran very hot!) and the performance showed in both Sisoft Sandra and Everest.
The write speeds were also consistently very high with the Kingston according to Everest, although the Corsair PC2-9136 running at PC2-9600/1200MHz speeds offers a slighter faster result than both the Kingston running at the same 1200MHz speed and overclocked at 1232MHz. In fact, the Kingston at stock speeds performs the slowest of all the settings and modules tested.
Memory Latency
-
Kingston HyperX KHX9600D2K2/2G (800MHz)
-
Corsair CM2X1024-9136C5D (800MHz)
-
Corsair CM2X1024-10000C5D (800MHz)
-
Corsair CM2X1024-9136C5D (1142MHz)
-
Kingston HyperX KHX9600D2K2/2G (1200MHz)
-
Corsair CM2X1024-9136C5D (1200MHz)
-
Kingston HyperX KHX9600D2K2/2G (1232MHz)
-
Corsair CM2X1024-10000C5D (1250MHz)
0
10
20
30
40
50
60
70
80
90
nanoseconds (lower is better)
-
Kingston HyperX KHX9600D2K2/2G (800MHz)
-
Corsair CM2X1024-9136C5D (800MHz)
-
Corsair CM2X1024-10000C5D (800MHz)
-
Corsair CM2X1024-9136C5D (1142MHz)
-
Kingston HyperX KHX9600D2K2/2G (1200MHz)
-
Corsair CM2X1024-9136C5D (1200MHz)
-
Kingston HyperX KHX9600D2K2/2G (1232MHz)
-
Corsair CM2X1024-10000C5D (1250MHz)
-
55.7
-
60.4
-
60.8
-
56.8
-
56.2
-
56.6
-
56.4
-
55.6
nanoseconds (lower is better)
Sisoft Sandra seems to over exaggerate the latency of the Corsair memory, but Everest still shows both Corsair DIMMs as having a slightly higher memory latency at 800MHz "low latency timings", even though the timings set are very similar. We can attribute it to the Corsair not liking a 1T Command Rate on our Inno3D nForce 680i SLI, even though it was able to achieve some extremely low 3-3-3-6 timings.
Want to comment? Please log in.