Testbed Configuration and Methods
We performed out performance tests on the following platform:
- Intel Core 2 Duo E6400 CPU (Conroe, 2,4 GHz);
- Thermaltake Blue Orb 2 CPU cooler;
- ASUS Commando (Intel P965) and Gigabyte GA-N680SLI-DQ6 (nForce 680i SLI) mainboards;
- EVGA e-GeForce 8800 GTX graphics card;
- Seagate Barracuda 7200.9 80 GB HDD;
- HIPER Type-R HPU-7S730 PSU;
- MS Windows XP Pro SP2 OS.
Our testing methodology is intended to show in the most extensive way the memory modules functionality in different modes.
First, we checked if they correspond to the claimed specifications (frequency, memory timings and voltages) in the Intel P965 (ASUS Commando) and nForce 680i SLI (Gigabyte GA-N680SLI-DQ6) platforms. I would like to say right away that the testing participants passed this check flawlessly.
Secondly, we tested the performance of these solutions in work modes typical for inexpensive memory kits. For this purpose we performed a set of benchmarks on ASUS Commando mainboard with the voltage set at 2.1V. In this case we checked the minimal timings in DDR2-667 and DDR2-800 modes and the maximum frequency with 5-5-5-15 timings settings. For a more illustrative comparison, we have also included the results of the simplest memory kit as a reference ?we took a pair of Samsung M378T2953CZ3-CE6 (DDR2-667) modules.
Thirdly, we tested our memory kits with a more appropriate voltage setting for solutions of this category: 2.4V. We used the same mainboard as a test platform and measured the maximum frequencies with different timing settings, namely 3-3-3-8, 4-4-4-12 and?-5-5-15.
And finally, the fourth part of our test session would be a similar check performed on an abit IN9 32X-MAX mainboard with Command Rate set to 1T and?T.
