All the memory modules were tested at their rated latencies and voltages and we enabled their Intel XMP profiles where available. The exception is for the OCZ Reaper which was not stable at its specified 7-8-8-24; instead we had to run it at 8-8-8-24. We also included a reference (1.65V) DDR3 1600MHz memory module from last year's shootout (Kingston's HyperX DDR3 1600MHz Dual-Channel Memory Kit - KHX12800D3K2/4G) to compare and contrast against these lower voltage memories.
For the overclocking segment, we increased the frequencies gradually while putting the modules through a stability test. Voltages and latencies were adjusted accordingly to get the best overclocked frequencies. Finally, for the temperature testing, we looped the memory modules through 3DMark Vantage for at least three rounds before measuring the external temperature of the heat spreader using a laser thermometer.
The following system was used to test all the memory modules:
- Intel Core i7-875K (2.93GHz)
- ASUS P7P55D-E Premium (Intel INF - 22.214.171.1245, Intel Matrix Storage Manager - 126.96.36.1993)
- NVIDIA GeForce GTX 260 (ForceWare 258.96)
- Seagate 7200.10 SATA 200GB HDD
- Windows 7 64-bit Ultimate
Our benchmarks consisted of some synthetic tests and two modern games:
- PCMark Vantage - PCMarks and Memories Suites
- SPECviewperf 10 - proe-04 and sw-01 viewset tests
- SiSoft Sandra 2010 Lite (ver 16.67) - Memory Bandwidth and Latency tests
- Crysis Warhead - Gamer profile, 1280 x 1024 resolution, 4x AA
- Far Cry 2 - Ranch Small Map, 1280 x 1024 resolution, No AA, DX10, High Settings for Render and Performance