Testing Methodology
I based my testing today on the memory testing methodology that was developed for Icrontic’s recent PC2-8500 memory roundup. It will be slightly different, however as this is the first four gigabyte kit to find its way into the lab. Due to the memory ceiling in 32 bit operating systems, the 64-bit version of Windows Vista Ultimate was installed for testing. The hardware configuration is also slightly different and listed below.
- CPU: Intel Core2Quad Q6600 @ varying frequencies.
- Motherboard: Asus P5K-E based on the Intel P35 chipset.
- Graphics Card: ATI Radeon X850XT.
- Hard Drive(s): 1x Seagate 7200.7 hard drive.
- Optical Drive(s): Pioneer SATA DVD-RW drive.
- CPU Cooling: Noctua NH-U12P with 53CFM Noctua NF-P12 fan.
- Memory Cooling: Corsair Dominator “Airflow Fan”.
- Case: Antec mid-tower with rear 120mm 45CFM exhaust fan (side left open)
- Power Supply: Corsair TX750
- OS: Windows Vista Ultimate (64-bit)
Environmental controls were also put in place. Ambient temperature was maintained at 18.5°C for all tests and was not permitted to deviate more than +/- 0.5°C.
For our thermal testing, I placed an external temperature probe sitting on the vacant memory slot between the two populated ones. Full load tests were conducted with the Corsair “Airflow fan” present and not present.
Finding a way to fully load four gigabytes of RAM was actually somewhat challenging. Running a custom Prime95 test has always been my poison of choice—unfortunately, a single instance (even the multi-threaded v25) can not allocate more than ~1.7GB of RAM. To address this, I ran two instances of v25 with two cores being utilized each. Each instance chewed up about 1.7GB and left just enough for Windows Vista to operate without swapping to virtual memory. This method of “loading” the memory is used for both power consumption and thermal tests.
Articles RSS