Testing (continued):
Sisoftware Sandra Lite 2008.1.12.34:
SiSoftware's Sandra was used to analyze other facets of the memory's performance. Values for integer (in MB/s), float (in MB/s), and latency (in ns) are shown below, with higher values being better for integer and float, while lower values are better for latency.
This test is another close one, but the tighter timings seemed to help produce some better numbers at stock timings. The results at 1500MHz show that relaxing the timings on the OCZ Technology 2GB PC3-10666 ReaperX HPC DDR3 Dual Channel Memory Kit takes away its edge, as it puts up lower scores in all three categories.
3DMark06:
3DMark06 was used to compare each kit to see if either one makes a big difference in gaming scenarios. Memory really shouldn't have a huge impact in this graphical test.
I really didn't expect a huge difference with 3DMark06, and at 1333MHz the OCZ Technology 2GB PC3-10666 ReaperX HPC DDR3 Dual Channel Memory Kit got six more Marks with the rated timings than the Aeneon kit. It is better than nothing, but you can see that the tighter timings help gain just a few extra marks. And as you can see, overclocking doesn't get you much, either.
Temperature:
As a final, less than scientific test, I took a look at the heatspreader and heatsink. The plan was to get a thermal probe on or next to one of the memory chips, but the interface between the chips and the heatspreader was too good. It is the tightest fitting and best installed heatspreader I have seen. On top of that, it is more robust than your typical heatspreader, so even without the heatsink up top, the cooling might be better than the typical thin strip of Aluminum. Then while searching the Internet to see what other people were doing with their kits, I saw how destroying the modules was very possible if you tried to get "under the hood".
In an open system, with no case fans running, the memory was stressed again while a thermal probe was used to check temperatures. MemTest86 was executed for an hour to try to get the chips nice and hot, but the heatsink barely got warm to the touch. With an ambient temperature of 25C, the skin of the heatspreader reached only 29 C, and the fins of the heatsink registered only 27C.
|
|
