Benchmarking:
Benchmarking consisted of running the test system through two memory-centric benchmarks, one gaming benchmark, and then running a test to see how calculations are aided by different memory configurations. As a point of reference, the Crucial Ballistix 2GB PC2-6400 DDR2 kit was put head to head with the recently reviewed Transcend 2GB PC2-6400 DDR2 kit. Both kits offer the same capacity and default speeds, but it will be very interesting to see what the improved timings (4-4-4-12 versus 5-5-5-15) of the Crucial kit can provide in terms of performance.
The Tests:
» Everest Ultimate Edition 2006 Memory Benchmark
» SiSoft Sandra 2007 Memory Benchmark
» 3DMark06
» Super Pi / Mod1.5 XS
All tests were executed on both sets of memory in two conditions. First, with the memory at the default speed of 800 MHZ (DDR2), and then with the memory overclocked to 900 MHz (DDR2). The Transcend timings were 5-5-5-15 at both speeds, while the Crucial Ballistix was run at 4-4-4-12 at 800 MHz and 5-4-4-12 at 900 MHz.
Everest Ultimate Edition 2006 Memory Benchmark:
The Everest Ultimate Edition 2006 Memory Benchmark provides four sets of results: memory read speed (in MB/s), memory write speed (in MB/s), memory copy speed (in MB/s), and memory latency (in ns). Higher values are desirable for memory read, write, and copy speeds, while lower values are desirable for latency.
As you can see from the data above, the lower latency Crucial kit generally held a slight lead in each test and at each speed. I was unsure what kind of performance boost to expect from the tighter timings, but as you can see it equates to a fraction of a percent in these results. The only place where the Transcend kit squeaked out a victory was in the default speed "Copy" benchmark, and despite repeating the test a few times the results stood up. However, once overclocked the Crucial kit regained the lead.
SiSoft Sandra 2007 Memory Benchmark:
SiSoft Sandra 2007's suite of benchmarks includes one that provides two results related to memory performance. The values are provided in units of MB/s, and higher values are better.
As you can see from the data above, this benchmark follows the lead set by the Everest benchmark and the Crucial kit scores better by a small margin across the board.
|
|
