Let’s get on to the tests. The specifications of the test rig:
NVIDIA nTune
First, I ran NVIDIA’s System Stability Test, which stresses the CPU, RAM, PCE-E Bus, hard disk and video card, for thirty minutes. Stressing the system in this fashion, which simulates the environment a computer would encounter while running a game, helps determine how well the RAM, as a crucial piece of the computer’s hardware, performs under the stressful conditions gamers impose on their computers every day. The System Stability Test passed with flying colors, holding true to the Crucial strategy of marketing Ballistix as the memory line for PC enthusiasts of all stripes.
Then I ran the Tune System utility, which takes a lot of the time, arcane knowledge and guesswork out of overclocking by adjusting the speeds of the memory, front side bus and PCI Express bus over a period of three hours until it comes up with fastest stable combination. Although this method, like any other overclocking method, relies on the stability of all the major components in your computer to achieve results, it is by far the most efficient way to test the overclocking capabilities of this motherboard.

The chart above shows the four (out of twelve) areas where Tune System utility made changes to the memory timings; all four changes increased performance. The most significant by far is the thirteen-cycle change in the Row Address Strobe (tRAS), which is the number of clock cycles it takes for the computer to find a particular row of data in memory before sending a command to open it. However, all four decreases occur in related areas of memory access and it’s clear from these changes that the Ballistix has a great deal of overclockable performance potential.

SiSoft Sandra
To translate those changes into real-world bandwidth numbers and compare the changes to the previous generation of Ballistix RAM, I ran the SiSoft Sandra Memory Bandwidth benchmark, testing both the standard and overclocked RAM timings. The results: