Testing

The Test System

The memory replaced some OCZ DDR3 PC3-12800. While the Intel P35 chipset has been used on different DDR2 and DDR3 motherboards, the memory naturally can only operate on a motherboard that specifically supports DDR3.

CPUZ

Mainboard

SPD & XMP Profiles

Default 1066MHz

Overclocking proved to be pretty beneficial as always. This kit easily operates 1600MHz with or without XMP profiles. It also scales close to 1760MHz at 8-8-8-24 with some more aggressive motherboard settings, but doesn't out perform 1720MHz at 7-7-7-24. Also, 1760MHz seemed to need a lot more chipset and memory voltage. At 1720MHz, the system didn't need any more voltage other than the 2.1 volts given to the memory.

1600MHz FSB

1600MHz XMP 1

1600MHz XMP 2

1720MHz

The memory's ability to maintain 1720MHz at 7-7-7-24 requiring 2.1 volts might not be too healthy for the average user, but does mean a lot more power for the enthusiast. When it comes to overclocking, the lowest timings by far really offer up some extra power especially in graphic intensive and number crunching programs. Keep in mind, your mileage may very depending on your chipset. I've seen kits achieve a bit higher results when going to the most extreme situations to find optimal voltages to maintain stability.

All benchmarks were performed at 1680x1050 with all settings left to the default installation High settings. Video drivers were also left to default. For comparison, the Crucial DDR3 PC3-12800 was tested along with some OCZ DDR3 PC3-12800. All memory timings were matched for the best results.

The first part of my tests always include what I consider the staples of testing one's hardware. Sisoft SANDRA's memory bandwidth, mathematical, and multi-media start off the tests followed by the tried and true 3DMark.

As per typical memory tests, the Ballistix kit's memory bandwidth was compared to the OCZ as well as against it's own XMP profile settings. This way you can see just what to expect from the different memory timings and another comparable kit.

There is very little difference between the kits in terms of memory bandwidth. Naturally, the lower XMP 1: 7-7-7-24 timings out perform the XMP 2: 8-8-8-24 timings. Lower latencies always offer better performance. That often equates in to more frames per second as well as better program crunching.

Got News? Send 'em in!