Test Setup

CPU:	Intel Core 2 Quad Q6700 G0 Stepping
Motherboard:	Asus Rampage Formula X38
Video Card:	eVGA 8800GTS (G92) 512MB Core: 670MHz Memory: 972MHz
Harddisk:	3 x 320 GB Seagate 7200.10 in RAID 0 Array
Power Supply:	Corsair HX1000W
Software:	Windows Vista 64-Bit, 177.83 Forceware

Performance

The performance of the kit was tested with six separate tests in order to give a well rounded idea of the real world performance buyers can expect. The tests were performed at a range of clock frequencies at each CL: 3, 4 and 5. To test the overclock potential of the kit at each CL, voltages from 1.8V all the way up to 2.4V were employed in order to indicate the require voltage a user can expect in order to achieve such frequencies.

The results are summarized in the following table and graphs.

A-Data Vitesta Extreme Edition 4GB 800+ MHz CL4
CPU Clock Memory Ratio	Memory Speed	Memory Timings	Everest Read	Everest Write	Everest Latency	Quake 3 Timedemo	3DMark 2001SE	SuperPi Mod 1M
10 x 266 2:3	400 MHz	5-5-5-15 1.8V	6658 MB/s	5012 MB/s	65.8 ns	732.4 fps	41422	17.56 s
10 x 326 2:3	489 MHz	5-5-5-15 2.1V	8712 MB/s	6347 MB/s	56.7 ns	912.7 fps	54411	16.02 s
10 x 355 2:3	533 MHz	5-5-5-15 2.4V	8864 MB/s	6548 MB/s	54.2 ns	924.9 fps	56560	15.10 s
10 x 266 2:3	400 MHz	4-4-4-12 1.8V	6898 MB/s	5122 MB/s	64.8 ns	748.1 fps	42499	19.92 s
10 x 307 2:3	460 MHz	4-4-4-12 2.1V	7701 MB/s	5522 MB/s	59.7 ns	801.0 fps	49898	17.97 s
10 x 327 2:3	490 MHz	4-4-4-12 2.4V	8103 MB/s	6231 MB/s	56.4 ns	860.4 fps	52371	16.48 s
10 x 222 2:3	333 MHz	3-3-3-6 1.8V	5557 MB/s	4598 MB/s	88.2 ns	600.2 fps	36613	26.04 s
10 x 256 2:3	384 MHz	3-3-3-6 2.1V	7012 MB/s	5911 MB/s	64.9 ns	756.9 fps	47412	17.82 s
10 x 266 2:3	400 MHz	3-3-3-6 2.2V	7314 MB/s	6011 MB/s	62.3 ns	812.1 fps	48454	17.12 s
10 x 272 2:3	408 MHz	3-3-3-6 2.4V	7413 MB/s	6021 MB/s	61.7 ns	819.2 fps	49349	17.03 s

While this kit overclocks quite linearly with regard to the supplied voltage, the graph makes apparent the small increase in clock frequency afforded by loosening the CL rating to 5. This makes running the kit at CL5 almost pointless as the small increase in MHz is hardly worth the more relaxed timings. I should finally note that, while these results are a good indication of expected performance, no two kits will perform identically and that these results are not guaranteed.