Test Setup

Test System
CPU:	Intel E6300 Conroe 1.8 GHz, 2 MB Cache
Motherboard:	ASUS P5K3 Deluxe BIOS 0910
Video Card:	PowerColor X800XL Pro 16 PCI-E
Harddisk:	Samsung P80 80 GB
Power Supply:	Ultra V-Power 450W
Software:	Windows XP SP2, Catalyst 8.3

Performance & Overclocking

As soon as the memory is installed, we run CPU-Z. The SPD programming is very relaxed and on the safe side. While the memory is rated for 1600 MHz CL7, the SPD maxes out at 1333 MHz CL9.

The first benchmark run was done at the intended setting. The 1.8V seems to be right on the mark. Getting a higher speed also means raising the voltage. Before we do that, the memory was tested using 1.5V at CL7. This is the default voltage for DDR3. With this much power, the X-Series memory manages 717 MHz (DDR3-1434). At a full 2.0V, the memory breaks the 1800 MHz mark and reaches 1834 MHz. This is certainly a very good overclock of almost 15%.

Tightening the timings is the next step. These were set to CL6, then CL5. The X-Series can run at 1600 MHz CL6-6-6-18 and even beyond that using 2.0V. A-Data has clearly chosen to go with slighly more relaxed timings and lower voltage. This is certainly a smart move to expand the possible user base of such a product.

Lowering the timings to CL5 the memory almost manages 1400 MHz at 2.0V. The X-Series beats DDR2 memory in every aspect. You will never manage these speeds at these CL ratings with the second generation DDR memory.

A-DATA DDR3 X-Series 1600 MHz CL7-7-7-20 2GB Kit
CPU Clock Memory Ratio	Memory Speed	Memory Timings	Everest Read	Everest Write	Everest Latency	Quake 3 Timedemo	3DMark 2001SE	SuperPi Mod 1M
6 x 434 3:5	717 MHz	7-7-7-20 1.5V	8875 MB/s	6833 MB/s	60.0 ns	717 fps	27626	21.01 s
6 x 466 3:5	800 MHz	7-7-7-20 1.8V	9907 MB/s	7650 MB/s	54.0 ns	803 fps	29152	18.81 s
6 x 483 1:2	917 MHz	7-7-7-20 2.0V	10276 MB/s	7320 MB/s	51.1 ns	792 fps	28834	19.45 s
6 x 400 2:3	652 MHz	6-6-6-18 1.5V	9152 MB/s	6919 MB/s	57.7 ns	739 fps	27837	20.75 s
6 x 450 2:3	666 MHz	6-6-6-18 1.7V	9286 MB/s	7097 MB/s	57.3 ns	748 fps	27992	20.19 s
6 x 466 5:8	750 MHz	6-6-6-18 1.9V	9883 MB/s	7452 MB/s	53.4 ns	791 fps	28952	19.17s
6 x 474 3:5	800 MHz	6-6-6-18 2.0V	10247 MB/s	7626 MB/s	51.7 ns	804 fps	29320	18.71 s
6 x 416 1:2	814 MHz	6-6-6-18 2.0V	9154 MB/s	6492 MB/s	56.9 ns	716 fps	27258	21.90 s
6 x 469 3:4	549 MHz	5-5-5-15 1.5V	8439 MB/s	6546 MB/s	60.1 ns	704 fps	27312	21.90 s
6 x 480 3:4	600 MHz	5-5-5-15 1.8V	8883 MB/s	7168 MB/s	59.1 ns	754 fps	28340	20.12 s
6 x 426 2:3	666 MHz	5-5-5-15 2.0V	9329 MB/s	7070 MB/s	56.0 ns	757 fps	28145	20.29 s
6 x 436 2:3	684 MHz	5-5-5-15 2.0V	9624 MB/s	7261 MB/s	54.5 ns	767 fps	28494	19.76 s

Our sample scales very well with voltage, but seems to reach the limit at 918 MHz, as even CL8 or CL9 did not change that. The Mainboard can do around 970 MHz though, its maximum is 485 MHz FSB. The CL8 has been added for reference only, as you can reach the maximum speed with CL7 as well. Consider that these numbers reflect only the results of our individual kit, so your milage may vary!