Test Setup

Performance & Overclocking

The memory SPD contains three CL ratings. These do not seem to be specifically picked, but are rather usual settings. 5-5-5-15 followed by 6-6-6-18 and 8-8-8-24. The RAS# Precharge is always triple the CL rating. Other manufactures go into greater detail and try to set tighter timings for this value. The three settings do hold a good base for each CL rating though. If you want to know how far a memory will go at each setting simply configure the specific speed in the BIOS. It would have been great if Aeneon would have also added a CL7-7-7-21 SPD.

The memory is rated for default 1.5V, so it was set at 1333 MHz in the BIOS with a 333 MHz FSB. This FSB rating was always used as a base to keep the CPU performance similar in each test runs, so that we can focus on the impact of performance with higher memory speeds. The PCIe clock was locked at 100 and the multiplier of the CPU was lowered to 6 instead of 7, to make sure that the CPU would not become the limiting factor. After running our benchmarks at this speed, we pushed the memory starting at 1.5V to its limit, slowly raising the Voltage by 0.1V steps up to 2.0V after each successful run.

The memory was tested at CL5-5-5-15, 6-6-6-18, 7-7-7-21 and 8-8-8-24. While it is possible to go higher with CL9 or CL10, it does not really serve any purpose, as such loose settings translate into slower performance. All milestones were benchmarked, as well as the highest possible at standard 1.5V or 2.0V - the maximum voltage one should use for DDR3 with sufficient cooling. It should be noted that Aeneon nowhere explicitly warrants "overvolted" memory. So you will use such settings at your own risk.

Wit the Aeneon DDR3 memory it became apparent quite fast, that voltage has no real impact on performance. The difference of maximum speed and any given CL rating between 1.5V or 2.0V operation is always somewhere in the range of 20 - 25 MHz. We were only able to push the memory further when the timings were relaxed.






The voltage scaling graph illustrates the afore mentioned lack of performance gain with higher voltage. The DIMMs only react to a change in latency. The modules max out at just over 1500 MHz at CL8 and 2.0V. This is quite respectable, considering a solid 12% boost at 2.0V. Remember that your milage may vary and this just reflects the performance of the samples.