Some Quick Memory Benches

[cnumartyr]

Ran some quick tests... DDR2-800 vs 1200 at stock and all. Q6600, Asus P5K-E, and Transcend DDR2-800 1.8v Kit. All tests were run 4 times and an average taken. Transaction Booster was on Enabled Level 3 for Setups 1 and 2... It had to be lowered to Level 2 for Setup 3.

Setup 1:
2.4 GHz, 1066 FSB
DDR2-800 4-4-4-12

Setup 2:
3.6 GHz, 1600 FSB
DDR2-800 4-4-4-12

Setup 3:
3.6 GHz, 1600 FSB
DDR2-1200 5-5-5-12


Results:

SuperPi 1M Times:

1: 21.48 Seconds
2: 14.43 Seconds
3: 14.11 Seconds

Memory Bandwidth Integer:

1: 5484 MB/s
2: 7156 MB/s
3: 8114 MB/s

Memory Bandwidth Floating Point:

1: 5490 MB/s
2: 7163 MB/s
3: 8123 MB/s

Memory Latency:

1: 88ns
2: 75ns
3: 60ns


So in synthetics, it's pretty obvious that faster memory really helps, but in the "practical" Super Pi Application it really didn't help that much. I'm sure if I had the time tonight to run 3D Mark 06 on all 3 setups it would prove to be even less of an improvement in gaming.

I posted this so perhaps people could understand a bit more that it really doesn't matter on an Intel setup as far as most things go. DDR2-800 is going to perform just as well as DDR2-1066. It's not that big of a deal in real world applications.

 

[Evilonigiri]

I would like to add to that.
http://www.anandtech.com/cpuchipsets/intel/showdoc.aspx?i=3184&p=8

The Origins of Static Read Control Delay (tRD)

With over a year of experience overclocking the Core 2 family of processors, we have learned a thing or two. One of the most important items we've learned is that higher FSB settings do not necessarily mean better performance. Understandably, this may come as a shock to some. For whatever reason, even a lot of well-regarded, seasoned overclockers seem to place great value in achieving the highest possible FSB. Based on what we know, we always establish our base target MCH overclock at the same spot - 400MHz FSB with a tRD of 6. The only other potential base MCH target value even worth considering is 450MHz with a tRD of 7, which should only be used when extra memory speed is needed or when a low maximum CPU multiplier becomes a limiting factor. Without getting into too much detail, let's examine what we mean by this.

It goes in further details, showing why higher fsb speeds was indeed faster before, but now with the new C2D, higher fsb speeds doesn't mean squat since the tRD timings can be changed.