Sync memory or faster async memory settings?

[ravynmagi]

I've been trying to decide if I should run my memory slower and synchronously or faster and asynchronously? Seems like the topics I've googled on this are kinda old, so wanted to get some current thoughts on this.

Here is specifically how I could run it..


3.0GHz CPU, 8 multiplier, 378MHz FSB, 2.0 memory multiplier.

756MHz memory, 4-4-4-12 timings (synchronous)


3.0GHz CPU, 9 multiplier, 335MHz FSB, 2.66 memory multiplier

891MHz memory, 4-4-4-12 timings (asynchronous)


3.0GHz CPU, 8 multiplier, 378MHz FSB, 2.66 memory multiplier

1005MHz memory, 5-5-5-15 timings (asynchronous)


I've tested these three scenarios to be most stable for me. So I'm debating the synchronous memory with the decent timings, the faster async memory with the same timings, or the fastest async memory with looser timings. Unfortunately lower CPU multipliers with a higher FSB are not possible for me.

 

[Cooperstown39]

I,m pretty new to this as well and read many through many posts on this subject. It seemed the more I read the more confused it became with many conflicting opinions. In the absence of a clear or prevailing opinion, I decided to test this myself.

I purchased a benchmarking program (around $30) and tested the various scenarios (multiple overclocks, with multiple memory timings). In the end, the CPU, Memory, and Graphics benchmarks showed no difference between running at 1:1 or having the memory running at a higher frequency. Strangely enough, the one benchmark that did show a significant difference was the hard drive. In all scenarios, the hard drive performed better with a 1:1 setting. I am guessing this has something to do with virtual memory and page swapping??

Again, I am very new to this and I am sure there will be many opposing opinions but I would suggest you benchmark your own system and see which settings work best with your components.

Hope this helps...