What's wrong? 5:4 ram ratio yields no better than 3:2...

[Naruto]

I have a abit is7-e with 2x256mb pc2700 ram.

@ 3.3ghz (275x12 & 3:2 ratio - ram @ 366mhz 2.5,3,3,7)

Sandra gets 5300MB/s

@ 3.0ghz (250x12 & 5:4 ratio - ram @ 400mhz 2.5,3,3,7)

Sandra gets 5300MB/s

Can anyone help me out? And are my ram bandwith results bad? Thanks.

 

[Soulkeeper]

haha what a coincidence that those 4 different mhz values yeilded you the same result in the end
the 3.3ghz setting should be the better deal if you weren't able to reach those speeds with 5:4

those bandwidth results look about right i suppose, but i wouldn't put much credence to the synthetic sifsoft mem benchies tho because all it really does is transfer data accross the fsb and basically time the transfer to calculate stuff then it prob doubles it or something. I dunno how effective it is in a real world situation, it's calculations are fairly close because the fsb increase would make the system more powerfull, just as the extra mhz will. But what it's not telling you is the system with 400mhz mem actually does have more mem bandwidth, it prob just isn't able use it as effective as the other setting due to the fsb and clock speed increase, which both help the memory subsystem in nearly every case.

 

[wicktron]

Perform an unbuffered test rather than buffered. It will give you a better indication of what's really going on.