Memory Performance

[authenticate]

I have recently assembled a dual-cpu system and am very happy with it. However, I thought that I would benchmark it using SiSoft Sandra just to see how it rates on a world-scale. The memory benchmark suggests that my single piece of 256Mb SDRAM performs rather poorly compared to RAM in certain benchmark machines. I don't know what to make of this. It's made by IBM so I thought that it should be good. The specifications section confirms that it is PC133 standard.
Should I be concerned anyone?

 

[Tweakin]

On the average, dual systems will benchmark lower then a single system in memory performance. I'm not sure, but I believe this is due to the latency from the dual processor pipeline. My Tyan was not a screamer in the benchmark area, but real world it smacked the you know what out of my single systems. Just a thought.

 

[Oyeve]

I have a similar problem but without the dual cpus. I have a rather highend system. My memory is 1024megs of pc133mhz cas3 but according to sisandra my memory runs slower than most of the other benchmark comparison machines sisandra has. I think the program just sucks.

 

[bernse]

Same here. I run a 1.2 Ghz T-Bird and 384MB of Inineon PC133 RAM. I get slightly better than the benchmarks it gives for a Duron 600. Absolutely every other benchmark Sandra gives my system just smokes though. I am glad to see that I may not be alone and the memory bench program might just be crap.

 

[Taz4158]

Try running the benchmark tool Performance Test by Passmark, available:

Here