Mixing memory didn't drop benchmark much

[videobruce 0]

On a 100MHz memory bus (AMD 751 chipset) mixing one PC133 and two PC66 (?) strips only dropped the benchmark 5% running SiSoft Sandra 2004.

 

I'm not sure about these other 2 strips (32MB each) of memory. The date on the chips is 1998. Can I assume it is PC66? I'm just surprised the hit was less than 5% vs. testing only the PC133 strip (128MB) by itself.

 

It went from 660MB/sec down to 630MB/sec when the additional strips were added.

 

Input? Am I correct about the '98 chips being PC66?

[Tomay 0]

Probably not, but you can check that with sandra.

Go to Mainboard Information and scrool down to somethink like this

 

Memory Module(s)

Memory Module 1 : 128MB 16x(8Mx8) SDRAM PC133U-333-542 (CL3 upto 133MHz)

Memory Module 2 : AMD 128MB 16x(8Mx8) SDRAM PC133U-333-542 (CL3 upto 133MHz)

Memory Module 3 : 256MB 8x(32Mx8) SDRAM PC133U-333-542 (CL3 upto 133MHz)

 

 

you can see all the modules in my system are PC133.

[videobruce 0]

I returned the box to the owner since it was up tp par. I was just wondering about the lack of a hit with the older memory.

I know since the bus isn't a 133 the difference wouldn't be that great, but was surprised to see the little difference.

Since there WAS a hit, is it ok to assume that the older memory isn't PC100? If it was PC100 there shouldn't be a 'hit' at all mixing it with the other stick (even though it was 'better' memory?