CMX-3200C2 Latency in Everest

[c13]

Everest Home edition (formerly AIDA32) is probably not the best tool for this, but I get 105.4 ns for my dual-channel setup of:

 

- 2xCMX256A-3200C2 revs. 6.1 (Bank 1) and 1.2 (Bank 2)

- Abit AS8 v1.0, BIOS rev. 16 with a P4 540 @ 2856 MHz.

 

Timings are currently 2-4-4-7. Just experimenting.

 

Anyone know if this is acceptable latency for these modules or not?

[Wired]

They're not guaranteed at their normal timings when different revisions are matched up, so those settings are ok.

[c13]

Would anyone say what latencies they are getting or would expect to get from either a matched pair of rev. 1.3 or 6.1's?

 

I had sent in a rev. 1.3 to RMA and got the 6.1 back but I believe I also read on this forum (IC Inquiry thread?) that a naming problem exists such that any rev. greater than 1.3 is actually 1.3, but I may be wrong about this.

[Wired]

You're thinking about an older thread here. The 2.x were just 1.x's mislabeled. rev 3 and above are actual revisions.

[c13]

I've done some looking around on the web and I believe I ought to be seeing much lower numbers, mixed revisions aside.

 

Wired, or anyone who knows their latency, care to post a number for 2xCMX256A-3200C2?

 

Strangely enough, my RAM is designated as PC3300 in Everest but that's probably the fault of their program interpreting my 204 MHz speed.

[Wired]

CAS Latency: 2 on Intel, 2.5 on AMD

RAS to CAS Delay (tRCD): 3

Row Precharge (tRP): 3

Active to Precharge Delay (tRAS): 6

 

Those are the TwinX or single stick timings. Not guaranteed when mixing revisions, but it is possible.