ram timing (again)

[littlekates]

can someone tell me whats wrong with this timing. i seem to get very bad superPI result, 1M = 36s. pc currently overclocked to 2530mhz (253x10).

CAS letancy(tcl) = 2.5
Row Cycle Time(trc) = 13
Row Refresh Cycle(trfc) = 14
RAS to CAS delay(trcd) = 3
RAS to RAS delay(trrd) = 2
Min RAS active time(tras) = 5
Row Precharge Time(trp) = 3
Write Recovery Time(twr) = 3
Write to Read Delay Time(twtr) = 2
Read To Write Delay Time(trtw) = 4
refresh Rate (tref) = 166mhz

memory is 1GB(2x512) PC3200 Geil value ram.

 

[man_from_uncle]

Is that with a 1:1 ratio or a 6:5 divider? If it's 1:1 you look to be a bit off the pace for a venice core (should be 32-34 by my reckoning). If you are using the divider then that's only just on the high side of what i'd expect. Not sure how the older cores do at 2.5 ghz, but i imagine they are a tad slower.

Do you have any other ram options to get to your max overclock with that Geil? (Can you raise the frequency by going to CAS3 for example)

 

[littlekates]

its 6:5 me think as i use 166 divider. this is SPD timing table

freq 166Mhz:
CAS latency = 2.0
RAS to CAS = 3
RAS precharge = 3
Tras = 5

Freq 200Mhz:
CAS latency = 2.5
RAS to CAS = 3
RAS precharge = 3
TRas = 6

 

[mc_bob]

That's the same result I get, Spec in sig. Our machines our similiar after Overclocks.
I do think it should be slightly quicker however.