Strange Overclock Results

[govtcheez75]

I ran my previous x2 3800+ at 2520mhz (250X10). I just recently upgraded it to a 4600+ which is stable at 2808mhz (234X12).

I thought the 4600+ at 2.8ghz would be faster (maybe by just a little bit, but definitely faster), but SuperPi times have actually gone UP!

I thought that 300mhz overall would easily be enough to cover for the 16mhz in FSB/HTT.

....any thoughts?

 

[Thor86]

That is weird, but what are your ram settings/timings?

 

[Duvie]

I agree it seems weird....It is not unocmmon for faster cpus to be slower due to worse ram timings, dividers, etc...however not to the tune to effect near 300mhz in cpu increase...It would have to be running cas 3, 2t and a divider that made the ram run below 400ddr versus 500ddr with the one that ran at 250HTT...

1) make sure it is same version...I have seen differences from the different mods and versions...
2) make sure whatever core is running it is not sharing with something else....you could set affinity on it ...

 

[cmrmrc]

my superpi 1M time is 30.7s?...same clock as you...how about yours?

 

[Polish3d]

13 seconds

 

[cmrmrc]

Originally posted by: Frackal
13 seconds

not helping here........

 

[myocardia]

Same motherboard? Same RAM, same divider? Same version of SuperPi? All of those things will make a difference.

 

[cmrmrc]

Originally posted by: myocardia
Same motherboard? Same RAM, same divider? Same version of SuperPi? All of those things will make a difference.

im waiting to see what is his pi score so that i can judge....

you're right, but even if we have totally different components, the score won't differ alot...i'd say max 3s....

 

[govtcheez75]

Yep. All else is the exactly the same. I didn't have to mess with anything other than clock speeds and voltage (this thing does 2.8ghz at much less voltage than my 3800+ did 2.5).

It passes all stability tests including occt/snm/stress prime 18+ hours.