P5K Premium, Q6600 G0 Stepping H2O Build

Umm, Jason, could you try again to get below 13s on SuperPi? It's for Paul, not me.

QUOTE (cool_case @ September 06, 2007 09:16 pm)

Umm, Jason, could you try again to get below 13s on SuperPi? It's for Paul, not me.

Today was a rough run(s) to feel out the parameters, potential and limitations. I'll push a bit more tomorrow with more ammo at my disposal.

QUOTE (jmkays @ September 07, 2007 12:21 am)

QUOTE (cool_case @ September 06, 2007 09:16 pm)

Umm, Jason, could you try again to get below 13s on SuperPi? It's for Paul, not me.

Today was a rough run(s) to feel out the parameters, potential and limitations. I'll push a bit more tomorrow with more ammo at my disposal.

Really nice work, Jason! It's great how you are -*test*-('")ing the G0 and the QX6850 vs. each other.

And - whew - Paul will be relieved to hear that the 13s barrier will eventually fall.

QUOTE (Merc @ September 07, 2007 12:06 am)

I won't be seeing 4.0 either. BSOD at 1.675v. Playing with RAM right now. Have it at 480MHz 2.2v 44410 2T on P95 blend.

Man - these monster-high vcores. I didn't dare go above 1.625. Ah, the advantages of H20.

Did you try 8x with higher fsb? That's where I got my highest oc on the B3 quad.

It will fall. What sucks is SuperPi uses only one core. Why?

No haven't tried the lower multi. I'll give it a shot.

QUOTE (cool_case @ September 06, 2007 09:33 pm)

Did you try 8x with higher fsb? That's where I got my highest oc on the B3 quad.

Yes, I did. Suffice it to say, it was not a pretty sight.

An excellent day's work by Jason and Merc. The p5k and the G0 are both showing that they'll live up to the expectations for that pairing.

QUOTE (cool_case @ September 06, 2007 09:16 pm)

Umm, Jason, could you try again to get below 13s on SuperPi? It's for Paul, not me.

To WGF, with the love:

SuperPi: 12.906s. 3.960 GHz

SuperPi: 12.859s. 3.978 GHz

SuperPi: 12.812s. 3.995 GHz

SuperPi: 12.766s. 3.995 GHz

SuperPi: 12.735s. 4.005 GHz

Diode temps after running Orthos for 2 hours:

Very Nice!

That's great Jason. How did you break the 4.0 barrier at so low a Vcore?

To get a real stress -*test*-('") on your overclcok you'lhave to run 4 P95s simultaneously. Orthos doesn't hack it. Wanna put a link in my thread for a program that will do that automatically.

Temps have dropped significantly. Very nice.

Super, Jason!

QUOTE (Merc @ September 07, 2007 08:54 am)

That's great Jason. How did you break the 4.0 barrier at so low a Vcore?

I'm not sure, to be honest: just the right combination of settings (as published in my results). Anything higher than 1.625v, and the chip was very unstable.

It was relatively easy to overclock once I established the right combination of settings, but very tricky to arrive at that combination; hence, the two hours of trial and error.

QUOTE (jmkays @ September 07, 2007 02:18 pm)

QUOTE (Merc @ September 07, 2007 08:54 am)

That's great Jason. How did you break the 4.0 barrier at so low a Vcore?

I'm not sure, to be honest: just the right combination of settings (as published in my results). Anything higher than 1.625v, and the chip was very unstable.

It was relatively easy to overclock once I established the right combination of settings, but very tricky to arrive at that combination; hence, the two hours of trial and error.

Sure ... it was straight trial and error science.

You conveniently forgot to mention that the song "It's Magic" was playing in the background the whole time.

Resetting the strap seems to be the key Jason. Like I said in the PM I hadn't touched it, it was on auto. I reset to 333 and Vcore to 1.625v w/ a 434 FSB and she is running P95x4 at 3901 right now. I'll keep playing with it as I have a lot to learn about this new board, the CPU and the chipset. A LOT!!

UPDATE: It BSODed after 3 minutes of P95x4 at 1.59v Vcore (1.625 set)