more testing done.
Bear in mind i haven't fully primed anything yet, just don't have the time right now to chuck it through 8 hours of stress. Each test 'pass' that i refer to, i am reffering to orthos Small fft's for THIRTY MINS ONLY. I also haven;t fiddled with the voltage, just whacked it up to 1.45v in the bios and started playing (1.44v real) Temps are measured using real temp, and is the maximum temp reading of the hottest core (core0, every time.. core1 is 10°c colder...)
Ok, so it passed 379x9.5 no problems, and today i decided that i'll have a play with the fsb, so, multi down to 6 (also updated the bios of this p5k-vm... i now have temps reading in the 30's instead of -6°c) and i can select x6, x7, x7.5, x8, x8.5, x9 and x9.5 for the multi =)
So, as it was having trouble at 400x9.5, i tried 400x6.. PASS (52°c)
then i thought i'd just chuck it straight up there... 450x6 = PASS (54°c)
475x6 = PASS (quite chuffed with this considdering it's a mATX board...) (56°c)
500x6 = FAIL, and an epic one at that... wouldn't een post. Haven't tried anything in between yet, or bothered to up the voltage, this is about as high as i'm willing to go (for the time being
)
So i went on to try the cpu speed, now it didn't like 3800mhz as 9.5x400, so i thought i'd aim for that...
Straight in at 450x8 (3600mhz, i know it was ok at this speed before) = PASS - delighted! (59°c)
450x8.5 = PASS - Horah, 3.85ghz on a 450 fsb on an £80 cpu, will hapily stay at this speed if i can fiddle the voltage down a bit (62°c)
450x9 = BSOD upon entering windows, notch the voltage up to 1.5 (1.48 real - i know, i'm either brave or stupid...) loads windows, and is currently sat at 4050 mhz now, but won't prime for longer than a minute (core1 fails every time, core0 will keep going for the 30 mins though) (68°c)
So, i think i'll now pop it back down to 450x8.5 and nudge the fsb a little (470 infact
)
Water should be installed within the next couple of weeks, so i'll let you lot know how it goes.