Q6700 OC tips/info

[Leon_Solitario]

I'm relatively new to overclocking and all I can seem to find is mixed info on the matter of Q7600 safe temps. I've done a bit of fooling around with OC before but nothing too indepth yet, mostly just CPU OCs. I'm using an old nvidea 680i SLI board that came with some upgraded cooling along with an older closed-loop watercooler (lawl, I know, but it still seems to do better than normal stock would have), the thermal paste is old (somewhere ~3 years) so I'll be replacing that soon and hope to get ~5c cooler from that. the board stays under 27~28c, the CPU has hit up to 66c when under max load from prime95/OCCT's linpack, but normally stays in the mid 50c range under normal load and mid~high 40s with minimal load, idling in the mid 30 range (stayed in the mid 50 range with the normal OCCT test too).

from what I get after doing a google search on the matter 72c seems to be the limit for these chips, but I'll want to stay in the low 60s (hoping that will be my max after reseating the cooler with some new goop) for better lifespan.

playing with the voltages (did this to reach stability with a 3.15GHz OC) I get some odd temp differences. raising the vcore the temp generally rose, which should be normal. so I decided to increase FSB voltage and it allowed me to reach stability with a lower vcore, the temp dropped a good several C. is this normal? no other temps seem to have raised in it's place, so I'm a bit weirded out atm, lol.

quick edit: using HWMonitor to read core temps.

 

[Leon_Solitario]

what I'm basically asking is, is this something I should have to do to achieve stability (and lower core temps)? raise the fsb voltage. or should I revert it to stock?

I can only raise it by 0.1v at a time and stock @ 1.2v is the lowest possible. I had to raise vcore above 1.35 (stock 1.2something) to achieve stability @ 1.2v fsb, but @ 1.3v fsb I can achieve stability under 1.35 vcore and get lower core temps (still testing how low I can take it).

edit: I also experience a ~0.03 voltage drop in vcore under extreme load. I know that it's normal to experience a drop, but what's normally to be expected?

 

[Leon_Solitario]

update: I'm getting up to 3.168GHz before the system no longer passes POST @ 3.17GHz, tried increasing voltage and still wouldn't give, the board is known for having trouble with higher OCs though (haven't messed with memory yet because it's going to be replaced soon and I'll get to start all over \o/ ). right now I'm just fine tuning the voltage to see how low it can hit before I start getting instability again.

quick stress tests showed 67c core max temp, the rest being mostly the same with under a couple degrees of difference.

I find it kind of odd again because quick testing @ 3.16GHz the temps were nearly several degrees cooler than they were at 3.15GHz. is this another odd occurrence that tends to happen at times? I had disabled speed step, thermal control and enhanced halt state earlier for tweaking/testing so the processor shouldn't be getting throttled down.

 

[Leon_Solitario]

I've gotten it stable at 3.167GHz@1.27v now. had prime95 running for over 7hrs with no errors/warnings and running 62~63C at max load with other temps scaling more or less the same (still need to reseat the cooler too).

fsb set to 1266 for 3.165 cpu frequency. vcore 1.3 for 1.25v, ftt 1.3v.

should I stick to this setting or try to drop ftt back to 1.2v and raise vcore? core temp will rise, but I'm not sure what other affects it will have (still can't seem to find much on raising ftt v. vcore voltages).

edit: vcore droop seems to be 0.03~0.04v max.