Not under load, just having the computer on 24/7.

I've overclocked my computer and the autovoltage is around 1.44 (if I don't use auto it will crash prime95 within 5 hours).

Is this too much voltage? Will it significantly shorten my cpu lifespan or risk harming my mobo/cpu?

I'm using an e6850. 3.0@3.6.

well if you use autovoltage while overclocking you deserve to get your chip burnt

just set it manually to the lowest stable voltage, autovoltage usually sets it unnecesseraly high...

What are your temps?

What motherboard are you using?

Did you try manually setting it to 1.44v?

If not I would do that and run Prime and if it's stable then gradually reduce the voltage one step at a time until it becomes unstable.

With increased voltage comes an increased chance of Electromigration http://en.wikipedia.org/wiki/Electromigration

You have a very good CPU and depending on which board your using and what BIOS it has you should be able to get it stable at a lower voltage than that.

p35 ds3r, award bios



temp is 64C under MAX load, prime95 smallfft. blend is about 58C.

i hate testing prime95 cuz it takes friggin hours to find the instability. the thing is, auto-voltage auto-regulates all voltage on my mobo, not just CPU. when I adjust fsb voltage it can affect stability, as well as all that other stupid voltage. fsb, mch, and ddr voltage confuse as to their necessity when oc'ing.

Manually, I've tried 1.4125 and prime95 crashes after about 5 hours. temps with 1.4125 are virtually identical to autovoltage temps at around 1.44. My board also has significant vdroop.

1.44V is fine for 24/7 use, as long as you don't intend to keep the CPU until it reaches antique status.

It's within norm:

VID Voltage Range: 0.85-1.5v

http://processorfinder.intel.com/d [...] Spec=SLA9U

So what is considered antique use? Would the cpu last about 4 years at 1.44?

It should last for years but 1.44 is too high for 65nm processor. On the positive side unless you turned off speed step it will throttle way down when idle so it's likely not an issue. The reference voltage above is not the acceptable voltage for CPU like say ram voltage range.

1.44v is too high? People keep their systems clocked at 1.45-1.55v using conroes. From what I've read I should be able to get 3.6 @ 1.4 or 1.41. Not sure why I can't.

I don't know about speed step, I looked for it in the BIOS but I couldn't find it, unless it's called something else in gigabyte mobos. CPU-Z reports vcore for the cpu LOWER under load and HIGHER when idle..

I think it's C1E if I remember right. It downsteps the multiplier to 6 on my motherboard when there's no load. My E4300 runs at 3 Mhz (9 x 333), and when no load, multiplier goes to 6 and I'm running at 1998 Mhz, slightly above stock 1.8. My MB is p965-DS3 rev 2.

It's only a 10% overvolt, well within 'acceptable' limits. You see some people running 1.5V with high end air or water cooling, and I don't see many dead C2Ds? Hmm...

As long as you have plenty of cooling, it's fine.

it's fine

It's quite high (my opinion) for a 600mhz overclock.
You should try to reduce the voltage to stock (or around 1.25V) and work your way up until it's stable (saves a lot of time).

Should I touch FSB/MCH voltage?

Try stock settings up to say 1.35V cpu.

If it's still unstable you could test with a very small fsb overvolt.

MCH shouldn't really need overvolting at 400MHz.

So set voltage to 1.35V and see the highest I can get running prime95 smallfft?

I couldn't imagine getting remotely close to 3.6ghz considering I would get errors a few hours in at 1.38.

Really the limit for air cooling is 1.5 Vcore. any higher than this and you processer won't last. If your using a high quality water cooling system then up to 1.6 Vcore is possible.

2003 i set my 3.0c to 1.45v

they said the same things!

2007 24/7 it still works


c2d should be 1.41-1.45v

1.44 is pretty high u know.
I managed to get 2.7Ghz out my q6600 with 1.15v (bios)
It ranges 1.09-1.12 (load-idle) 100% stable

EDIT: Thats on a B3 stepping too.