Opinions on safe daily voltage for SB

Quote:

---

Originally Posted by Metroid

If it was a gem then it would not have died. :D

---

Yeah, cause I was only aiming for 5Ghz with this new plantform, if it was a gem it would have done that with less 0.1v at least :D

Quote:

---

Originally Posted by st0ned

Just to warn all of you, my CPU has officially degraded. I haven't played with the other volts so it mush ( should ) be vcore related :cool: . I wouldn't say also that 1.45v is safe, since 1.5 isn't. Maybe those 1.38v wondering around the internet aint so that unreal :rofl:

---

Sorry to hear about the chip degrading... Since you said it still runs stock will it run 4.5 now? I'm thinking maybe clocks with 1.35v to 1.375v load for longer term use.

I dont know yet, but I'll give it a go, and live feedback after it ends this stock run :up:

I will probably be running in the 1.45-1.49 range under water for 24/7, but I won't cry if the chip dies... People give all kinds of overly conservative estimates. I ran 1.5 vtt and vcore through my i7 860 and that chip lasted a whole year before it died. If you really don't want your chip failing then I would stay under 1.40.

I ran my i920 in that conditions too. But a two day degradation is too much for my to handle.

Quote:

---

Originally Posted by st0ned

I ran my i920 in that conditions too. But a two day degradation is too much for my to handle.

---

Yeah but you ran Linx for a long time. This would make it degrade faster I would imagine. Nonetheless, 1.5+ does not seem safe. I'll try 1.45.

The safe voltage is ever decreasing as the size of the transistors gets smaller. With the 2600K's being manufacturered on 32nm verses the 45nm of the orignial i7's its safe to say the true safe 24/7 voltage is less than 1.45V as recommended at the time. On the older i7 intel's recommended safe voltage was 1.35V. Those 2 year old chips have for the most part survived (at least in my case) at 1.4V constantly.

1.5V seems high for a 24/7 build, I'd say 1.4V would be more appropriate, that still represents a 40% increase in voltage. For suicide benchmarking 1.5V might be a little low!

I would have to say 1.375v for air cooling and 1.425v for water cooling. If you have top air cooling and don't get more than 70 degrees C with 1.4v should be okay aswell.