updawg said:
Well my temps arn't my concern I can run the cpu stress test and it won't go over 40 C its just when I run the blend test it crashes. I've honestly gone up to 1.5825 and it still fails, would I be safe going higher? I mean my temps are fine even under load.
Well, assuming you have ruled out the memory erroring out, and you know its the CPU, you may be at the end of the line using "reasonable" voltage.
I personally think that 1.5825V is pretty high if you use the PC a good bit, esp. if its 24/7. If only used a few hours a day, (and your temps are very good, I see) I don't think I'd worry too much about it, however, esp. if your board undervolts a little like mine does.
Like me, I think your CPU (and alot of other peoples' C2Ds) are doing well to a point, then the required voltage goes way up for modest FSB gains.
I only brought up lapping the chip because in my experience, it can improve OCs more than whatever temp drop you get out of it would indicate. My thought is that a really flat / good contact between chip and cooling surface prevents / reduces "hot spots" within the chip, not just the temp diode spot. So you may "only" get a 7c drop after lapping, but you may get better stability.
I've tested this be "simulating" lapping by putting a small AC unit on my radiator, chilling the water to achieve a 10c drop in load temps, then lapping the chip, removing the AC unit, and re-loading the chip, where the temp. is still +/- 1c from where it was, but after being lapped I can get away with a few more FSB for a given Vcore.