Originally Posted by gsk3rd
Yea I am going to do this tonight and see what happens.
Lapped the TRUE last night and got it hooked up. idle temps were 27c load were 46c @ 1.535volts.
I am also testing to see if my sensors or chip are faulty cause more tests revealed the 52c barrier with my chip. Something is wrong and I am bound to find out what.
100 pages of knowledge!!!!!
My chip seemingly does not like to go above 52 degrees either, at least for the voltage I've been pumping through it so far. Most I've ever tried is 1.56v, and I stopped testing after that. Case in point, I hit 3.9ghz OCCT stable under at least three different configurations, and each time the processor never exceeded 52-53 degrees. However, when I was testing my highest oc after my fans broke, my chip would seem to crash under load when @ 53 degrees everytime. And I know my cpu/motherboard can usually handle that oc, as it passed OCCT twice.
From what I've seen I would say that these chips have certain temperature thresholds, given a particular frequency and voltage. To achieve a certain clock frequency you would need a certain amount of vcore, however this is also dependant on the cooling method used and the temperature of the processor itself. The problem as we all know lies in the fact that as you pump in more voltage to a cpu, heat increases. I believe that is one of the factors in why these chips might be difficult to overclock. That's also the most likely reason as to why people on phase seem to hit stable oc's of the same frequency as those on air, yet with a much, much lower vcore. Chew over at XS proposed the theory that water makes little to almost no difference over extreme air setups, and given that a water setup will not cool much further than a very good air setup this would make sense. The temps would be lower, but would not be enough to overcome the various 'temperature thresholds' so to speak.
I may be very wrong, but take my example. Using 1.5v, I can hit 3.9ghz OCCT stable with my temps never exceeding 51 degrees. I can also hit 3.9ghz using 1.53v, once again not exceeding 51 degrees. Same for 3.9ghz using 1.55, it just hits 52 degrees. The interesting thing about the last 3.9ghz OC I did however, was that at 1.55v it was only stable once. I noticied after OCCT finished that my sound drivers failed, which I never had never seen before prompting me to believe that it was not completely stable. True enough, when I tested it again I got a BSOD @ 53 degrees. Variations in ambient temperatures could account for that 1 degree difference.
However many people here have run their chips @ well over 52 degrees, and currently because my system is buggered it'll hit 58 degrees on 34 degree days. I'm obviously not stress testing it under 34 degree ambient temps, but it does handle 'normal high load' such as gaming. So in the end it may be particular to a few chips, or just the overclocks me and Gsk3rd have been playing with, but my cpu seems to dislike going over 52-53 degrees as well. As for temp monitoring, I'm currently using coretemp as well as the primary program. Wizzard, OCCT (and it's graphs/logs), PC probe II however also all report the same temperature reading for me.