Q6600 Core Temp Question ..

blackninja

blackninja

blackninja

Associate
Joined
13 Nov 2007
Posts
1,483
Hi,

Just got my Q6600 & TRUE today - fitted it and it went rapidly to 3.7 GHz (9x412) @ which I am happy with for now. Gives 14735 with the OCed GTX in 3dmark06 so I know that all is well. 3.8GHz was just over the noise threshold for the initial install :D

My core temps with CoreTemp under p95 are showing as C0 = 66, C1 = 58, C2 = 58 & C3 = 66 .. at idle the temps are (now in desktop) : C0 = 44, C1 = 37, C2 = 44 & C3 = 37 -I can accept this as a possible core/ihs interface issue OR a TRUE concave issue (and it did look a bit off off of flat but not greatly). I do intend to lap it though.

Strangely though, Speedfan reports the same temperature within 2 degrees for all cores consistantly. Any ideas as to why speed fan reports closer temps for each core whilst CoreTemp has 2 that seem to be 8 degrees below the other 2 ???

I am inclined to believe that the usual delta of 15C to speedfan is correct (which matches 2 of the cores almost exactly) because the delta between the cores unler load AND idle is the same with CoreTemp ... which I think would not be the case ...!!!

Cheers
Paul
 
About 1/2 hour P95 @ 3.8 but it took 4 notches of voltage on the CPU and one on the MCH to get it there though .. that's from 3.7GHz which requires 1.425 Vcore (=1.37 measured) and 1.29 MCH to be stable. To keep it below 70C (CoreTemp max core reading) in my closed case required a bit of welly on the deltas :p Which was an issue :D
 
if you download the new speedfan beta (you will need to register) should read the same as coretemp (have sorted the T-Junction read issue)..

Lapping sorted a lot of my uneven temp issues... also you have to keep in mind that unless that ultra is very tight it will pull away slightly from the CPU (assuming you have you motherboard vertical and the ultra "hanging" off it
 
You must log in or register to reply here.
Back
Top Bottom