You are killing that CPU

75-78c max with a Q66 using coretemp or realtemp to verify and Speedfan must be calibrated to read that correctly... realtemp indicates you are 8c from a dead CPU

Speedfan will not read the CPU core diode without the offset placed into the calculations from its settings

I suspect the CPU cooler is not able to cool the processor properly. A Q6600 on the right CPU cooler can run 1.4-1.45vcore and 3.6GHz if the BIOS has speedstep or CE1 shut down so the full 9x multiplier is used

With the right HSF and properly mounted it should -full load test- out to 75-78c max even after an hour... some can do better depending on the tower fans and ambient


With a PSU reading 11.25, its toast or too small for the system and that will also drive temps up

Dont 'set' anything in CoreTemp... let it read the CPU diode on its own

it should see that Q66 and read it correctly as 100c tJM and that means your MAX full load test temp after as long as an hour+ should remain 20-25c UNDER tjM or 75-80c MAX which the system will never see running FSX or other apps

Your PSU is not able to run this system, thats for sure and that will also drive temps up

I also suspect the HSF is not able to clock a quad as high as you are trying to go in combination

your version of Speedfan is OLD.. they may have corrected the offset in the newer version 4.35

install that and check it against CoreTemp... but you are reading CPU CORE 0 and NOT the default speedfan temp for the processor which is the CASE temp of the slug. You muct change that in the configuration to read CORE 0

Config - Temperatures - check INTEL CORE 0 to be displayed and verify that with Coretemp


On the older versions of speedfan the calibration is donw through the ADVANCED dropdown..

INTEL CORE at $0 on ISA

and set the temp offsets to make up for the case to diode temp variance using Coretemp to show you what the correct temp is


But as I said, the newer version of Speedfan may already have fixed the offset problem


here is what you should see and it should be close to what Coretemp reads for the Diode althouh it may not be exact... if you see a variance of more than 3c then you know speedfan is off and needs to be calibrated. The older versions of speedfan required a +13-15c adjustment







I just noticed looking at my speedfan readout on 12v its not right... I know my PSU is fine so its possible Speedfan is not reading yours right either. Grab Everest and verify the PSU on the 3.3, 5 and 12v outputs





as long as its above 11.75 you are OK but 11.70 is borderline.. 11.8 would be OK

congo wrote on Oct 4^th, 2008 at 1:26pm:



yea yea yea

its a typo..

LOL

fixed

luke wrote on Oct 4^th, 2008 at 6:24am:


Perhaps you didn't see it.

You should reference your mainboard manual about this Luke.

Chapter 4, page 22.

Do some research on the several options listed on that bios guide page.

On that note, it may be advisable to just read the whole bios section again as a refresher.

Until you know your bios inside out, you don't have the ability to manage the board properly.

Make bios your friend.

Your Computer property page is odd Luke. Why not install SP3 ?

Nick is right about the temps, don't run any stressing apps under the overclock until you get a cooler, there isn't any point as the overclock will likely change after you get a new PSU and cooler anyway, you only risk damage stressing it now.

NickN wrote on Oct 4^th, 2008 at 9:37pm:


Is that because of the mounting system Nick?

I saw someone who wanted that cooler until they researched the mounting procedure, and they decided to get a different cooler because they didn't like the way it installed.

NickN wrote on Oct 4^th, 2008 at 9:37pm:


From my understanding, load line calibration stops vcore voltage spiking when enabled, (reducing risk of cpu damage), is that right?

I'm sorry if the above seem obvious, but I just want to clarify Nick, there's so much bad or incomplete/non-specific info out there.

Cool!  

I was looking at those direct pipe contact coolers, and I wondered about differential thermal expansion between the metal types.
(probably not eough temperature extremes to worry about that though)

How do you find that type of cooler compared to the 120? They are pretty inexpensive here in Australia.