q8400 running over 100c?????!!

Hi Guys,

My q8400 is currently overclocked to 3.2 ghz from stock of 2.66 and has been for the best part of almost a year and is runs super stable with no crashes or problems this entire time.

However - I just downloaded and ran realtemp to find out if I could push it any further once I get my new cooler and the readings coming back were rediculous!

On idle the temp is sitting around 40 - 50c and once ingame (battlefield 3) the temps went up to around 102 - 103!

Interesting to note that the computer is rock solid stable and has never shut down or displayed any symptoms of over heating. But just to be on the safe I went into bios and ran the default clock settings again with absolutely no change in idle temperature.

Is my sensor gone or is realtemp giving me a false reading? OR is my chip and mobo running on kryptonite..?

System Setup:

q8400 @ 3.2gz
P5q Pro
cool master 700w power supply <--- edited today!
Gigabyte 460GTX
4gz DDR2
1x Hard disk

 

Are you running a better than stock cooler?

I dare say the temperature reading is inaccurate! Would have thought at stock the idle temps would have been at least a bit lower though...

 

If that is right (i doubt) then your cooler must not be mounted right. I think it's more likely that it is a false reading. How hot does the cooler feel to touch while it is under load?

 

Just done a stress test and as you can see the load temps are insane. I also cracked open the case and noticed my mobo heat pipes were very hot (I couldnt touch the cooling fins for long)

This is running at stock speeds with a stock cooler and additional head removing fan.

The strange thing is that I have been running this system OC'd @ 3.2ghz for almost a year absolutely 100% fine and only just decided to check the temps now.

How likely is it the readings are incorrect or is possible the chip could opperate find without burning out at those temps?

(SEE LINK FOR JPEG UPLOAD OF SCREEN SHOT)

http://postimage.org/image/fsewqsd0f

 

To add to the above the PC has NEVER shut down on me in an attempt to save itself from overheating...???

 

False reading...

Try Real temp (another good program)

and see if it gives the same results..

 

If the heatsink is too hot, you need to either fix the heatsink fan and/or improve case air flow.
Be aware that letting a CPU run too hot for a long time will cause power conducting channels to develop in the silicon a lot faster, the CPU may now permanently run a bit hotter.

Mak sure the fan is set to run at max speed, not based on temperature.

 

I've checked the fan speed and it running at max which from what I gather is 2000rpm.

I've taken another screen shot with the various programs and it is possible that I need to allow for a -15 offset which coincides with the ASUS AI Suite temp.



I think I do recall checking the temps back when I first overclocked and they were completely normal according to the masses hence my reason for leaving it for so long which may have indeed developed power conducting channels.

What I'm stumped on is why the PC has never shut down or showed any signs of overheating over this entire time while I've been heavy gaming??

(SEE LINK FOR NEW SCREEN SHOT)

http://imageshack.us/photo/my-images/18/hightemp2.jpg/

 

I would buy a cheap custom cooler if i was you. If you can't touch your cpu cooler then it's far to hot. Mine barely gets warm unless i do a stress test but still it is 20-30c from being too hot to touch and hold. As for not getting thermal shutdowns/instability, it is puzzling if those temps are accurate.

 

I will be purchasing a Hyper 212 plus tomorrow with dual fans to really help bring those temps down even if my theory of -15 offset is correct.

And to clarify my current stock CPU cooler was perfectly fine to touch it was simply the cooling fins on my actual mobo itself connecting to the heatpipes that was pretty hot. I don't know if this is serious or not - i imagine probably not serious.

 

you must test to find the right offset on certain chips
http://www.techpowerup.com/realtemp/docs.php

 

btw, y ur cpu load at 100%? did u run any background program?

 

Yes I had prime 95 going in the back ground

 

Those are some insane temps. my advice: unmount cooler, clean paste, put some new in (arctis silver 5 or ceramique 2), reseat, use a lower OC the first 100hs, then reapply current OC. check temps.
Option 2, get a better cooler.

 

Even if the temperatures displayed by realtemp/coretemp are incorrect, your CPU is still overheating as it is hitting Tjmax.

Tjmax is the zero point of the core temperature sensors and all core temperatures are displayed as the distance from Tjmax. When you hit Tjmax the CPU will start throttling to prevent damage, which is why it seems to sit at 99C (Tjmax -1).

 

Its melting like an ice cream on a hot summer day

 

Maybe dust?

 

Try Aida64 (ex Everest)
That is one of the few apps that really shows my temps correctly.

 

Also, Coretemp and Realtemp both correctly select TjMax for known CPUs so will display correct temps too.
As close as the accuracy of the temp sensor circuits anyway, which isnt necessarily very accurate unless you are near TjMax.
Unfortunately, the C2D, C2Q series are very inaccurate.

Forgot to say, as the heatsink is really hot, it looks like you are at TjMax.
This isnt due to lack of quality heatsink paste otherwise the cooler wouldnt be hot.
The problem is that the cooling isnt good enough, the CPU is using too much power through too high voltage or the CPU is knackered and conducts too much current to ground through overheating damage.

 

Yeah, it cuold be, maybe use a compressed air can to dust off?