• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Inaccurate temp reads on various programs

E

ElvisG

Limp Gawd
Joined
Sep 1, 2005
Messages
326
My current rig...
E6600
Big Typhoon CPU cooler
Arctic Silver 5
ASUS p5w dh deluxe

When I overclocked the CPU over 2 years ago I remember my temps being 35 - 39 degrees C. After a couple of days of burning in, the temps was always within the normal range. Since then I only monitored the temps in the BIOS every 6 months or so. The reason being is because I just assumed everything was OK as long as Windows was booting up.

Well, two weeks ago, the power went out and the motherboard reset to factory default. After a week, I decided to check the temps again. I used Realtemps and CPU-z and noticed that the temps was in the 60 - 65 degrees C range. A little concern, I decided to reboot into the BIOS and see what the BIOS said the temps were. So when I rebooted the BIOS said that the temps were in the 35 - 38 degrees C range. When Windows booted up I checked AiBooster to see what the temps were and it was what the BIOS said range.

Because of the electricity going out, is a temp sensor messed up or is this normal because the program is reading a different sensor?
 
Was it under load when it hit 60-65? It's not going to be under any in BIOS so it's impossible to tell load temps from there.
 
The 60+ degree C temps where just after a boot-up under no load.
 
Real Temp is the most accurate way to check your temperatures, so I suggest you rely on it as far as that goes. Try loading your CPU with Prime95 or LinX/IntelBurnTest/OCCT and see what kinds of temperatures you get from that. You may need to reseat your heatsink if they are too high.
 
You might need to remount your CPU, using a fresh application of thermal paste. It seems that thermal paste may lose efficiency after a long period of time + heat. It has been 2 years for you after all. Make sure to clean both the cpu and heatsink first before applying the new coating of thermal paste.
 
Just to give everyone a update, this problem has been solved. It turns out to be a power supply issue. My power supply finally died this morning. At first, I couldn't decided if the motherboard or the power supply was the issue. The only thing I had to go on was my ASUS P5W DH motherboard stand-by light was blinking instead of being on with a steady light.

So I went out and bought a new power supply and booted up the machine. Once Windows booted up, I checked the temps again using real temps, speed-fan and what do you know. My temps are in the 30 degrees C instead of 60 degrees C upon boot-up.
 
Can someone please educate me ... do different temperature readings from various programs essentially reflect differing assumptions of TjMax? If so, does that mean that the chip manufacturers do not publish TjMax values (otherwise every program would reference that value...). Do I have this straight? Thanks.
 
RobertP said:
Can someone please educate me ... do different temperature readings from various programs essentially reflect differing assumptions of TjMax? If so, does that mean that the chip manufacturers do not publish TjMax values (otherwise every program would reference that value...). Do I have this straight? Thanks.
Click to expand...
Programs that read temperatures from the DTS like Real Temp, Core Temp, Speedfan, etc., only differ in their readings when there is some contention as to what the real value of Tjmax is. Intel has published Tjmax values for some of their CPUs, but not for many of their older ones. Therefore, the target Tjmax values used by temperature programs for 65nm Core 2 CPUs are generally just best-guess numbers. However, the Tjmax values for all 45nm Core 2 CPUs and anything newer are known, so with those CPUs, the readings you get will be as accurate as is possible.
 
You must log in or register to reply here.
Back
Top