• 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.

Core temp, realtemp etc.

jmartis

n00b
Joined
Aug 9, 2008
Messages
60
I have an E7200 CPU.
RealTemp shows idle temp at ~35°C while Core temp and CPUID Hardware Monitor show exactly 10C more than Realtemp.

Anyone guess which temp is "real"?
 
We will have a definite answer either today or tomorrow after Intel provides the true value of TJmax, odds are with Real Temp.
 
I have an E7200 CPU.
RealTemp shows idle temp at ~35°C while Core temp and CPUID Hardware Monitor show exactly 10C more than Realtemp.

Anyone guess which temp is "real"?

"A man with one watch always knows the time. A man with two watches is never sure."
 
So the TJmax is 95C? I read on Intel website its 74.1C. Just one question as Im new to this matter, the "Distance to TJmax" reading is always right?
 
So the TJmax is 95C? I read on Intel website its 74.1C. Just one question as Im new to this matter, the "Distance to TJmax" reading is always right?

74.1C would be your thermal spec. So you want to keep Tcase under that. I would say you shouldn't exceed 70C to be safe. That would cover inaccuracies in the Tcase sensor.
 
The actual Tjmax for most 45nm Intel CPUs (except for the QX9650 and QX9770) is 100C as was released by Intel yesterday. If you manually set it to that number in Real Temp or Core Temp, you'll get the most accurate reading possible from the DTS.
 
So you want to keep Tcase under that. I would say you shouldn't exceed 70C to be safe. That would cover inaccuracies in the Tcase sensor.
:eek::eek: Case at 74C? Talk about cooking hard drives! I think I've never seen any case to get over 40-50C ever...
 
The actual Tjmax for most 45nm Intel CPUs (except for the QX9650 and QX9770) is 100C as was released by Intel yesterday. If you manually set it to that number in Real Temp or Core Temp, you'll get the most accurate reading possible from the DTS.

Well, not exactly the true temperature. The problem is that with bigger distance from TjMax you are getting the values with bigger and bigger error (in other words, if you make a straight line for temperature, then the DTS reporting can be anywhere in triangle which have it's most correct spot at Tjmax, and with lower temperatures the error increases). But using 100C for 45nm processors are best result you can get for now.
 
Well, not exactly the true temperature. The problem is that with bigger distance from TjMax you are getting the values with bigger and bigger error (in other words, if you make a straight line for temperature, then the DTS reporting can be anywhere in triangle which have it's most correct spot at Tjmax, and with lower temperatures the error increases). But using 100C for 45nm processors are best result you can get for now.

That's why I said the most accurate reading possible from the DTS instead of saying that you'll get the actual temperature ;). I'm aware that the DTS isn't entirely accurate, but it's the best way we have to measure temps save for drilling a channel into the heatspreader and sticking in a thermal diode like the [H] and Intel do it :D.
 
What is the purpose of such inaccurate thermal sensor? On my old P4 Prescott I got always accurate temp reading, why did they mess it up in Core 2 CPUs?
 
What is the purpose of such inaccurate thermal sensor? On my old P4 Prescott I got always accurate temp reading, why did they mess it up in Core 2 CPUs?

No, your P4 Prescott doesn't have accurate temp reading. DTS in all Intel processors are same - used for shutdown in case of overheating. Fact, that you can read it's value and present it as a temperature is just a side effect.
 
hmm, so thats why coretemp oddly shows no matter what frq. I set it at will all ways so the same temp.
 
What is the purpose of such inaccurate thermal sensor? On my old P4 Prescott I got always accurate temp reading, why did they mess it up in Core 2 CPUs?

The P4 used an old thermal diode, which worked in a different way. There was less temperature variations with the old diodes, however the new DTSes that Intel uses are far more accurate when it comes to their actual purpose, which is to send the PROCHOT signal if Tjmax is reached. The farther away from Tjmax the temperature is, the higher the error; however, at Tjmax the DTS is far more accurate than a thermal diode. They're not designed to provide accurate temperature readings.
 
Back
Top