Conroe Temperature Sensors / Readings

I'm having trouble finding an accurate reading for my CPU given that there appears to be 3 distinct sensors in a conroe that people are reporting their temps from. I've seen many posters quoting their temperatures in forum posts here and elsewhere, sometimes giving the name of the application they use, but no explanations or discussions anywhere about the differences each sensor reports or which should be used as an accurate measure.

Bearing in mind I'm using an Asus board, I've used PC probe to get a reading from one sensor, that seems to sync with the one reported in the BIOS. The BIOS is usually a couple of degrees higher than idle in windows, which may be to do with speedstep. Other people also refer to CoreTemp and TAT, both these applications seem to give the same readings from 2 sensors, one embedded in each core. When temps are quoted from these apps, I'm assuming the poster is averaging them given that usually they're within a degree or two of each other. Everest lists all three, and the values seem to relate to the other apps I've used. Asus probe / BIOS always gives me a reading 10-15 degrees higher than CoreTemp / TAT, and yet I've seen other posters say it's the other way around on their setup and some who say they give the same readings. There seems to be no way to get accurate readings.

So which of the three sensors should be used to gauge the CPU temperature, and which do overclockers use as their guide when making sure things aren't getting too hot?

My temps - Idle, Loaded


Cheers
10 answers Last reply
More about conroe temperature sensors readings
  1. This is a repost of Computronix's info from aother thread,it should answer post of your questions. IRT temps, use TAT. It was designed for intel processors and takes the core temp reading from in die (old term) dont use PCprobe for OC

    You are correct about the thermal monitors. each die has one as well as the 'socket" more old terms. They are known as TM1 and TM2 TM= Thermal Monitor. In die is the prefered monitor for OC.

    For more info on overclocking the C2D

    As for diferent usrs temp variances..they dont mean anything. Case, cooling fan manufacture, HSF mounting all impact and have caused readings to be all over the place. If you are going to over clock, get a good aftermarket cooler. If not the stock HSF is fine. In either case, make sure you mount the HSF properly

    Wusy's THG C2D overclocking guide

    Quote:

    Here's Intel's Thermal Specification for the E6600 which is 60.1c as per the following link:

    http://processorfinder.intel.com/details.aspx?sSpec=SL9S8

    Quote:
    Thermal Specification: The thermal specification shown is the maximum case temperature at the maximum Thermal Design Power (TDP) value for that processor. It is measured at the geometric center on the topside of the processor integrated heat spreader. For processors without integrated heat spreaders such as mobile processors, the thermal specification is referred to as the junction temperature (Tj). The maximum junction temperature is defined by an activation of the processor Intel® Thermal Monitor. The Intel Thermal Monitor’s automatic mode is used to indicate that the maximum TJ has been reached.


    The first part of the 60c spec implies a single measuring point, which would be in contact with the CPU cooler, via the heat sink compound. Since there is no sensor at this measuring point, then BIOS temps are indicated from CPU look-up tables, referencing the motherboard's CPU socket temp sensor, to calculate this single measuring point. Additionally, one core always runs hotter than the other, so this calculated measurement is at best, an average value.

    Wusy's Overclocking Guide suggests using Core Temp, which displays slightly higher temps than Intel's Thermal Analysis Tool, (TAT). The second part of the spec refers to mobile CPU's measured at junction temperature (Tj) with Intel's Thermal Monitor (TAT), which reaches thermal red-line at 80c. Core Temp also measures internal CPU core (Tj) sensors, which correctly displays the (Tj max) spec as 85c.

    Obviously, there are some calibration and specification ambiguities here. Since Core Temp displays higher load temps than TAT, which is a Notebook tool, then TAT's values are inacurate for PC's. It's vague as to how the ~ 15c discrepancies between BIOS, motherboard utilities and SpeedFan, versus TAT and Core Temp should be interpreted. The difference seems to approximate BIOS + ~ 15c = Core Temp.

    Thermal Design Power (TDP) is exceeded by a considerable margin when vCore and CPU clock are increased over spec's at 100% load, so 50c BIOS and 65c Core Temp should be safe and reasonable thermal values for heavily OC'd worst-case gaming loads.

    CPU Temps *C.

    BIOS/CoreTemp
    -60-/----75-----
    -55-/----70-----
    -50-/----65-----
    -45-/----60-----
  2. Quote:
    Obviously, there are some calibration and specification ambiguities here. Since Core Temp displays higher load temps than TAT, which is a Notebook tool, then TAT's values are inacurate for PC's. It's vague as to how the ~ 15c discrepancies between BIOS, motherboard utilities and SpeedFan, versus TAT and Core Temp should be interpreted. The difference seems to approximate BIOS + ~ 15c = Core Temp.


    Is there an explanation as to why the P5B Deluxe reports temperatures the opposite way round to that described above? Are my Core Temp values as accurate as those on other boards?
  3. Quote:
    Obviously, there are some calibration and specification ambiguities here. Since Core Temp displays higher load temps than TAT, which is a Notebook tool, then TAT's values are inacurate for PC's. It's vague as to how the ~ 15c discrepancies between BIOS, motherboard utilities and SpeedFan, versus TAT and Core Temp should be interpreted. The difference seems to approximate BIOS + ~ 15c = Core Temp.


    Is there an explanation as to why the P5B Deluxe reports temperatures the opposite way round to that described above? Are my Core Temp values as accurate as those on other boards?


    Yes. We (a bunch of regulars) went through this months ago so I dont want to dredge it up again. In short, TAT is an Intel tool. Coretemp is someones personal project. Trust TAT
  4. Thanks JumpingJack, interesting article. I guess I'll rely more on the Core Temp / TAT values from now on.

    Hopefully I'll find an explanation at some point about why the P5B BIOS and PC Probe readings are 15C higher, rather than lower like on other boards. Just to put my mind at ease.
  5. Quote:
    Thanks JumpingJack, interesting article. I guess I'll rely more on the Core Temp / TAT values from now on.

    Hopefully I'll find an explanation at some point about why the P5B BIOS and PC Probe readings are 15C higher, rather than lower like on other boards. Just to put my mind at ease.


    PCProbe reads from TM2 (Thermal Monitor) outside the core
    CoreTemp and TAT read from TM1, "inside" the core
  6. Quote:
    Thanks JumpingJack, interesting article. I guess I'll rely more on the Core Temp / TAT values from now on.

    Hopefully I'll find an explanation at some point about why the P5B BIOS and PC Probe readings are 15C higher, rather than lower like on other boards. Just to put my mind at ease.


    PCProbe reads from TM2 (Thermal Monitor) outside the core
    CoreTemp and TAT read from TM1, "inside" the core

    Wouldn't you expect a reading outside the core to be below a reading inside the core?
  7. Quote:
    Wouldn't you expect a reading outside the core to be below a reading inside the core?


    Under most circumstances yes, but it really depends on the air flow, and temperature differentials for the setup. For example if you have a horizontal (wrt MB) mounted fan, most always the extrenal should be lower. If you have a vertically mounted fan on the HSF, then air flow around and beneath the socket may not disspate the heat away from the socket sensor compared to the TM1 sensor.

    Jack

    Ah, things are beginning to make sense now. I can't quite picture your horizontal/vertical mounting, but I know what you mean and would assume you would call the Zalman 9700 I use as vertically mounted.


    Cheers Jack and turpit
  8. At last - someone posting what I'd like to know aswell - probably stating it better than me.. :lol:

    Also got a P5B, and I saw some strange things indeed on this mobo.

    The way I read the whole temp thing is:
    a.) Core Temp most accurate - DTS "hardwired" to read off
    b.) PC Probe - DODGY
    c.) Tc = the 60C spec, dead center between the IHS and Cooler - but as stated already in this post, there's no sensor there, so LUT are used. Most apps use this as the temp reading one sees in Speedfan / Pc Prove etc.

    DTS will show higher than Tc, and this is because of surface area and heat transfer from each core to IHS -> thanx to turpit ( :wink: ) I now finally know what type of relation I'm looking at between DTS and LUT values. (correct me if I'm wrong)

    Now the real shocker:

    My initial BIOS (06xx) had a 4-5C offset from Core Temp. Going to 0804, and the added PECI option, temp differed (-)15-17 from Core Temp. Mailed Asus, and they said this is normal :roll:

    Then, I installed a 120mm fan to address the high MB temp (42-45C) - blowing from out my CD/DVD tray in my Centurion case. This dropped my mobo temp to 29C. What I didn't notice, was that the CPU fan speed would never go above 2300RPM (qfan disabled) from this point.

    SO, when I did notice it, I tried everything but the fan speed would not budge. After e-mail to Asus, 0908 BIOS addition, fan still behaved the same. So, I disconnect my Roll Royce Merlin fan (it's loud cos it's summer here), and then only I saw that the mobo temp is used as reference for CPU fan speed (oddly in my view).

    No matter what temp the CPU is, the fan will be fixed at a 2300RPM (+/- 200 RPM / 2C) - I guess the mobo temp sensor is located just before the DIMM slots. Already sent them a mail, hoping that this relation can be changed future wise, hopefully by the user himself.
  9. Quote:
    Hopefully I'll find an explanation at some point about why the P5B BIOS and PC Probe readings are 15C higher, rather than lower like on other boards.


    Have you reviewed the Asus P5B forums for information regarding your inverted temperature readings, and have you tried flashing your BIOS? As for CPU temperature alphabet soup, a little more explanation might help to level the thermal playing field.

    Intel's spec for Thermal Case Maximum (Tc max) on the Tcase sensor is 60c, and can be considered as a soft redline. This is the temp displayed in BIOS, motherboard utilities and SpeedFan.

    Intel's spec for Thermal Junction Maximum (Tj max) on the Tjunction sensors is 85c, and is a hard redline limit. 75c is hot, 80c is overtemp where throttling begins, and 85c is shutdown temp. These are the dual core temps displayed in Intel's Thermal Analysis Tool (TAT) and Core Temp. Emphasis is on the core with the highest temp.

    This colored temp guage helps to illustrate the 15c difference typically observed of these measurements:

    CPU Temps *C.

    BIOS/CoreTemp
    -60--/--75-75--
    -55--/--70-70--
    -50--/--65-65-- Load
    -45--/--60-60--
    -40--/--55-55--
    -35--/--50-50--
    -30--/--45-45-- Idle

    I hope this helps to make sense of CPU temps, and good luck! :D
  10. Blah, I don't trust those stupid core sensors; CoreTemp shows a difference of 10°C between my cores, both on idle and load and that simply can't be right. :?
Ask a new question

Read More

CPUs