The Notebook Review forums were hosted by TechTarget, who shut down them down on January 31, 2022. This static read-only archive was pulled by NBR forum users between January 20 and January 31, 2022, in an effort to make sure that the valuable technical information that had been posted on the forums is preserved. For current discussions, many NBR forum users moved over to NotebookTalk.net after the shutdown.
Problems? See this thread at archive.org.

    Significant difference in core temps? (i5-430M)

    Discussion in 'Hardware Components and Aftermarket Upgrades' started by abaddon4180, May 28, 2010.

  1. abaddon4180

    abaddon4180 Notebook Virtuoso

    Reputations:
    1,229
    Messages:
    3,412
    Likes Received:
    39
    Trophy Points:
    116
    I just got my HP dm4t and one of the first things I did was download HWMonitor and I was just wondering if it was strange for one core to be higher than the higher by 10C. Here is the pic

    [​IMG]
     
  2. Gremo

    Gremo Notebook Geek

    Reputations:
    25
    Messages:
    91
    Likes Received:
    0
    Trophy Points:
    15
    Never seen something like this (only 2-3C difference in my system). I suppose that integrated GPU does not have its sensor, right?
     
  3. BruBoo

    BruBoo Notebook Evangelist

    Reputations:
    99
    Messages:
    486
    Likes Received:
    0
    Trophy Points:
    30
    Not all motherboard sensors are understood correctly by the usual monitoring applications.

    Before 'looking' at the readings set the monitor app to load at startup (e.g. put the shortcut in programs startup), then shut down and leave several hours before doing a cold start in a cold environment.

    Now

    Which sensors showed a minimum reading not far above the ambient temperature - they are possibly good

    Which sensors are not increasing in a slow and steady fashion as the laptop warms up - they are probably incorrect

    Which sensors are giving physicaly improbable readings (e.g. in a dual core processor at either idle or running a stress test application are the temperatures starting at the same or similar low value and tracking together (accuracy may get better as the temp rises)) These things should be true because physically each core is thermally connected a common cooler assembly using materials that should not allow the cores to move more than a 1- 2 degrees apart. Because of monitoring programme algorithms for temperature a 2 -3 reporting offset is also possible.

    So if your cores start at room temperature and track each other up and down within 2-3 degrees under the same load that is normal.

    Where any of these things are not happening - yet the system works fine the problem is likely to be monitoring not the actual temperatures
     
  4. kosti

    kosti Notebook Virtuoso

    Reputations:
    596
    Messages:
    2,162
    Likes Received:
    466
    Trophy Points:
    101
    It's normal for these chips to have different core temps at idle. Put some load on it with a stress test like Prime95 and I'm sure the difference will be a lot less. Your idles temps seem really low btw. Maybe try Hwinfo32 and see if it gives you a different reading.
     
  5. abaddon4180

    abaddon4180 Notebook Virtuoso

    Reputations:
    1,229
    Messages:
    3,412
    Likes Received:
    39
    Trophy Points:
    116
    I think it is a monitoring problem, then. When I started it up this morning and let it sit for a few minutes (ambient temp of about 21C) one core fluctuated between 26-31C while the higher one is between 37-39 and changes a lot more often.
     
  6. BamAlmighty

    BamAlmighty Notebook Consultant

    Reputations:
    47
    Messages:
    132
    Likes Received:
    0
    Trophy Points:
    30
    when you use hwmon you have to set the config file with the correct thermal spec for the processor.

    i.e. CPU_0_TJMAX=90.0

    90.0 = value you need to change based on your processors thermal spec