1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

Other [GUIDE] Windows Hardware Error Architecture (WHEA) (NOOB Friendly)

Discussion in 'Hardware' started by aramil, 11 Oct 2012.

  1. aramil

    aramil One does not simply upgrade Forums

    Joined:
    10 Jul 2012
    Posts:
    961
    Likes Received:
    57
    Windows Hardware Error Architecture (WHEA) & Are you really stable? (NOOB Friendly)


    What it is?

    “The Windows Hardware Error Architecture (WHEA) raises an event tracing for Windows (ETW) event whenever a hardware error occurs.

    The WHEA hardware error events that are defined, and the data templates that are used to describe those hardware error events, are different between Windows Vista and Windows Server 2008, Windows Vista SP1, and later versions of Windows.

    Starting with Windows Vista, the operating system maintains a list of all of the hardware error sources that can be discovered on a particular hardware platform. WHEA uses a discovery mechanism when the operating system starts to determine which of these hardware error sources exist on a particular system. The means by which this information is exposed to the operating system is platform-specific. The operating system collects this information from a combination of ACPI tables, firmware interactions, and other platform-specific mechanisms.

    A hardware error source is typically represented by the following:
    • One or more hardware error status registers
    • One or more hardware error configuration or control registers
    • A signalling mechanism to alert the operating system that a hardware error condition exists

    WHAT…..? Can I have that in English?

    Hidden inside of modern CPU’s and Operating Systems are very useful self-checks and reporting tools.

    EVENT VIEWER is Windows reporting tool.

    To open Event Viewer go to: Control Panel >Administrative Tools>Event Viewer

    In here you can see many various logs that windows keeps about your system and services that run on it. For this Subject we are only interested in “Windows Logs>System” (left hand pane).
    [​IMG]

    Now when Overclocking, We all know that stress testing is key (Prime95/IBT/”insert your test here”), and that running your system hard for a long time is a good test of being stable.

    Or is it?
    Here is a picture of a 12hr stable prime95 systems event viewer:
    [​IMG]

    As you can see we have “Hardware Error correction on the CPU Processor core”, this is where the CPU has detected an error and rerun the process to correct it.

    What does this mean?

    That this item had to go through the Processor TWICE. So taking twice as long to do, and importantly NOT causing an error in my stress test. I.e. I can pass stress testing without errors but I do have Hardware errors.

    To correct this I added one” +” (0.005v) to the Vcore rebooted and re ran the test: (if you still get errors “+” Vcore again)

    [​IMG]

    No errors. No wasted CPU Cycles. Faster, (and less likely to randomly crash or BSOD when gaming etc.)

    This is not an exhaustive description of how WHEA & Event Viewer (as it does so much more), but it shows enough that people should be able to get more stable without the “I ran prime for 8Hrs but BF3/ “insert game here” crashes my computer”.

    I hope this helps.

    WHEA TECHNOBABBLE on MS
     
    GoodBytes and Jaybles like this.
  2. Somer_Himpson

    Somer_Himpson New Member

    Joined:
    22 Jan 2010
    Posts:
    1,091
    Likes Received:
    53
    What the **** is this shite?!?!
     
  3. jizwizard

    jizwizard Well-Known Member

    Joined:
    24 Dec 2010
    Posts:
    1,903
    Likes Received:
    135
    in the words of the virgin mary. come again?
     
  4. Cei

    Cei pew pew pew

    Joined:
    22 Mar 2008
    Posts:
    4,717
    Likes Received:
    122
    I think I get what you're going on about, but there's one question, does it have an impact on benchmarks? If you can stop the CPU wasting cycles as you describe does it, for example, do SuperPi a bit faster?
     
  5. IvanIvanovich

    IvanIvanovich будет глотать вашу душу.

    Joined:
    31 Aug 2008
    Posts:
    4,870
    Likes Received:
    252
    Basically, they are letting us know that just because you ran prime and it didn't crash, that you can't necessarily call it good to go. Just another place to check to make sure your system is running at peak, after you have ran your stress test.
     

Share This Page