IRQL_NOT_LESS_OR_EQUAL BSOD NF7-S v2.0

Discussion in 'Abit' started by tHatDudeUK, Jan 1, 2004.

  1. tHatDudeUK

    tHatDudeUK Guest

    I wondered if someone could help me troubleshoot this very annoying problem.

    I have had this for a week or two but it's starting to get worse. I am
    certain it must be a hardware problem of some sort. I get the BSOD with
    IRQL_NOT_LESS_OR_EQUAL displayed for less than a second then my machine
    reboots and everything I was working on is gone. V. annoying.

    I recently replaced the processor from a 2200+ to a 2800+. The errors didn't
    happen straight after this and they don't seem to happen until the machine
    has been switched on a while and warmed up. The errors don't appear to
    happen when under load, just randomly...

    Using Abit NF7-S v2.0 motherboard. PSU is an Antec Truepower 550 watts.

    Possibilities.

    -Bad CPU/CPU Cache (will consider this after all other options discounted as
    replacing HS/fan annoying.)

    -Bad memory stick. It's a Samsung DDR333 stick and has been given a bit of a
    caning I guess being run at 450mhz DDR 3 - 4 - 4 - 10 either 2.8/2.9 volts.
    What's the chance this could have developed a fault due to being overclocked
    and it would manifest itself now in BSOD's? I have used an active
    Thermaltake DDR cooling kit on the Samsung stick which I have felt and it
    never appeared to get particularly hot. A little warm not warmer.

    -Graphics card. Occasionally (instead of IRQL_etc etc) I received errors
    suggesting the graphics card was at fault (something like NV4_disp.dll could
    not complete a drawing operation) after which I clocked the card, a leadtek
    a250 ti4200 agp4X 128mb card, to default speeds.

    -Bad Motherboard - Now I certainley hope this isn't the case but I guess it
    will have to be considered if all other possibilities are discounted.
    (another possibility could be poor cooling of SB or NB? What is the System 1
    temp shown in Winbond hardware doctor?)

    -Corrupt Windows XP installation - Same as Bad motherboard but easier to
    handle I guess.


    My first feeling is the memory stick is at fault. I ran Memtest when the
    bluescreens came up last night and errors were generated (but this could
    also be due to a motherboard or cpu cache fault could it not? If the cache
    passed on bad info to the RAM then the RAM isn't at fault?!) When I get the
    chance I'll replace the Ram stick see if that helps then the GFX card.

    Anyone have any other ideas or a clear idea of where to look first for the
    problem?

    Any advice on cheap but good cooling solutions for NB and SB (at moment I
    have 3 coolermaster DDR aluminum cooler things attached to the SB and the NB
    has the supplied fan.)

    Thanks in advance and Happy New Year!!!

    Kind Regards,

    tHatDudeUK
     
    tHatDudeUK, Jan 1, 2004
    #1
    1. Advertisements

  2. tHatDudeUK

    Dr Teeth Guest

    Stop win XP rebooting so that you can actually see the BSOD would be a
    good start, as well as doing a search on Google for the error message.

    Start > Control panel > system > advanced > startup and recovery -
    settings > deselect the 'automatically restart' box in the system
    failure section.

    Good Luck.

    Cheers,

    Guy

    "If you want to find out what is wrong
    with democracy, spend five minutes with
    the average voter." - Winston Churchill
     
    Dr Teeth, Jan 1, 2004
    #2
    1. Advertisements

  3. tHatDudeUK

    RobD Guest

    If I had this problem, I would look at the bsod to determine if the same
    driver was causing the problem every time. I would also take a mini-dump
    and look at it with windbg to determine if the memory page was swapped out
    or a pointer was bad. More than likely it is a memory-related hardware
    issue, but it could be a bad driver. The bottom line is some kernel driver
    tried to access a page of memory that was swapped out.

    - Robert -
     
    RobD, Jan 1, 2004
    #3

  4. Hi,

    My feeling is that the errors are generated as a *side-effect* of using
    flaky memory, or memory that has the wrong settings. If it fails the
    default MemTest x86 then I would adjust the memory frequency/timings/voltage
    until it did pass, then I would do the *Full* MemTest x86 test for about 3-4
    passes (24Hours) and expect to see no errors.

    I recently had to trouble shoot a similar problem (IRQ_Not_Less_etc) and it
    was a memory problem.

    Good luck
    --
    Wayne ][

    Barton (AQXEA) XP2500+ @ 2.2GHz (10x220) - 1.775vCore
    CoolerMaster Aero 7 Lite - 3,200rpm
    ABIT NF7-S (v2.0 - BIOS#14)
    512MB Dual TwiSTER PC3500 @ DDR440 1:1 (9,3,3,2.0 - 2.7v)
    Sapphire Atlantis 9800 - 3.3ns Samsung (325/290 Default)
    240GB (2x120GB) WD-SE SATA RAID-0 (NTFS - 16k Stripe)
    Antec SX630II Mini-Tower Case Inc 300w PSU
    2 x CoolerMaster 80mm Blue Neon Fans
    WinXP-PRO-SP1
    Cat 3.7 - DX9.0b
     
    Wayne Youngman, Jan 1, 2004
    #4
  5. tHatDudeUK

    tHatDudeUK Guest

    The annoying thing is it's completely random and doesn't happen when under
    load as might be expected. Anyhow I've changed my memory settings and kept
    my CPU the same speed to see if its the memory or the CPU. I'll see if it
    happens again and run memtest all night and consider some new memoery
    sticks. Any recommended memory sticks? Are the twister sticks any good
    because I tried some twinmos sticks recently and they had m-tec chips on and
    were the biggest pile o **** I ever used. Didn't even run at default speed.
     
    tHatDudeUK, Jan 1, 2004
    #5
  6. I too think your memory is most likely the culprit.

    What memory timings are you currently running ? There was mention of CL3 but
    was this on another machine ?

    IME, having used 4 different nf2 boards is it doesn't like CL > 2.5
     
    binaries_freak, Jan 1, 2004
    #6
  7. "tHatDudeUK" wrote
    speed.


    Hi,
    well I was looking for the best value PC3500cas2 that's why I went for
    TwinMOS TwiSTER. Works well (very stable). Used it for a friends mobo as
    well. At 220MHz FSB I am getting 9,3,3,2.0 (Aggressive preset) using
    2.7vDimm. Passed the full MemTest x86 run over 24-36hours. Not too shabby
    timings considering that the memory is rated at 217MHz. Strangely my system
    will fail Prime95 after 24hours if I use 2.8vDimm, but all is fine @
    2.7vDimm.

    Didn't have any luck trying for 2,2,2.0 timings (at any FSB Speed), so if
    these timing are desirable to you then maybe you should be getting Corsair,
    OCZ, Mushkin etc. . but there will be a price hike!

    I'm getting 1530 in SiSoft Sandra *Unbuffered*Memtests after tweaking
    everything possible. CPU interface has to be *disabled* above 200MHz, which
    has a negative effect on benchmarks (about 50 points), but this is in turn
    *offset* by the increase in FSB from 200MHz to 220MHz. . .
    --
    Wayne ][

    Barton (AQXEA) XP2500+ @ 2.2GHz (10x220) - 1.775vCore
    CoolerMaster Aero 7 Lite - 3,200rpm
    ABIT NF7-S (v2.0 - BIOS#14)
    512MB Dual TwiSTER PC3500 @ DDR440 1:1 (9,3,3,2.0 - 2.7v)
    Sapphire Atlantis 9800 - 3.3ns Samsung (325/290 Default)
    240GB (2x120GB) WD-SE SATA RAID-0 (NTFS - 16k Stripe)
    Antec SX630II Mini-Tower Case Inc 300w PSU
    2 x CoolerMaster 80mm Blue Neon Fans
    WinXP-PRO-SP1
    Cat 3.7 - DX9.0b
     
    Wayne Youngman, Jan 1, 2004
    #7
  8. tHatDudeUK

    SteveH Guest

    Both the times I've seen that error (on my old PC and on one I was fixing
    for someone) it was down to dodgy video drivers.
    I changed the drivers and the problem went away.

    SteveH
     
    SteveH, Jan 1, 2004
    #8
  9. tHatDudeUK

    tHatDudeUK Guest

    I changed the video drivers and still had the same problem.
     
    tHatDudeUK, Jan 1, 2004
    #9
  10. tHatDudeUK

    chris Guest

    WhenI went from a thoroughbred to barton on my nf7s I had to update the bios
    to get it to work stably, no idea why but its rock solid now. I updated to
    the d20 bios, the latest at the time.

    chris
     
    chris, Jan 1, 2004
    #10
  11. tHatDudeUK

    Fishface Guest

    Fishface, Jan 1, 2004
    #11
  12. Your memory subsystem is faulty. Try another RAM stick. Set everything
    to defaults - don't overclock until you have found the fault.

    You've changed the CPU from a 2200+ (133FSB) to a 2800+ (166FSB). How
    sure are you that the memory is DDR333? Can you test it in another
    machine? Running it at 450/2.9v may well have cooked it.
    It *could* be the motherboard, but if the system was OK prior to your
    changing the CPU and you haven't fiddled with the memory, then I doubt
    it. You can try disabling L1 and L2 cache in the BIOS before running
    Memtest86 - if the faults then go away, it implies a problem with the
    CPU cache.

    BTW, you are running the latest Memtest86 (v3.0), aren't you? I found
    v2.8 could report false errors on some board/CPU combinations. Have a
    google for "Alegr Memtest" and try that - I can't comment on it as I
    have not used it.

    Have you flashed the latest BIOS and set CMOS defaults since changing
    the CPU?

    Finally, don't forget that if your memory and/or L2 cache is duff, your
    installation of Windows is suspect, as files will inevitably have been
    written incorrectly back to disk. More so if you've done a defrag.
     
    Mike Tomlinson, Jan 1, 2004
    #12
  13. tHatDudeUK

    Dr Teeth Guest

    I used to get the same error message regularly, which magically seemed
    to fix itself without any hardware changes or reconfigs.

    Cheers,

    Guy

    "If you want to find out what is wrong
    with democracy, spend five minutes with
    the average voter." - Winston Churchill
     
    Dr Teeth, Jan 1, 2004
    #13
  14. You bring up a valid point, one of the problems I have read plenty about is
    that even when the board / CPU / RAM is supposed to support upto 200 Mhz FSB
    often times many users find that switching from 133 to 166 will require a
    little increase in chipset voltage (vdd on the NF7) above stock, could be
    worth a try since this the Northbridge is now being stressed a lot more due
    to the FSB increase and it happens to be the bridge for the CPU / RAM / AGP
    buses ?
     
    binaries_freak, Jan 1, 2004
    #14
  15. tHatDudeUK

    --oZ-- Guest

    When I first got my nf7-s v2.0 I found that random reboots would fail with a
    similar message. However I tracked my problem down to a flaky lan header.
    When I turned off the onboard lan in the BIOS it would work "flawlessly"
    (with the obvious exception of networking)

    It was nothing to do with drivers as other people were using the same
    drivers as myself without a problem. End result: returned the board and
    swapped in the new one, NO reinstall of the OS and it worked ok with LAN
    enabled.

    It may not be your problem, but worth a quick restart and bios setting to
    find out I would imagine.

    As for memory types... bollox. I am running 2x 512MB samsung CL3 and it is
    fine. Some people seem to make glib statements about compatability without
    any basis in reality. If you want something approaching a compatability
    chart use this:

    http://www.abit.com.tw/abitweb/webjsp/test_report/NF7/02.php

    Hope that helps
     
    --oZ--, Jan 2, 2004
    #15
    1. Advertisements

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.