Computer Hardware Clusterfuck

Discussion in 'OT Technology' started by red, Nov 28, 2009.

  1. red

    red New Member

    Joined:
    Jun 30, 2005
    Messages:
    87,095
    Likes Received:
    0
    So I'm trying to diagnose my machine. I started getting random bluescreens a while back, and I assumed it was a RAM problem based on the random nature of the occurrence, and whatever the error message was. I confirmed this by running a memory test and getting a buttload of errors. Oddly, I also noted that any time I unseated an re-seated the RAM, the problems (and the errors) would go away for about a day. Anyway, I bought new RAM, and called it a day.

    A few days after I replaced the RAM, I started getting hard lockups. No bluescreen, the screen would just freeze (and whatever sound had been playing would have the last quarter-second repeated indefinitely).

    I thought, "well fuck, if it wasn't a RAM problem, the motherboard must be toast. What else would produce RAM errors?"

    So I replaced the motherboard. The (new) RAM in the new motherboard tests A-OK hunky-dory in memtest86, but the lockups are still happening.

    Processor? Video card? HD seems to be fine. Machine runs just fine most of the time (until it locks up). I can't think of how I can test those components individually without replacing them, and the cost of doing that is going to get really annoying really fast. I tried running a surface-test on the HD, just to confirm, but the machine inevitably locks up before it can finish (though I never saw any errors on the test).

    I also have a spare processor/mobo/vid-card combo, but none of the old shit is interchangeable. I could swap the whole shebang out and see if it works, but I suppose it would only confirm that it's not the hard drive, and I'm getting sick of swapping out motherboards.

    Anyone have any bright ideas?
     
  2. red

    red New Member

    Joined:
    Jun 30, 2005
    Messages:
    87,095
    Likes Received:
    0
    I could also add that, prior to this pain in the ass issue, the whole setup had been running fine for a year or so.
     
  3. Stilgar1973

    Stilgar1973 New Member

    Joined:
    Aug 12, 2006
    Messages:
    8,533
    Likes Received:
    0
    Can I make the assumption that you have cleaned out all the dust from the CPU and video card?

    This COULD be a heat problem. Get some new thermal paste, unseat the processor, put down new thermal paste, put the fan back on.
     
  4. red

    red New Member

    Joined:
    Jun 30, 2005
    Messages:
    87,095
    Likes Received:
    0
    It's not a heat problem. For one thing, I can monitor the temps on everything, and they're low and stable. The CPU was freshly mounted on the new mobo
     
  5. Frequency

    Frequency New Member

    Joined:
    Dec 30, 2004
    Messages:
    7,504
    Likes Received:
    0
    Location:
    PA
    heat was my first thought, so that's 2 votes for heat
     
  6. Yop

    Yop New Member

    Joined:
    May 26, 2006
    Messages:
    620
    Likes Received:
    0
    Location:
    Oviedo
    Heat or Possibly a memory leak can cause a Hardlock.

    It sounds like since you've replaced the mobo and it still happens that its pointing towards a memory leak. I would try reformating and then go from there.
     
  7. red

    red New Member

    Joined:
    Jun 30, 2005
    Messages:
    87,095
    Likes Received:
    0
    It's not a heat problem. That would be easy to diagnose. I can monitor temps a bazillion ways.

    And a memory leak would imply that I've installed some new application with a memory leak in it. I think I would notice if the machine only crashed after something gobbled up 2 gigs of memory.
     

Share This Page