also @ TechSpot: Tea Party Republicans and 'liberal weenies' alike celebrate Texas email privacy law

Games Suddenly Crash

Discussion in 'Gaming' started by slugonamission, Aug 11, 2005.

  1. slugonamission Newcomer, in training Posts: 29

    OK, ive just built a new PC, heres the specs

    500W PSU
    120GB HDD
    40GB HDD
    Creative Sound Blaster Live!
    AMD Athlon 64 3000+
    Sparkle NVidia 6600GT
    512Mb RAM

    Im having a problem. Games have just started crashing on me, battlefield 2 used to work fine, now i can play it for a minute or two, then it crashes my computer completely (not to desktop, it does a straight restart). This is the same with 3DMark05. Again, it used to work fine. I dont think its an overheating issue, although i have no sure way of finding out. Also, i have recently re-installed directX 9 (these problems seem to have been occouring since installing UT2k4), but reinstalling DX9 doesnt seem to have helped the problem.

    Any ideas would be welcome.
  2. DonNagual TechSpot Ambassador Posts: 3,564

    Are you sure you don't have heat problems?

    Try installing this: http://www.techspot.com/downloads/26-motherboard-monitor.html

    Also, just to stick to the basics, you do have the latest bios for your motherboard as well as all the latest drivers for mobo, vid card etc., right?

    Have you installed any new hardware lately?
  3. slugonamission Newcomer, in training Posts: 29

    actually, im getting to think this is a heat issue. The nights here seem to be getting hotter, and it hasnt happened all day.

    I cant install the latest BIOS - no floppy drive unfortunately, but i do have latest GFX card drivers

    Just tried again, 3D mark 05 crashed again. Seems to be a minute or so in. The screen went all like, blocky, but not like usual pixellation. There were a few pixels in strange colours. Thats really the best way to describe it. Anyway, i will try again tomorrow, and try that motherboard monitor you posted.

    Thanks, slug
  4. DonNagual TechSpot Ambassador Posts: 3,564

    Yah, it really does sound like a heat problem. That pixelation you describe is a good indicator. I'd fix your heat issues asap, or you are going to damage your hardware (if you haven't already).

    Good luck!
  5. slugonamission Newcomer, in training Posts: 29

    ok, ive come to the conclusion that it is a heating issue. This time, 3dmark05 got onto the 2nd test, then bailed near the end of that, this is in the morning here, when it seems to be a bit cooler than at night.

    So then, whats the best cooling method?
  6. slugonamission Newcomer, in training Posts: 29

    BTW, 3dmark had frozen when it came up with that strange pixellation stuff
     
  7. DonNagual TechSpot Ambassador Posts: 3,564

    But the problem is, at this point we don't know WHY you are overheating. (Let alone WHAT is overheating).

    It may be just your video card, but your rebooting sounds like your CPU is the guilty one. That pixelation has me confused. We need someone with more experience to give some imput.

    Best case scenario, you add more (or more powerful) case fans and that fixes the problem.

    But it is possible that your current CPU heatsink fan needs to be re-seated with some new thermal paste. If you have never done this before it can be tricky.

    And if you are going to go that route (removing your CPU heatsink), you may as well plop a better heatsink on there while you have the stock one off. A good heatsink on the CPU makes a HUGE difference.

    There are many good makers out there these days. Are you using a socket A motherboard? Check out the Zalman fans. They are pretty good.
  8. DonNagual TechSpot Ambassador Posts: 3,564

    By the way, what bios are you are running? That is another possibility worth looking at.

    You mention that your vid drivers are up to date, what about your motherboard drivers?
  9. slugonamission Newcomer, in training Posts: 29

    nope, its a new socket 939 one

    ive noticed a couple of times when it has crashed, the monitor has gone into standby, but the sounds still work for a bit, then it crashes. Also, after crashing, it seems to take a little longer to go through POST than it does normally. After crashing though, ive looked through my BIOS, and the CPU has been running at around 50-55...
  10. slugonamission Newcomer, in training Posts: 29

    Do you mean BIOS version? Im not sure, i'll reboot and find out for you.

    As for the VIA 4-in-1 drivers, there doesnt seem a new version for them out...
  11. slugonamission Newcomer, in training Posts: 29

    ok, just taken case side off, it feels pretty cool inside there. Also, ran 3dmark with it off, it crashed (again). Felt my graphics card's heatsink. It was warm, not hot or anything. Im starting to wonder if it is actually a cooling problem....
  12. DonNagual TechSpot Ambassador Posts: 3,564

    I'm starting to agree with you. If your CPU is only getting to 55c, that is nowhere near hot enough to be shutting you down.

    There are those in this forum who can help, but they are not a jedi like I am. You must turn to the dark side, as I jedi can not teach you what you need to know...

    In the meantime, I would run a diagnostic on your hard drive, and on your memory to rule that out. Go to the Windows OS section of this forum, and have a look at Rick's sticky about getting windows to work. In the first post he describes how to check HDD and memory. I think the best thing to do for now is just start ruling things out, one at a time.

    P.S., if you are running an old bios..... it could be as easy at that. :eek:
  13. slugonamission Newcomer, in training Posts: 29

    ok, thanks for your help

    Ive also posted on the nVidia forums. It seems that the 77.72 drivers can be a bit buggy (or apparantly so anyway) so i waiting from a suggestion from there.

    Thanks again, Slug

    BTW, approx what temprature would cause a crash in the GPU and CPU?

    EDIT: Theres new nvidia drivers out (77.77), im gonna try them now...
  14. DonNagual TechSpot Ambassador Posts: 3,564

    It depends on your motherboard settings in most cases for when it will auto shut-down, but I think 70c is getting to the danger point. I get worried when my CPU reaches the 60c mark.

    For GPUs, I am learning first hand that they burn a lot higher. Depends on your card of course. Mine is a 6800GT, and it runs at 60c idle, 70c at full load. Yeeesh.
  15. slugonamission Newcomer, in training Posts: 29

    just been stress testing my CPU with prime 95, its running at about 61-63 - fan running at around 4000RPM. Its been running for about 50 mins now with no errors
  16. slugonamission Newcomer, in training Posts: 29

    ok then, just installed rivatuner for debugging purposes (just so i can have stat logging)

    Basically, under load, my GPU is running at 66 degrees. Core clock under load is 501.43 and the memory clock is 1002.38. Before i started 3dMark05, the core clock was 300.86 though, and the temp was 49.00
  17. olefarte TechSpot Ambassador Posts: 1,425

    If you have "Cool and Quiet" turned on, go in the bios and turn it to off. See if that helps.
  18. slugonamission Newcomer, in training Posts: 29

    AMD K8 Cool and Quiet control? It was set to auto, just disabled it
  19. slugonamission Newcomer, in training Posts: 29

    well, it seems to have sorted it :)

    Just run 3dmark, no problems. Thanks to you two for helping me sort it our

    I'll be back if it occours again...

    Thanks again, Slug
  20. DonNagual TechSpot Ambassador Posts: 3,564

    olefarte, just out of curiousity, why was cool n' quiet causing him to overheat?