Advertisement
If you have a new account but are having problems posting or verifying your account, please email us on hello@boards.ie for help. Thanks :)
Hello all! Please ensure that you are posting a new thread or question in the appropriate forum. The Feedback forum is overwhelmed with questions that are having to be moved elsewhere. If you need help to verify your account contact hello@boards.ie

Problem OCing all of a sudden

Options
  • 29-07-2008 1:12pm
    #1
    Closed Accounts Posts: 41


    Hey, can anyone help?

    I built myself a machine a few months ago, a Q6600, overclocked it to 3.2GHz, had it running without a glitch for 24hrs running Prime95... so I considered it 'stable'.

    Since then its been running at full load for sometime days, sometimes a week or two, and all seemed fine.

    Then I get itchy fingers, and try replacing the cooler to see if I could make the thing a little quieter. Wasnt any quieter so tried to go back to the old cooler.

    But I can seem to get it stable for more than 8-10 hours running Prime95 at any speed anymore - not even the stock 2.4GHz!!

    I've reapplied thermal paste 4 times now using various suggested methods. Ive tried upping the voltages to their rated limits. Ive tried dropping the RAM speed way down.

    I think I had it running at 2.4GHz for maybe 20 hours with reduced ram speed, so I tried it at 3GHZ last night, again with ram way below spec, but I woke up to find Prime95 gone - crashed.

    Right now its running a mem check.

    I suspect the ram isnt the problem. I'd rather not have to go out and buy any replacements, but if I do, I'd rather not have to buy more than I have to. Any suggestion on how to identify exactly what is the problem. Or any suggestions what might be wrong?

    If it helps... its a Q6600 G0 stepping, Abit IP35-Pro, OCZ 8500 Reaper HPC 2x 2GB kit. Andy Samurai Master Cooler.

    Cheers!


Comments

  • Closed Accounts Posts: 12,401 ✭✭✭✭Anti


    Well for starters that cooler isnt really upto cooling a overvolted quad, but it should be alright. But it could be faulty ram, unstable rails on the psu. Id start off with changing the bios on the board to the latest or best one.

    What temps are all 4 cores hitting after say 20 mins of 2 instances of orthos, as iirc orthos cant max out 4 cores unless 2 instances are running with the correct affinities set.

    Are you sure your not using too much or too little thermal paste ?

    This is how much that should be used:

    z48ljaaj0ujqbqex64s.jpg

    pxabhfeabqvyj7wml1p8.jpg


  • Closed Accounts Posts: 41 shepherd


    thanks

    there are a few different way suggested to apply paste. one way says a thin line of paste down the center of the cpu. the other says spread it out. ive tried both.

    i use prime95 for stability check since one instance will runs across four cores.

    the temps i get depend on the speed i clock it at. at standard 2.4Ghz it runs maybe 48-52 degrees. At 3Ghz its maybe 55-59. At 3.2Ghz its maybe 65 ish. I think those are all ok since the spec specifies a limit of 71 degrees i think.

    if its faulty ram, I suppose i could test one at a time maybe? But I thought running ram way below spec would rule out the ram though.

    I'll try new bios.

    But how do I test unstable rains on psu without getting a new/borrowing another psu???

    Can I test the cpu without stressing ram?

    Any other suggestions?

    Thanks again.


  • Closed Accounts Posts: 12,401 ✭✭✭✭Anti


    A raytracing benchmark will only test the cpu. Testign the psu will be a bit hard, but think abit have a voltage monitor will will measure vcore, 3.3v 5v and 12v rails. ITs usually built into the temprature monitor software that comes with the motherboar. Uguru or something.

    65c is getting a bit toasty, id prefere not to go over 60 but as you said it is in tollerance.


Advertisement