Overclocking Assistance with an unstable overclock

Discussion in 'Hardware' started by c-hri-s, 15 Mar 2012.

  1. c-hri-s

    c-hri-s What's a Dremel?

    Joined:
    15 Mar 2012
    Posts:
    4
    Likes Received:
    0
    Hi,

    I've bought a pre-overclocked system from PC Specialist, and it seems to be suffering much the same as the Scan rig that was mentioned in this thread: http://forums.bit-tech.net/showthread.php?t=224056

    I don't know too much about overclocking, hence me originally paying someone to do it for me, but I was wondering if I could learn from your expertise!

    My system has been running okay in the couple of weeks that I've had it with just a few occasional bluescreens. It will complete 3DMark benchmarks with no issue, however it's always bluescreened after only 20 seconds or so of running IntelBurnTest ("A clock interrupt was not received on a secondary processor within the allocated time interval"). The same thing happens with Prime95, but you need to leave it a couple of minutes.

    The system is as follows:

    • Overclocked Intel Core I7-2600K Quad Core (3.40Ghz @ Max 4.80Ghz)
    • Asus® Sabertooth P67 (New Rev 3.0)
    • 16Gb Kingston Hyper-X Genesis Dual-Ddr3 1600Mhz, X.M.P (4 X 4Gb Kit)
    • 1.25Gb Nvidia Geforce Gtx 570
    • Corsair 650W Tx650 V2-80 Plus® Bronze
    • Coolit Eco Ii A.L.C Fat Boy (Advanced Liquid Cooler)

    And these are the OC settings in the BIOS:


    • AI Overclock Tuner: manual
    • BLCK Freq: 100.00
    • Turbo Ratio: enabled (all cores)
    • By All Cores: 46
    • Internal PLL OverVoltage: enabled
    • Mem Freq: 1600MHz
    • EPU Power Saving: disabled
    • Load-Line Calibration: high
    • VRM Freq: auto
    • Phase Control: extreme
    • Duty Contorl: extreme
    • CPU Current Capability: 110%
    • CPU OverVoltage: manual mode
    • CPU Manual Voltage: 1.40V
    • DRAM Voltage: 1.650V
    • VCCSA Voltage: 1.1250V
    • VCCIO Voltage: 1.150V
    • CPU PLL Voltage: 1.81250V
    • PCH Voltage: auto
    • CPU Spread Spectrum: disabled
    • CPU Ratio: auto
    • Intel Adaptive Thermal Monitor: enabled
    • Intel Virtualization: disabled
    • Intel SpeedStep: enabled
    • Turbo Mode: enabled
    • AMD Turbo Core Technology: disabled

    Temperatures seem reasonable. It runs about 32 degrees, going up to about 66 running Prime95 on 'full heat mode'.

    I'd appreciate any advice.
     
    Last edited: 15 Mar 2012
  2. rollo

    rollo Modder

    Joined:
    16 May 2008
    Posts:
    7,887
    Likes Received:
    130
    Does it blue screen in games or intensive CPU tasks, if so I'd be contacting the vendor.

    Your timings look fine as does CPU voltage if i was checking I'd start with the memory
     
  3. c-hri-s

    c-hri-s What's a Dremel?

    Joined:
    15 Mar 2012
    Posts:
    4
    Likes Received:
    0
    It seems fine in games - and in general use, although I've had it BS a couple of times when using general apps, mostly when I've been doing a bunch of stuff at the same time.

    I tried dropping the CPU voltage to 1.35 and changing the "By All Cores:" value from 46 to 42 and it seems to have improved things. IntelBurnTest doesn't bluescreen in the five minutes I've left it running for ... does that help any?

    Obviously it only ramps up to 4200Mhz now rather than the 4600 it went to previously. Temperatures seem much the same.
     
  4. S1W1

    S1W1 Minimodder

    Joined:
    25 Mar 2011
    Posts:
    235
    Likes Received:
    29
    PC Specialist have a duty to fix this.

    Get in touch with them, tell them your problem and they will no doubt offer to take the PC back at their cost and tweak the BIOS settings to achieve more stability. Scanning over your settings everything looked OK so it may be the case that the CPU is faulty.

    Either way PCS should be able fix it if you send it back for repair.
     
    Last edited: 15 Mar 2012
  5. c-hri-s

    c-hri-s What's a Dremel?

    Joined:
    15 Mar 2012
    Posts:
    4
    Likes Received:
    0
    To be honest they may well suggest that - but I'd rather not be without it if it's something I can get sorted, especially now all my data is on it :(
     
  6. boltonuk007

    boltonuk007 What's a Dremel?

    Joined:
    29 Sep 2011
    Posts:
    441
    Likes Received:
    13
    I have a friend who bought an i5 2500K, pre overclocked to 4.5. He had problems with BSOD, random at first and then more frequent and after changing the rest of the hardware except the pre overclocked mobo/cpu and ram. I dropped the OC to 4.2 and he has had no problems since. Some chips just cant cut it, even of the same type. My 2500K runs happily at 4.8, no probs. I would try to drop the multiplier by 2 or 3 and see if it improves things, you wont notice and real difference in performance, just in benchmarking.
     
  7. thetrashcanman

    thetrashcanman Angel headed hipsters

    Joined:
    18 Nov 2010
    Posts:
    2,716
    Likes Received:
    76
    I would very much agree with this, the chip probably needs more volts, or to drop the multiplier, I would suspect the company you purchased it from didn't want to put more the 1.4v through the chip (intel's supposed limit).

    Either that or a mixture of they didn't do there testing properly.
     
  8. boltonuk007

    boltonuk007 What's a Dremel?

    Joined:
    29 Sep 2011
    Posts:
    441
    Likes Received:
    13
    they probably put together a hundred a week, same hardware, same settings, the odd chip cant take it. I dont believe they put each one through a 24 hour burn test.
     
  9. c-hri-s

    c-hri-s What's a Dremel?

    Joined:
    15 Mar 2012
    Posts:
    4
    Likes Received:
    0
    Mmm, just turned it up to 44/1.350 and it ran a complete IntelBurnTest sequence on High.

    Perhaps as you say 46 was just that little too much. I think for the sake of a very slight bit of performance that I'll probably never notice, I'll stick at 44 rather than trying to get 46 stable and pushing the voltage past 1.40.

    Thanks to all for your advice. Much more use than the vendors forums.
     

Share This Page