Help diagnose this unexpected crash!???

Discussion in 'General Hardware' started by CPC_RedDawn, Jan 16, 2014.

  1. CPC_RedDawn

    CPC_RedDawn Ancient Guru

    Messages:
    7,389
    Likes Received:
    7
    GPU:
    MSI GTX1080 +110/+500 H20
    Playing some DayZ my whole system just shut down and restarted completely.

    Here is the message I got when starting windows back up.

    My understanding is that this line...
    BCCode: 124

    124 means my CPU needs more vcore... is that right?

    I have stressed the CPU out completely using OCCT for 2 hours in Linpack mode, all things ticked, and 90% memory usage and max temp was 71,72,68,69 across all four cores.

    I think its just this game that does not like this overclock as I know DayZ is VERY CPU intensive.

    So does Code 124 mean I need to through a little more Vcore on the CPU.

    My overclock literally is the most simple of overclocks, just used X.M.P on my memory for 1600MHz and the correct timings and set CPU Multi to x45 and vcore to 1.2v (CPU-Z reads 1.98v). Everything else is left on auto, no thermal throttling features have been disabled either all left on auto/default position.
     
  2. ---TK---

    ---TK--- Ancient Guru

    Messages:
    22,135
    Likes Received:
    0
    GPU:
    2x 980Ti Gaming 1430/7296
    raise cpu volts a tad and test with latest p95 blend max memory
     
  3. clawhamer

    clawhamer Ancient Guru

    Messages:
    2,579
    Likes Received:
    0
    GPU:
    EVGA 780 Classified
    FC3 proved to be my toughest stress test... if I could pass 12 hours of FC3 I knew I was good to go. That was a rough week of testing :p
     
  4. -Tj-

    -Tj- Ancient Guru

    Messages:
    13,872
    Likes Received:
    9
    GPU:
    ZOTAC GX980Ti Amp!Extreme
    either more core or higher LLC or higher cpu current aka wattage (mine is in % and 120-130% is usually more then enugh) .
     

  5. Extraordinary

    Extraordinary Ancient Guru

    Messages:
    17,952
    Likes Received:
    88
    GPU:
    GTX980 SLI
    I remember back with my 965, I could pass P95 for hours, but Crysis Warhead would bluescreen me every time

    I was reading that these stress tests like P95 don't stress every instruction set on the CPU (Something like that), so when you play a game that uses those ones you didn't test, you find out you're not stable, which is why real life use is the best stress test
     
  6. CPC_RedDawn

    CPC_RedDawn Ancient Guru

    Messages:
    7,389
    Likes Received:
    7
    GPU:
    MSI GTX1080 +110/+500 H20
    On this CPU is it ok to leave on

    C1E, Thermal Monitor, C-State, etc...?

    I have them all on default, and I just ran OCCT again with 1.2v using AVX linpack, 90% memory, and after 5 minutes I got a hard reset... strange as I tested it before and it was stable!

    I tested for 30mins at 1.21v and it crashed AGAIN!

    I noticed in CPUZ that the vcore was bumping up to 1.238v (obviously LLC doing its thing)

    I then tested at 1.24v and nows its passed OCCT for 30mins. CPUZ showed 1.26v being used, again this is probably LLC doing its thing.

    1.24v is a MASSIVE increase from what I had of 1.2v which I swear to god was stable before!
     
  7. clawhamer

    clawhamer Ancient Guru

    Messages:
    2,579
    Likes Received:
    0
    GPU:
    EVGA 780 Classified
    Yeah, there is no substitute for real usage over time, stress tests are just an indication.

    I wouldn't bother running Prime95 on Haswell either.

    Are you running adaptive or manual voltage?
     
    Last edited: Jan 16, 2014
  8. CPC_RedDawn

    CPC_RedDawn Ancient Guru

    Messages:
    7,389
    Likes Received:
    7
    GPU:
    MSI GTX1080 +110/+500 H20
    Manual voltages.

    My Mobo has three options in the Vcore type.

    "Override" or "Adaptive" or "Auto"

    This is left on Auto.

    Just disabled C-State (which also disabled C1E?) and disabled Thermal Monitor as I do that my self.

    CPUZ now shows 4.5GHz @ 1.248v (1.24v set in BIOS)

    Also for the life of me I can not find LLC in the BIOS at all! I think the board has its own way of applying this settings and I think its on by default.

    Checked the manual and its not there either.

    Right now its stable,

    4.5GHz (45x100)

    C1E, Speedstep, Turbo Boost, C-State, Thermal monitor, etc ALL DISABLED

    Vcore = 1.24v

    XMP = Enabled

    Temps are fine too idleing 25,26,24,26 after 30mins in OCCT its 77,74,79,77

    All this for ONE game LOL. It was completely stable at 1.2v in every game I tried and that's over 120 steam games, and 20 installed games (INCLUDING FC3 LOL).

    DayZ you bitch.

    Will leave this settings on now and see how things go over the next few days if it fails again I will start from scratch.
     
  9. rflair

    rflair Don Commisso Staff Member

    Messages:
    3,771
    Likes Received:
    4
    GPU:
    GTX 980/290X
  10. CPC_RedDawn

    CPC_RedDawn Ancient Guru

    Messages:
    7,389
    Likes Received:
    7
    GPU:
    MSI GTX1080 +110/+500 H20
    Yep I use that also, and it ran perfectly fine for 10 runs straight at 1.2v but DayZ just does not like it at all.

    By the way I was playing normal DayZ and NOT DayZ standalone.

    EDIT

    RFLAIR how did you let it run for 20 runs? The counter only goes to 10..
     
    Last edited: Jan 17, 2014

  11. rflair

    rflair Don Commisso Staff Member

    Messages:
    3,771
    Likes Received:
    4
    GPU:
    GTX 980/290X
    Was remembering off the top of my head, but your right.
     

Share This Page