Computer can't keep OC anymore.

Discussion in 'Overclocking & Cooling' started by vegeta535, Jul 29, 2019.

  1. vegeta535

    vegeta535 2[H]4U

    Messages:
    3,051
    Joined:
    Jul 19, 2013
    So over the past couple weeks my computer will not boot anything but stock settings. Even rasing memory about the base 2133 would cause to get the failed to apply OC setting. This is with a MSI x299 tomahawk with a 7820x and g.skill 3200 c17 memory. Even a slight bump OC will cause it to fail to boot. I also seem to be getting hicups during gaming at stock setting. I am using a Seasonic FOCUS Plus Series SSR-850FX 850W 80+ Gold PSU which is only about 2 years old. I haven't been too happy with the MB since I got it but which would be the issue. The PSU or MB?
     
  2. VanGoghComplex

    VanGoghComplex [H]ard|Gawd

    Messages:
    1,713
    Joined:
    Apr 5, 2016
    What kind of overclock were your getting before? How's your cooling system performing?
     
  3. vegeta535

    vegeta535 2[H]4U

    Messages:
    3,051
    Joined:
    Jul 19, 2013
    I use to sit at 4.8 at 1.25v. Using custom water loop with temps going to 80s under prime and other heavy loads. Didn't get past 60s in games and stuff. Temps under stock dont go past 50s under normal loads. Idles around 30c.
     
  4. hititnquitit

    hititnquitit Limp Gawd

    Messages:
    438
    Joined:
    Apr 16, 2018
    How do the rails on your psu look? Ideally loaded but thats probably not feasible right now.
     
    vegeta535 likes this.
  5. vegeta535

    vegeta535 2[H]4U

    Messages:
    3,051
    Joined:
    Jul 19, 2013
    Under HWinfo I get 5.0, 3.312 and 12.06. So verything looks ok under stock settings. MB might just be poop.
     
    Last edited: Jul 29, 2019
  6. pendragon1

    pendragon1 [H]ardForum Junkie

    Messages:
    13,278
    Joined:
    Oct 7, 2000
    test/replace the cmos battery then toss more voltage at the cpu and ram.
     
    vegeta535 likes this.
  7. vegeta535

    vegeta535 2[H]4U

    Messages:
    3,051
    Joined:
    Jul 19, 2013
    Ill try the battery tomorrow but yea I cranked the voltage and didn't change. Went up to 1.35 on cpu and 1.5 on the ram and still fails.
     
    pendragon1 likes this.
  8. Gamer X

    Gamer X Limp Gawd

    Messages:
    288
    Joined:
    Jul 5, 2019
    How old is the PSU?
     
  9. DooKey

    DooKey [H]ardness Supreme

    Messages:
    8,038
    Joined:
    Apr 25, 2001
    Smells like a failing motherboard to me.
     
    vegeta535 and dragonstongue like this.
  10. vegeta535

    vegeta535 2[H]4U

    Messages:
    3,051
    Joined:
    Jul 19, 2013
    PSU is not even 2 years old yet.

    That is where I am leaning towards. It failed to boot this morning at stock setting this morning. Another reboot got it into windows afterwards tho.
     
    dragonstongue likes this.
  11. Nenu

    Nenu [H]ardened

    Messages:
    18,789
    Joined:
    Apr 28, 2007
    First instincts are motherboard issue or CPU issue.
    Its possible the memory controller on the CPU has a problem.
    Or perhaps a corrupt cmos.

    I would reset the cmos and start from scratch, dont use old settings, work them out again.
    If you still get issues, try with a single memory stick to see if that does any better.
    Not as a permanent solution but to help troubleshoot.
     
    vegeta535 likes this.
  12. vegeta535

    vegeta535 2[H]4U

    Messages:
    3,051
    Joined:
    Jul 19, 2013
    Yea I already tried resetting the cmos. Any change beyound stock settings cause the system to fail booting. I hope it is not the CPU or otherwise I am just going to have to scrap they system since 7820x are so expensive now.
     
  13. DooKey

    DooKey [H]ardness Supreme

    Messages:
    8,038
    Joined:
    Apr 25, 2001
    That CPU is still under warranty. Intel is good about sending out stuff quickly in my experience.
     
    vegeta535 likes this.
  14. sirmonkey1985

    sirmonkey1985 [H]ard|DCer of the Month - July 2010

    Messages:
    21,359
    Joined:
    Sep 13, 2008
    question is how long have you run that overclock? if it's been for a year or two then it could be the board especially with it being a custom loop i'd assume there's not a whole lot of airflow around the CPU socket. probably to many of the solid caps have failed and there's just enough to keep it running. that's my assumption though, just odd that overclocking the memory is failing as well, so it is possible that it's the memory instead or there's something else failing on the board.
     
    vegeta535 likes this.
  15. vegeta535

    vegeta535 2[H]4U

    Messages:
    3,051
    Joined:
    Jul 19, 2013
    Not when it is delided =/.
     
  16. vegeta535

    vegeta535 2[H]4U

    Messages:
    3,051
    Joined:
    Jul 19, 2013
    I tried different combos of the 4 sticks I have and the results are all the same. Even with 1 stick it does it. Airflow should be fine. I have 2 Noctua fans intake and 3 out on 2 280 rads. GPU is also under water so it is not dumping hot air into the case either.
     
  17. vegeta535

    vegeta535 2[H]4U

    Messages:
    3,051
    Joined:
    Jul 19, 2013
    I had the system for about 1 1/2 years. I didn't keep the OC 24/7. Only when I going to be gaming. About a month ago I started getting the failed OC and ended up backing the memory down to 3000 allowed it to boot. I haven't been gaming that much recently so the CPU has been running stock. It is just yesterday I couldn't boot even with unless memory was at base 2133. I couldn't get cpu to boot with over 4ghz all core either at that point. I am at work now but like I said earlier it failed to boot this morning even on all stock settings. So it is either the CPU or MB. I can get a Asus prime deluxe for $180 so I might go give that a shot if I can't get anything going today after work.
     
  18. Nenu

    Nenu [H]ardened

    Messages:
    18,789
    Joined:
    Apr 28, 2007
    Booting is a fairly light load for your PSU, it should be handled with ease. Your issue occurs while booting.
    If you have a multimeter, measure the 12V rail while the computer boots from power off.
    If you see a change above 0.1V once it has settled on its 12V value, the PSU is suspect.
     
    vegeta535 likes this.
  19. dragonstongue

    dragonstongue 2[H]4U

    Messages:
    3,113
    Joined:
    Nov 18, 2008
    what about Hard drive / storage?

    A bad drive / failing drive can easily fubar with things when the rest of stuff "might not be bad"


    sounds to me is very likely the motherboard (always cooled well enough without cooking the shit out of the motherboard while overclock the "system") ?


    PSU "flaking" IMO not as likely to be the culprit (if you not pushing above ~85% max of rated load on "name brand" PSU OEM/maker .. unless bad cable from PSU to a different faulty part like a sata power cable being bent/our of whack eventually cause system to freak... As well (just to say as was on my mind ish) since Intel since Core i first launched (core i7-950 as example) their voltage control and "happy spots" is much more tightly controlled or it "freaks out" ..


    I know a few buds various generations of, the voltages etc need to be "just right" or X amount of time (seconds, minutes what have you) insta power down /refuse to power up an so forth.. though this affects all systems slightly different as all systems are not "clones"


    hope can get sorted out, system down sux [H]ard not beneficial to anything but a game of patience o_O
     
    vegeta535 likes this.
  20. vegeta535

    vegeta535 2[H]4U

    Messages:
    3,051
    Joined:
    Jul 19, 2013
    I change the cmos battery. Changed PSU cables and still a issue. I did notice when I went into the case the vrm heatsinks were burning hot after being on for a bit at stock settings. I am positive that it is the MB at this point. I'm going to swing by MC and grab a new MB this weekend. Saw a open box ASRock i9 for $120.
     
    pendragon1 and thebufenator like this.
  21. thebufenator

    thebufenator [H]ard|Gawd

    Messages:
    1,182
    Joined:
    Dec 8, 2004
    Good luck. It's great having a MC nearby ain't it?
     
    vegeta535 likes this.
  22. vegeta535

    vegeta535 2[H]4U

    Messages:
    3,051
    Joined:
    Jul 19, 2013
    Sure is. I hardly buy anything computer wise on the internet anymore.
     
    Last edited: Aug 1, 2019
  23. vegeta535

    vegeta535 2[H]4U

    Messages:
    3,051
    Joined:
    Jul 19, 2013
    Well I changed out my MB to a ASRock i9 and all is well now. Except I have to chop the corner of my radiator fan on top to clear the damn io shied guard. Still going to spend some time tuning but I am able to boot with ram at 3200 speed an a slight OC to 4.5ghz. Such a pain working in system with a custom loop without breaking it down lol.
     
    Last edited: Aug 4, 2019
  24. vegeta535

    vegeta535 2[H]4U

    Messages:
    3,051
    Joined:
    Jul 19, 2013
    So I had a look at the my old MB and there is a greasy film around the vrm heatsinks. On both sides of the MB. Guess one or more capacitor shit the bed? Visual I don't see any damaged caps.
     
  25. SvenBent

    SvenBent 2[H]4U

    Messages:
    2,976
    Joined:
    Sep 13, 2008
    Why is the change of cmos battery advises so often when it has such a little purpose and not impact on overclocking?
    You dont eveb need tge freaking battery for computer operations as long as you dont have an power outage or put you computer in G2 power mode
     
  26. Azrak

    Azrak Gawd

    Messages:
    820
    Joined:
    Oct 4, 2015
    Doubtful it's from any caps. Usually the grease/film is from the thermal pads between the VRM and heatsinks. They have silicone grease/oil inside them and it oozes out over time.
    That's not to say that the caps haven't failed - it's just harder to tell because they are the "solid" kind. If the VRM is getting really hot and you know it did not used to when you first built the system then something has failed somewhere causing the VRM to work harder and it usually is a cap somewhere.

    There are some cases where a failing CMOS battery causes the BIOS to go all wonky and stop working properly, which causes the whole system to act strangely. I agree it doesn't make a lot of sense, but I think it's more than an old wives tale. There have been a few cases here on [H] where replacing the CMOS battery fixed issues. It's rare, but I guess it's a thing sometimes. <shrug>
     
    vegeta535 likes this.
  27. SvenBent

    SvenBent 2[H]4U

    Messages:
    2,976
    Joined:
    Sep 13, 2008
    In the aprox 10k machines ive have hardware diagnosed this has never been the case.
    Power form the battery is only needed if the pc is in G@ mode aka mechanical off. Turning of the pc (front of chasises )does not need the battery to keep cmos settings. You have to unplug it.

    Any power states above G2 that includes standby hibernation and soft-off all retains cmos setting with no battery at all
     
  28. GoldenTiger

    GoldenTiger [H]ard as it Gets

    Messages:
    18,662
    Joined:
    Dec 2, 2004
    I've had one of my own quit booting and swapping the cmos battery is what fixed it.
     
  29. STEvil

    STEvil 2[H]4U

    Messages:
    2,819
    Joined:
    Oct 17, 2000
    95% chance ram went bad. Very typical for bad ram to cause these issues.

    Lack of boot due to bad CMOS battery is just usually due to settings not holding between POST cycles. Some boards unlatch CMOS power before POST cycle to help load correct settings and keep soft settings/data from hanging around.
     
  30. funkydmunky

    funkydmunky 2[H]4U

    Messages:
    2,336
    Joined:
    Aug 28, 2008
    Why wasn't a clean-up and re-paste suggested first?
    FIRST THING TO DO!
     
  31. vegeta535

    vegeta535 2[H]4U

    Messages:
    3,051
    Joined:
    Jul 19, 2013
    Have no issues so far after the MB swap. Been playing oc last night and was able to settle at 4.8ghz again @1.23v. I relied it while I had it apart and temps don't go over 80c under burn-in tests with fans running at low speed so I am happy with results. All the hiccups I have been getting in games and so far no random crashes either.
     
    Last edited: Aug 6, 2019
  32. sirmonkey1985

    sirmonkey1985 [H]ard|DCer of the Month - July 2010

    Messages:
    21,359
    Joined:
    Sep 13, 2008
    given he has 3000+ posts first assumption would be he's probably not an idiot and thought of that already.. ;) but really most of the signs pointed to other things being the problem given the memory clock issues that it was highly unlikely to be a paste related.

    glad to see it's working again.
     
    vegeta535 likes this.