BSOD after CPU swap - eviltoast

I’ve swapped a CPU going from 5600g to 5900x, unfortunately the system seems to bluescreen from time to time (usually takes hours in-game, otherwise stable)

For some reason it gets slightly worse when I enable XMP. Significantly worse if I undervold the CPU even a bit. Temps go no further than 80-85C under full load.

Would appreciate your thoughts on potential reasons.

Specs:

  • 5900x
  • B550m DS3H (Swapping tomorrow to B550 Tomahawk)
  • 3600Mhz 2x16Gb Kingston Fury (2400mhz if JEDEC)
  • 6700xt Saphire Pulse
  • 750W Zalman GigaMax

Will also be reinstalling Windows after motherboard swap.

  • bassman2710@kbin.social
    link
    fedilink
    arrow-up
    8
    ·
    1 year ago

    I would recommend reseating before reinstalling windows. Is your cooling solution also capable for a high tier chip?

    • Mistic@lemmy.worldOP
      link
      fedilink
      English
      arrow-up
      1
      ·
      1 year ago

      I’ll try that when swapping mobos.

      Yes, the cooling is enough. I did say 85C was absolute max out of every core. And that’s without fans ramping up much.

  • _thrax@lemmy.world
    link
    fedilink
    English
    arrow-up
    4
    ·
    1 year ago

    First thing that comes to mind is to maybe try flashing/updating your bios? Obviously you have a version compatible with 5000 series CPU’s, but perhaps there is a newer version with bug fixes for this issue?

    Hope to see an update if you figure it out. Good luck!

    • Mistic@lemmy.worldOP
      link
      fedilink
      English
      arrow-up
      3
      ·
      1 year ago

      Could be, there are newer versions that I didn’t bother upgrading to (namely the security patch after a big loophole was found)

      If that’s the case, mobo swap will be solving it regardless then. Will be updating the post if I find the root of it.

  • Jubas@lemmy.world
    link
    fedilink
    English
    arrow-up
    2
    ·
    edit-2
    1 year ago

    Reseat CPU and RAM, clear all BIOS settings to factory defaults, update BIOS (if there is any new BIOS available) should be steps that you hopefully have already performed?

    Also make sure that if you have chipset drivers installed that they’re the newest version possible.

    If things are stable, then you can start looking at running at XMP speeds and otherwise overclocking/undervolting/w.e.

    • Mistic@lemmy.worldOP
      link
      fedilink
      English
      arrow-up
      1
      ·
      1 year ago

      Motherboard swap is something I wanted to do regardless. Especially since I have a good use for the old one.

      I will try different RAM later on. Although it wasn’t causing issues with 5600g.

  • kryptonianCodeMonkey@lemmy.world
    link
    fedilink
    English
    arrow-up
    2
    ·
    1 year ago

    Overheating, maybe? Are your fans picking up right before it happens? You might run with a temperature monitor app to watch for rising Temps.

    If that’s the issue, check your cooler. Make sure your thermal paste is applied and has good coverage. If using a pre-applied thermal pad that came on the cooler, make sure you pulled off the plastic cover (a common mistake). Make sure you didn’t leave your cooler fans disconnected by mistake. You didnt mention changing anything else in the case, but just make sure your case fans are facing the right way and that nothing is restricting airflow inside and outside the case. If you had bad airflow before but your cpu didn’t run as hot, could be that you got away with it before and just aren’t anymore.

    • Mistic@lemmy.worldOP
      link
      fedilink
      English
      arrow-up
      1
      ·
      1 year ago

      Did that, it’s in the post. 85C max out of every core under full load. That’s without fans ramping up.

      Stock temp is also normal, 40-45C.

      Plastic cover’s off, thermal paste is sufficient. Fans are properly connected, they are facing the right way (at least I hope they are, w/e, will be finding that out soon anyway)

      I did swap the fan with the CPU. Current one is Thermaright PA 120.

      I am a bit concerned about VRM temps, however. It could be that they give in under extended load. I wonder if undervolting puts an additional strain on those, but why would it if all I do is lessen CPU voltage? (Am new to undervolting, BSOD happens even at stock settings, though)

  • Mistic@lemmy.worldOP
    link
    fedilink
    English
    arrow-up
    1
    ·
    1 year ago

    Since my lemmy crashes when trying to edit the post, I’ll leave an update as a comment.

    For now I’ve limited wattage supplied to CPU, which solved the issue.

    So, basically, there are only 3 options which could be the cause.

    1. Wrong automatic motherboard settings. Which seems to be the most likely option.

    2. PSU protection gets wrongly triggered. That would be unusual, this particular model should be reliable.

    3. Faulty CPU. This is even less likely due to the way they’re manifactured.

  • derpydavy@lemmy.world
    link
    fedilink
    English
    arrow-up
    1
    ·
    edit-2
    1 year ago

    I had a similar experience before. I tried swapping a i3-2120 to a i5-3300 on a cheap Asrock H67 board. I have installed the latest BIOS and support page said it should work but it just kept blue screening.

    It was fixed after I replaced the MB to a slightly less cheap H77 board.

    Friend who helped me fixed the computer said that MB manufacturers don’t really test their MBs with all the CPUs they support. Some niche CPUs may not work, even though they are supposedly supported.