Help Me Solve My Desktop Computer Issue Before Spending Money.

geekcohen on 27/06/2024 - 08:40
Last edited 08/07/2024 - 08:13

Looking to get some assisting from the OzB Community on a computer problem I cannot current resolve. I want to seek thoughts first before I start replacing stuff, buying new stuff and pulling the desktop apart.

Here is a summary:

  • Windows 11 Custom Built Machine
  • Custom Build by me in November 2022
  • i9-13000KF
  • 96GB of RAM
  • RTX 2060 D6 12GB Video Card
  • 3 x 4K 27" Dell Monitors
  • M2 1TB (recently replaced due to SSD Died)
  • 2TB HDD (secondary drive)
  • 750W PSU

Issue? = Many!

  • Randomly reboots, I will be working away and it will just reboot
  • Google Chrome Often gives me a "status_access_violation" error constantly across multiples different sites
  • Chrome will just crash and shut completely and doesn't matter what site(s) I have open or what I am doing
  • I will be working away and then all of sudden all 3 screens go black and then come back
  • Windows 11 Updates won't install and gives me errors or BSOD/GSOD (Green/Blue Screen Of Death) when doing "reboot for updates"
  • Get Random Windows BSOD errors like "PAGE_FAULT_IN_NONPAGE_AREA" < This happens a lot, like this morning it has already happened twice!
  • Recently (not sure if related on not to something else) but the network (via LAN) will just drop. Reboot Windows and it is back. It just drops internet connectivity.

What have I done to try and resolve:

  • Reduce RAM and try to see if it is the RAM sticks
  • Done a Memory Test - No issues
  • Done a GPU Load Test - No issues appear to occur
  • Reseated GPU to the motherboard
  • BIOS reset
  • New Thermal paste on the CPU and reseated it plus the fan
  • Removed Chrome Add-Ons
  • Reinstall Chrome
  • Reinstall Windows 11 (from a fresh image downloaded from MS website)
  • Reinstalled Drivers and always doing updates when they are rolled out, I always like to update as soon as something is available, I don't leave it for ages.

The issues started maybe in the last 6 months and never had an issue before hand. The computer is used for work and some pleasure. It isn't a gaming machine or used for gaming. Heavily used with Chrome primarily and video calls since changing jobs. Was previously used with some Adobe work, but not soo much anymore. Primary Google Chrome with some WinSCP, VS Code and Putty.

Now I am thinking it is either the GPU failing or the PSU is failing or insufficient. I don't have a bigger PSU or same size (only 650w or 550w in some spare machines).

I possibly think the PSU because as the GPU increases requirements then it doesn't have sufficient power so it just reboots.
I possibly think GPU based on some of the errors that are happening.

I don't think it is a heat or cooling issue because I have the side of the machine off and it has 6 fans on the case plus 1 massive one for the CPU. CPU runs at about 50 to 60 degrees. It is only auto mode for fans, so you can hear the fans increase upon additional load at times.

I don't want to go and buy new parts (like new PSU or GPU) for it not to be an issue. So just wanted to get feedback from others as to ideas or if anyone has experience a similar issue. There is so much stuff when you Google with conflicting opinions.

TLDR: Thanks to ChatGPT

User seeks advice on resolving multiple issues with their custom-built Windows 11 desktop (i9, 96GB RAM, GT2060, 3 x 4K monitors, M2 SSD, 2TB HDD, 750W PSU). Problems include random reboots, frequent Chrome crashes and errors, screens going black, Windows update failures, BSOD errors, and intermittent LAN connectivity drops. Despite various troubleshooting efforts (RAM reduction, memory/GPU tests, reseating components, reinstalling software, etc.), issues persist. Suspects GPU or PSU failure but seeks community feedback before replacing parts. The machine is primarily used for work with Chrome and video calls.

Update everyone - 28/06/2024 at 10:50am

I've swapped out the GPU and the same thing happens with Chrome, but it I had no windows crashes.

I am now thinking motherboard as the 3 month old M2 drive became unbootable suddenly yesterday. It is detected in the BIOS and I go to do a Windows Reinstall, it doesn't detect the M2 or another SSD I have plugged in. The BIOS sees both drivers, but not the windows drive and that is a fresh one via the creation tool. Tried multiple USB ports, it boots but sees no drives.

At this point, I might as well get a new machine.

Update 28/06/2024 at 11:50am

Ordered a new motherboard and new RAM today from CentreCom. Will see when it arrives whether it fixes the issue. Got a new M2 Drive coming already.

So it will be new RAM, Motherboard and HDD. Will see how that goes, I don't think it is GPU anymore or PSU. I am thinking Motherboard.

Update 08/07/2024 at 8:15am

So, an update!

After installing new RAM, a new motherboard, a new PSU, and a new CPU, I encountered a combination of issues. Initially, the RAM and motherboard still caused some BSODs, occasional Chrome crashes, and "status_access_violation" errors. Leaving the system on overnight often resulted in reboots or BSODs, a persistent problem even before the new RAM and Motherboard. I adjusted the BIOS to lower power settings to mitigate potential power-related issues and prevent complete reboots due to high power draw from the CPU.

Further research led me to this Microsoft forum thread, where multiple users identified the CPU as the culprit. Following that find, I ordered a new processor (albeit, a 12th Gen i9, 13th Gen was getting pricey).

While waiting, I initiated the warranty process with Intel. However, since Intel required the defective CPU before sending a replacement and I couldn't afford extended downtime, I purchased a new CPU.

The result? Problem solved! The CPU was indeed the issue. It's unclear if high power draw damaged the original CPU, but all tests with Intel support showed no issues, which was frustrating. Nonetheless, replacing the CPU resolved all Chrome issues and random restarts.

On Friday, I received and installed a new 1300W PSU. I kept the BIOS settings unchanged for now and plan to adjust them later to test for further issues. Over the weekend, the system powered everything smoothly without any problems under load.

Ultimately, the CPU was the root cause. The total cost for the new parts was $1,325. Thankfully, the GPU appears to be fine, with no random screen blackouts.

Computing Computer Desktop Computer

Comments

  • arms on 27/06/2024 - 08:58

    Previously I have had a computer with random restarts, albeit after artifacting first. The issue was a GPU VRAM failure. From memory that was hard to diagnose as some of my load testing didn't trigger the restart, but after seeing restarts during gameplay I thought it might be the GPU. The issue was resolved with RMA with the manufacturer, thankfully.

    In your case it sounds like it maybe is a drive or RAM issue if you're getting page fault errors. Have you tried migrating storage drives, or to boot off your HDD?

    • geekcohen on 27/06/2024 - 09:04

      Previously I have had a computer with random restarts, albeit after artifacting first. The issue was a GPU VRAM failure. From memory that was hard to diagnose as some of my load testing didn't trigger the restart, but after seeing restarts during gameplay I thought it might be the GPU. The issue was resolved with RMA with the manufacturer, thankfully.

      I get a feeling that it is the GPU as well.

      In your case it sounds like it maybe is a drive or RAM issue if you're getting page fault errors. Have you tried migrating storage drives, or to boot off your HDD?

      I've recently replaced the drive from an SSD to an M2. The SSD failed and it was still giving me all sort of errors post fix. Others say it is a "Driver Issue" for the page fault errors, and I think that maybe it is the GPU and a driver conflict because of the CPU failing.

  • Jimothy Wongingtons on 27/06/2024 - 09:02
    +3

    1. Have you considered that your pc may be cursed and or haunted? / have you accidentally said beetlejuice 3 times in a row, messed with a Ouija board or upset Indian burial ground in the last 3-10 business days?

    2. You’ve already done everything else I can think of. Randomly reboot, could it be a power supply issue too? I’d prob just buy a new pc at this point.

    • geekcohen on 27/06/2024 - 09:05

      Have you considered that your pc may be cursed and or haunted? / have you accidentally said beetlejuice 3 times in a row, messed with a Ouija board or upset Indian burial ground in the last 3-10 business days?

      Yes, I have considered it.

      You’ve already done everything else I can think of. Randomly reboot, could it be a power supply issue too? I’d prob just buy a new pc at this point.

      Tempted too especially for EOFY. If I was to do that, I would probably buy a laptop. But not sure.

      • Jimothy Wongingtons on 27/06/2024 - 09:06
        +1

        I went from gaming pc to gaming laptop from metabox, so much more convenient imo- I will say though if you are gaming I had an issue with my metabox where the design was insufficient for venting heat from gpu - after some time the keyboard had keys over the heat vents unresponsive lol.

        Still works fine other than that , and tbh I prefer mx keys anyway

        • geekcohen on 27/06/2024 - 09:13
          +1

          I am a big Metabox fan and would most likely move to a Metabox. I have had a similar issue as well with GPU getting hot with my older model.

          I would just use a Laptop as a Desktop really. 1 x Laptop Screen with two external monitors and then plug all my peripherals into it.

      • Minimum chips on 28/06/2024 - 08:01
        +1

        He said it once, then by quoting it you said it again.

        Boy, I hope nobody says Beetlejuice a third time.

    • altomic on 27/06/2024 - 09:27

      Have you considered that your pc may be cursed and or haunted?

      dumbest thing I've read on the internet today.

      not "may be", it definitely is cursed.

      obvious gypsy curse.

  • Screamingeagle on 27/06/2024 - 09:03

    I had a similar issue a while ago where one of the cables from the power supply to the GPU was slightly unplugged (at the power supply end). This led to random bluescreens.

    With that being said, I would have a look at the power supply. An Nvidia 2060 recommends a 550 watt power supply and an i9-13900KF can draw as much as 253 watts in turbo mode.

    Insufficient power can lead to all sorts of issues. Try to put your configuration into a power supply calculator and see what it comes up with. You also have not said what brand it is.

    You can also check the GPU temperatures using MSI Afterburner (it works for most( if not all) brands of cards) and you can check the temperatures and adjust the fan curves if needed.

    • askbargain on 27/06/2024 - 09:06
      +1

      if it's a power draw issue, you won't see a blue screen

    • geekcohen on 27/06/2024 - 09:08

      PSU Calculator says 557 WATTS. But I wonder if the PSU is failing and not delivering full power. The machine is on 24/7, so it rarely gets a break.

      • askbargain on 27/06/2024 - 09:09
        +1

        what brand/tier is the PSU? it's very unlikely to be PSU (coming from a 750W + 3090 ti)

        • geekcohen on 27/06/2024 - 09:11

          Corsair CX750F

  • juzza87 on 27/06/2024 - 09:05
    +1

    Try a different GPU, even a basic one so you can rule out the GPU.

    EDIT: Also what motherboard is it and have you tried a different motherboard?

    • geekcohen on 27/06/2024 - 09:09

      I've got a basic GTX 750 I could try, but that is about it.

      Motherboard is a MSI MAG B660M, haven't tried another Motherboard as I don't have another one that is compatible for an i9.

  • askbargain on 27/06/2024 - 09:08

    go borrow someone else's ram. it's likely ram.

  • skid on 27/06/2024 - 09:08

    Take out all but 2 RAM sticks. Then try underclocking the RAM first and underclock the CPU second.

    • geekcohen on 27/06/2024 - 09:09

      Already tried that, no change.

      • skid on 27/06/2024 - 09:10

        edit: whoa, you guys type fast. Reading the above, if you think it is the GPU, try changing the GPU driver to a much older version.

        • geekcohen on 27/06/2024 - 09:11

          Already tried that too.

  • aussieprepper on 27/06/2024 - 09:08

    Is it DDR5 or DDR4? Have you done a memory test while not in windows? Are you running XMP?

    • geekcohen on 27/06/2024 - 09:10

      DDR4 and yes tried Memory Test, all passed.

  • apple2016 on 27/06/2024 - 09:13

    If the Bios does not log any errors, the only way to find out would be to replace gpu, memory, psu etc and see when it stops. Also check windows event log for any errors etc.

    • geekcohen on 27/06/2024 - 09:15

      The event log refers to NVIDIA a lot.

      The NVIDIA LocalSystem Container Service terminiated unexpectly. It has done this 123 time(s).

      Me is thinking GPU.

      • apple2016 on 27/06/2024 - 10:02

        Get a secondhand gpu and see if the errors continue. This would help you confirm or rule out the GPU.

        • geekcohen on 27/06/2024 - 10:03

          For the cost of a new CPU, may as well buy a new one. If it isn't it, I can use it in another machine.

  • Loot N Plunder on 27/06/2024 - 09:20

    Have you looked at the Windows 11 Crash Log?

    Boot to a live cd (usb) and see if it still hapens

  • Master Bates on 27/06/2024 - 10:22

    some pleasure

    ( ͡~ ͜ʖ ͡°)

    Does your motherboard have an APU? If not can you borrow an external GPU to test with?

    • Grunntt on 27/06/2024 - 14:42
      +2

      Does your motherboard have an APU? If not can you borrow an external GPU to test with?

      Pretty sure the i9-13000KF doesn't have integrated graphics.

  • JustinFR on 27/06/2024 - 10:27
    +5

    Navigate to C:\Windows\Minidump

    See if you have any files there. Most blue screening will create an informative crash log which will point directly to the cause of the issue. Somehow send a link with the files and I'll diagnose it for you if you'd like.

    • geekcohen on 27/06/2024 - 11:07

      Yeah, it does I was trying to read them but not getting much luck. I'll zip them up and PM you a link.

    • pharkurnell on 27/06/2024 - 11:31

      Nice offer. Have a vote.

    • Lord Ra on 27/06/2024 - 13:56

      Keep us posted.

  • jk82 on 27/06/2024 - 12:27

    Have you tested that SSD? If you have another one I'd try replacing it and do a fresh windows install on the new SSD and see how you go

    • geekcohen on 27/06/2024 - 12:28

      M2 1TB (recently replaced due to SSD Died)
      Reinstall Windows 11 (from a fresh image downloaded from MS website)

      • jk82 on 27/06/2024 - 13:18

        Yeah I read that - but have you checked that it's not a faulty one?

        • geekcohen on 27/06/2024 - 14:01

          I don't have any other new hard drives. Only option would be to buy a new one and then see if it works…… then claim the other original one back on warranty.

          Based on this thread and replies, I am replacing almost every part of the computer!

        • geekcohen on 27/06/2024 - 14:24

          It could be faulty, now getting unbootable device issues after multiple restarts and some investigation of bios settings.

          • jk82 on 27/06/2024 - 16:04

            @geekcohen: Maybe try a different M.2 slot? or removing it and installing windows on your HDD (you could just partition a small segment if you don't want to wipe it)

  • solidussnake on 27/06/2024 - 12:59
    +1

    Checked for bent pins under the cpu ??

    • geekcohen on 27/06/2024 - 13:16

      Most processors now don't have pins, as per this

      And it has been fine since build but last 6 months had issues. So if there were anything wrong from a CPU perspective, then I would've thought it would had been from the get go.

      • solidussnake on 27/06/2024 - 16:07

        Yes CPU's don't but I was referring to the Motherboard's pins (I should have said) as reseating could have accidently pushed one slightly out. Worth checking at least ?

  • flagger on 27/06/2024 - 13:19

    There is a known problem with the 13900/14900 where motherboard bios's were upping the power limits them to get the maximum performance out of them: https://www.reddit.com/r/intel/comments/1awpon0/is_your_inte…. Basically they could be drawing up to 400A. Either adjust your bios power settings or update the bios with power fixes for your CPU.

    I should also note, that if the motherboard has been over stressing the CPU, it could have caused permanent damage to it and will continuously crash after pulling back the power settings.

    • ldd-mn on 27/06/2024 - 14:08

      This also applies to the 14700/13700.

    • geekcohen on 27/06/2024 - 14:23

      Oh dear, I wonder if this is my issue……. it would explain the windows crashing, chrome crashing, the thread errors, the works.

      If that is the case, then may as well get a new machine when I am going to probably have to do a CPU and Motherboard replacement.

  • stickingly on 27/06/2024 - 13:25

    Have you tried a different GPU?

    A while back I had an issue with a PC randomly crashing or freezing once or twice a day. After much troubleshooting I tried a different cheap GPU I had lying around and problem went away.
    Had the original GPU replaced under warranty, installed the new GPU and no more issues

    They initially tested it and said they couldn't find any fault, but I persisted and had to explain that it was an intermittent issue and the problem didn't happen with another card and they eventually sent me a new one.

    • geekcohen on 28/06/2024 - 10:49

      Have you tried a different GPU?

      Yes, and issue persists.

  • outlander on 27/06/2024 - 14:22
    +1

    Power supply. It can lead to all sorts of issues.

    The issue is not just the power rating, it's that all the power cables are capable of delivering the right voltage at the amps being drawn. Sometimes you get grime or corrosion on terminals that make a high resistance joint which works some of the time, but tends to crap out as things suddenly rise in power draw.

    Similar thing with SATA leads and such.

    You can usually use something like speedfan to monitor voltages. If you find them getting way out of spec (theres a 5% tolerance on the 12v line, so 11.4v is the bare minimum) then I'd recommend cleaning the terminals with a small rag/qtip and some alcohol

  • Jolakot on 27/06/2024 - 14:44

    Shot in the dark, but do you have an Asus motherboard?

  • joshash on 28/06/2024 - 08:20

    You need to try another motherboard

  • ipv6ready on 28/06/2024 - 09:50

    You can rule out insufficient power RATING if this combo has been working without problem before the issue.

    Lack of wattage will equate to complete shutdown not errors in chrome or BSOD

    Above does not rule out a “bad” PSU.

    Seems you have other PC why not swap out the gpu.

    • geekcohen on 28/06/2024 - 10:49

      You can rule out insufficient power RATING if this combo has been working without problem before the issue.

      Yes, it has been working fine. I haven't had a PSU partially fail.

      Seems you have other PC why not swap out the gpu.

      I've since done that and it doesn't appear to be the issue.

  • skillet on 28/06/2024 - 09:55

    Do a firmware reflash on your mobo MSI MAG B660M.

    It works on many laptop. It's FREE worth a shot.

    Go into BIOS turn off all the auto over clock.

    Inspect your motherboard carefully for damage. Could be also part damage by ESD but you need to be ultra unlucky and did it on purpose(charging up manually).

    • geekcohen on 28/06/2024 - 10:47

      Do a firmware reflash on your mobo MSI MAG B660M.

      Done that and there was an update, but no fix.

      Go into BIOS turn off all the auto over clock.

      Done that and also tried to do a BIOS reset

      Inspect your motherboard carefully for damage. Could be also part damage by ESD but you need to be ultra unlucky and did it on purpose(charging up manually).

      I don't think anything is damaged, it isn't like it has been dropped or anything.

  • Loot N Plunder on 28/06/2024 - 10:14
    +1

    Have you even tried a live linux to eliminate a windows/software/driver problem?

  • geekcohen on 28/06/2024 - 10:51

    Update everyone.

    I've swapped out the GPU and the same thing happens with Chrome, but it I had no windows crashed.

    I am now thinking motherboard as the 3 month old M2 drive became unbootable. It is detected in the BIOS and I go to do a Windows Reinstall and it doesn't detect the M2 or another SSD I have plugged in. The BIOS sees it, but not the windows drive and that is a fresh one via the creation tool. Tried multiple USB ports, it boots but sees no drives.

  • afah0447 on 28/06/2024 - 13:35

    Have you tried turning it off and then turning it on again?

    • geekcohen on 28/06/2024 - 13:45

      Honestly……

  • ChickenLips on 28/06/2024 - 15:09
    +1

    Hey mate. Given the multiple and weird crashes and faults I would be highly suspecting PSU issues, not total power but something wrong with the power delivery. I see you’ve tested RAM which would have been my first guess. More rarely it will be the mobo and I would keep that as a secondary option as the PSU is much easier to test.

    Edit: if you are testing modular, make sure you use the cables of the associated PSU

  • djathome on 28/06/2024 - 16:21
    +1

    Just chuckin in a few cents here…
    I would say from all thats been said, psu could be the problem.
    Lootnplunder had the same thought I did…
    I would try , and do all the time for problem machines like this, is to rufus up lmde on a usb stick, jam that in and run it for a while off the usb stick.
    Jump into a terminal and sudo dmesg.
    This will give you oodles (too much) info…
    Ignore it.
    Run chrome or whatever for a bit… then run sudo dmesg again (use the up arrow in the same terminal and hit enter). Anything funky going on will be listed… Usb ports shutting down, pciex lanes changing speeds, hard drive errors etc. It will look like gobbledy gook. Dont stress. It will just help you diagnose.
    But you might also find the machine runs ok as well….
    if you need help with this, please reach out.

  • ReaperX22 on 28/06/2024 - 23:41

    Sounds more PSU to me.. But could be the board.

    CX750 is a pretty budget psu. Given its Corsair it should still be fine… But given your specs I would've expected a more premium PSU. Curious if your new board and ram fix the problem.

    • geekcohen on 01/07/2024 - 13:05

      I am thinking Motherboard, but I might eventually upgrade the PSU.

  • Succulent on 29/06/2024 - 01:03

    Why do you have such a powerful spec and 4k monitors mostly for videocall, chrome and coding related work?

    Its a bit of overkill, isnt it?

    • geekcohen on 01/07/2024 - 13:00

      I don't think it is overall, not when you could be running 40 to 50 Chrome Tabs, SSH windows, WinSCP, VS Code, Slack just to name a few things plus in other background apps like sync.com for files.

      My 16GB i5 Lenovo laptop is struggling with about 10 chrome tabs and a Google meet call.

      When I built the machine, I was more into development and running a lot of Adobe products, however, I've since moved role and don't have as much hands on as what I used to (especially with Adobe stuff), so it might appear overkill, but I'd rather have a great machine than one that just ticks the boxes.

      • Succulent on 01/07/2024 - 20:32

        Yeah adobe products made more sense for that build.

        What would be the use case in your role for opening 40-50 chrome tabs?

  • oneone001 on 29/06/2024 - 02:11

    have you tried unplugging everything and plugging it again.

    dusting and vacuuming.

    maybe change your powerboard and not have everything running on one wire :

    try changing usb ports and see if using just mouse and keyboard helps

    wisdom from one who knows jackshit about IT but sometimes does the job.
    '

    • geekcohen on 01/07/2024 - 13:01

      Thanks for the tips, but overall I have tried all of that.

      maybe change your powerboard and not have everything running on one wire.

      I've got my Desktop going into a UPS and then the UPS into mains, so it isn't powerboard related.

  • sntc on 29/06/2024 - 07:01
    +1

    I did some through reading/calculation and you are close to max around 600watt usage

    you are likely hitting psu limit now and then, which may have started to affect other hardwares

    the psu you have is 80% efficient, meaning only 600w is available out of 750w, and it will degrade over time

    the newer cumulative windows update/games will most of the time demand more power than the previous builds, and sometime messes up with your drivers/software likely the cause for your chrome issues
    - try google Chrome enterprise which is much more stable
    - PAGE_FAULT_IN_NONPAGE_AREA - there is a rogue application/driver causing this issue - check reliability monitor to see if it can easily identity the culprit of not event viewer is your code, you can also look into enabling blue screen dump files

    though the CPU is 253w limit when not overclocked, it can have brief instantaneous spike (could not find by how much) less than 10ms as per intel website, likely the cause for your screen going off (check all power cables to computer/monitors are thoroughly plugged in) or random reboots

    random reboots and blue screen are usually driver related and sometime starting of hardware failure

    things I would try in your case with what you got

    1. use the extra psu you have with the primary psu, to distribute the load - you can put it outside the computer, yes you can use two psu at the same time

    - ideally get atleast a single 1000w psu with atleast 80% efficiency which gives you around 800w - and room to account for any fluctuation, overclocking, loss of efficiency / upgrades or additional hardwares overtime

    1. in the advanced power options in the old control panel, you can decrease maximum CPU frequency to 50% when plugged in to throttle directly and bring the stress on the cpu/power usage down and see what issues you stop getting, if it improves, up the numbers to test your threshold

    2. try Windows 10 instead of Windows 11 if not already, the same product code will work, make sure to update Windows first completely before installing any drivers or software, install one app at a time which you need, go with minimum apps you need and test, install others next day or later if no issues

    3. try Linux? - make a live Ubuntu usb and boot from it to see if you still have the same issue

    when googling look for "solved" articles

    best of luck

    • ReaperX22 on 01/07/2024 - 13:55

      Just wanted to add…

      That's not how the efficiency rating works. 80% is suggesting that it can convert power from the wall at 80% efficiency. It's a power draw figure really.

      A 750w rated psu will still output 750w (assuming it's decent). It just means at say 500w power draw from the system you're pulling say 600w from the wall.

      A more important rating is say the 12v rail and how much a GPU can pull from the psu. But I don't necessarily think the CX750 they have can't do it, but it's possible because it's lower end that it still may struggle to provide the power required for specific components, period.

    • geekcohen on 01/07/2024 - 17:08
      +1

      the psu you have is 80% efficient, meaning only 600w is available out of 750w, and it will degrade over time

      I think this is the probably, degrading over time. It started off good and then has gone downhill. Then as the CPU spikes, bang she powers down

      I got a new Motherboard, RAM and M2 SSD today and it kinda resolved the issue. I did have the adjust the power settings in the motherboard because I think the CPU is pulling too much as per this post.

      I've just ordered a Gigabyte UD1300GM PG5 1300W 80+ Gold Fully Modular ATX 3.0 Power Supply which should arrive in a few days and then I should be right after that.

      Fingers crossed that will do what I need it too.

    • geekcohen on 01/07/2024 - 13:02

      Thanks? Not sure if what you said was at all helpful.

  • kaleidoscope on 29/06/2024 - 07:49

    750W PSU

    For a 2060, it is an overkill but what brand/model of psu is it? It seems most likely a power draw issue or gpu vram. Since you’re investing so much time & money into troubleshooting might as well give windows 10 a go on a separate partition and see. I’ve heard of so much stability issues on win11.

    On the psu front, I’ve used 2060 with sff 600 from corsair and currently the same psu powers
    4070 with no issues.

  • geekcohen on 08/07/2024 - 08:12
    +4

    So, an update!

    After installing new RAM, a new motherboard, a new PSU, and a new CPU, I encountered a combination of issues. Initially, the RAM and motherboard still caused some BSODs, occasional Chrome crashes, and "status_access_violation" errors. Leaving the system on overnight often resulted in reboots or BSODs, a persistent problem even before the new RAM and Motherboard. I adjusted the BIOS to lower power settings to mitigate potential power-related issues and prevent complete reboots due to high power draw from the CPU.

    Further research led me to this Microsoft forum thread, where multiple users identified the CPU as the culprit. Following that find, I ordered a new processor (albeit, a 12th Gen i9, 13th Gen was getting pricey).

    While waiting, I initiated the warranty process with Intel. However, since Intel required the defective CPU before sending a replacement and I couldn't afford extended downtime, I purchased a new CPU.

    The result? Problem solved! The CPU was indeed the issue. It's unclear if high power draw damaged the original CPU, but all tests with Intel support showed no issues, which was frustrating. Nonetheless, replacing the CPU resolved all Chrome issues and random restarts.

    On Friday, I received and installed a new 1300W PSU. I kept the BIOS settings unchanged for now and plan to adjust them later to test for further issues. Over the weekend, the system powered everything smoothly without any problems under load.

    Ultimately, the CPU was the root cause. The total cost for the new parts was $1,325. Thankfully, the GPU appears to be fine, with no random screen blackouts.

    • Grunntt on 09/07/2024 - 10:52
      +1

      Thank you for returning with such a detailed follow up.

      It is becoming less common on here to find out any resolution - people seem to be keen to ask for advice but impolite enough to not let anyone who helped know what resulted. (The ones that really annoy me are when they mark it resolved and delete their post details.)

      Glad you finally managed to get it sorted. Unfortunate that it sounds like a rather expensive fix.

      • geekcohen on 09/07/2024 - 15:18

        Not a problem at all! I agree with all that you have said.

        Whilst it is unfortunate how much it cost me, the headaches have stopped and it works extremely well now. So now the battle is with Intel about the faulty CPU. I was trying the cheap route first before going big on a new CPU….. anyway! Life goes and I am much happier with the performance of the machine.

  • geekcohen on 17/07/2024 - 15:44

    My RMA got approved by Intel! Now to build another PC and and use it somewhere, I won't put the new replacement in my current machine, its running fine on the 12th Gen. I've got some spare hardware, just need a compatible fan and maybe a cheapish graphics card.

    • flagger on 17/07/2024 - 18:42

      Nice! I would just sell it off and put the money towards a new system in the future.

      • geekcohen on 29/07/2024 - 12:31

        Not a bad idea, I might just do that. Practically got a new system anyway, besides the GPU. I've replaced everything else but the case and GPU.

        It'll help pay for the cost of all the other hardware anyway.

  • geekcohen on 29/07/2024 - 12:31
    +1

    Intel Hints 13th, 14th Gen Desktop Chip Bug Results in Permanent Damage saw this over the weekend. Exactly what has happened to me.

    Sending my CPU back for a replacement. As suspected, they want the original CPU before sending a new one.

Login or Join to leave a comment
Login Join