BSOD - At the end of my rope

[Ace T'Ken]

So first off, some background. I've been building systems since I was 12 and I'm 38 now. I've won overclocking competitions and have built hydraulic flight sim rigs. I owned a storefront for years until I moved away and currently own a managed service provider after being the director of one previously.

That all being said, I am completely stuck here. I feel like I'm going insane and nobody I've spoken to has any clue what the heck is happening here.

A year ago, I built a rig with the following in it:
Intel 9900k
Asus Maximus XI Hero
EVGA 2080 Ti SC
Samsung 970 Pro 1TB m.2
8 TB WD storage drive (simply in AHCI mode, not RAID as it's the only SATA device)
32 GB Corsair LP RAM
Win 10 x64 Home
1000w EVGA platinum PSU

 

Updated the BIOS on the mobo, configured it (stock processor speeds, XMP, disabled onboard audio & wireless, configured fans), used all current drivers (including chipset, LAN, video, etc.) on a clean Win 10 install after disconnecting from the internet and removing all the ones Windows tried to install. Used the Samsung m.2 NVMe driver, and removed all excess startup entries. I even updated the fan controller on the video card. My software is updated obsessively.

 

After a short while, I started getting BSODs. The frequency could be twice a day to every two weeks, but this is massively unusual for any system of mine. Checking the event log, none mentioned anything more than the standard Kernel-Power crash. Sometimes I would see things like IRQ NOT LESS OR EQUAL. Nothing helpful. I tweaked a bit, but after about 4 months and increasing frustration, I changed out the RAM for some new G.Skill RAM.

That didn't help. Over the next 8 months, I've come to replace nearly the entire system.

 

I've had to sfc /scannow (and more) the m.2 drive dozens of times repairing file corruption (ONLY on the m.2, never the 8TB).
I've added in a 1500w battery backup thinking maybe it was something odd with my home power despite nothing else having issues.
I've changed the PSU to a new 850w EVGA Platinum PSU.
I've reformatted 6 times.
I've run without the NVMe driver.
I swapped the processor cooler.

 

Recently, I got fed up after losing work for the umpteenth time. I sold off the motherboard, processor, case, and newest set of RAM at a loss. That person has having no issues for three weeks.

I purchased a Ryzen 3800x and put it on an Asus Strix X570-F motherboard with a brand new case.  A completely new platform! I purchased a new Firecuda 520 2TB m.2 drive and installed my Corsair RAM back in it since it was obvious that was not the problem.

 

I did another fresh install with my new parts, the 8TB HDD, and my 2080 Ti. I updated and configured the BIOS, installed chipset drivers, and...

 

It bluescreened again. I swapped in an old GTX 770 in that I had laying about. Bluescreen. This time I canught the words PFN LIST CORRUPT before it rebooted. I have just now disabled my SATA controller completely, but I don't have access to my storage drive. I don't know if this will solve it, but I can't help but doubt it.

 

Can anyone at all tell me anything? I feel like I'm going insane. What the heck is going on?

 

Additional info:

    I'm installing Windows from USB. I've created fresh keys several times and have verified checksum of my install media.
    My temperatures are within normal range.
    I have a TV attached to the system, 1 wireless mouse, and 1 keyboard.
    It is wired into my router.
    I don't keep more than 3 tabs open at once in my browser and I only use 1 browser.
    I have suspected Firefox as it seemed to happen more often on YouTube while I left videos paused. I installed Chrome and removed Firefox. Same issue.
    It has only crashed once during 3D gaming; it mostly occurs when doing desktop tasks.
    It does seem to happen more when I have a game called Idle Champions of the Forgotten Realms open, but that could be paranoia. It has still happened when the game is closed fully.

[Karlston]

Probably not going to be of much help but it must be some hardware and/or software that is in common between the systems that have experienced the BSOD.

 

From what I've just read, PFB_LIST_CORRUPT seems to be caused by bad drivers, faulty memory or memory corruption, or drive corruptness.

 

Tried booting in Safe Mode?

 

--- Other things to try maybe...

Disconnected the 8TB HD? You've disabled the SATA controller, may have the same effect.

 

Tried a different OS, a Linux for example? Just to see if it's stable or not.

 

Same case? MB standoffs... right number & locations?

[funkyy]

Just a (dumb) question....is the Windows on the USB the same one being clean installed each time?

[Ace T'Ken]

I appreciate any help I can get at this point! I edited my original post to mention that I got a new case as well as I forgot to mention it. No luck there.

 

While the new M.2 drive was present with the old SATA, it also crashed during the most recent Windows install (so while using a bootable pre-OS) and after the crash when I tried again, it wouldn't let me install to the M.2 until I rebooted and disabled the SATA drive. After that, it was fine.

 

No, the USB keys being used to reinstall Windows have actually been 3 or so models that I've always had the latest deploy on using the official MS tool:

https://www.microsoft.com/en-ca/software-download/windows10

 

It hasn't resurfaced since I disabled the SATA, but the crash can be sporadic. Also not helping is the fact that all my data is on the SATA and I need it to do darn near anything. Has anyone ever heard of a M.2 and SATA in UEFI having issues coexisting?

[coromonadalix]

Have you changed your power supply,  had this problem in the past  the psu was at fault ??

 

Ok you did    sorry

[mp68terr]

Might not be a great help, but from what I read you changed/modified your hardware while always keeping your win10 OS. Just wondering if your got (would get) these failures while using another OS?

Re-read the thread, Karlston already proposed to try another OS...

[andy2004]

Windows 10 is the biggest PITA  why i for one will never upgrade sticking to windows 7

 

[Dwa Piva]

Has anyone ever heard of a M.2 and SATA in UEFI having issues coexisting?

 

You are on the right track, i think. Odd combination Win 10 Home and a 8TB data Harddive.

Did a MBR to GPT conversion?  Gave big trouble for me. Had to delete all (hidden) partitions and

finally got it working by making a fresh install from Win 10 x64 Prof.

[Dodel]

Found this article, specific to your nvme

 

"You gotta switch Raid to AHCI in bios, but due to Windows quirks, it won't boot if you simply switch. You need to preinstall AHCI drivers or let windows do it from Safe Mode. As for these steps, took me 5 minutes:

1. First, we need to switch the controller from RAID to AHCI mode. We need to Windows to boot into safe mode after doing this switch, so it will automatically correct its boot parameters. To do this, open your Command Prompt (Admin) and type the following command – bcdedit /set {current} safeboot minimal
2. Restart and enter the BIOS by holding F2 during boot.
3. On the BIOS menu, go to System Configuration - SATA Operation - AHCI - Apply - Exit. Your computer will then enter the safe mode.
4. Once your computer finishes booting into safe mode, open Command Prompt (Admin) once again and type the command below. This will trigger your computer to go back to normal boot mode – bcdedit /deletevalue {current} safeboot
5. Reboot the laptop proceed to login like usual. The controller is now in AHCI mode, running the generic Microsoft driver. If you want to update the driver manually (I couldn't find one), then:
6. 6. Once you finish downloading and extracting the driver, open your Device Manager, then:

• Storage Controllers - Standard NVM Express Controller
• Right-click - Update Driver Software
• "Browse my computer for driver software"
• "Let me pick from a list of device drivers on my computer"
• "Have Disk.." - "(extracted driver zip folder)\x64\ocznvme.inf"
• Select your Driver - Next - Yes (on the warning dialog) - Restart

That is it! You should now be up and running with the controller in AHCI mode and using the driver of your choice.

I couldn't find the actual driver, so I downloaded Samsungs NVMe drivers from their website and their Magic app. Then it appeared in the Device Manager under Storage Controllers, but the actual disk EVO 970 still says it's running the 2006 Microsoft driver. Trying to find out if that's an issue right now."

 

Good luck.

[Avitar]

I had this same problem once.

 

Rewire your house.

 

Your power outlet is giving your systems 105V or less and at lower than 50Hz.

 

My best bet is faulty wiring, even with a good PSU it can't give you more juice than you're getting from the wall.

[spudboy]

Perhaps the obsessive driver installation thing is causing issues. Outside of your GPU and SSD, just keep what drivers Windows installs... unless there is actual reason to update other drivers other than just wanting the latest and "greatest" (which is a lot of times not the greatest).

[Ace T'Ken]

Just to respond to everyone I can in one post...

 

I've booted to a PE USB device and it still crashed. It's a hardware issue, not an OS issue. Windows 10 is extremely stable at this stage, it's my system that isn't.

 

I've reformatted from clean 6 times. That means I have deleted all other partitions on them, so it's not a matter of the drive not being formatted correctly. My m.2 drive is the boot drive, the 8TB is a SATA drive for storage and is in AHCI mode. It is not in RAID mode and never has been.

 

I have an excellent 1500w UPS that I have purchased 3 months ago. It is not my house power.

 

It could be a driver issue, but going from an Intel system to an AMD system with a fresh install in between, the only drivers shared between the two are nVidia drivers (with new cards in between tests and about a dozen revisions, and no GeForce Experience), and Intel NIC drivers. I would be extremely surprised if one of those two caused the issues.

[Karlston]

6 minutes ago, Ace T'Ken said:

It's a hardware issue

 

Looks like it.  It's something you haven't yet swapped.

 

Cables? Power cable, monitor/TV cable, ethernet cable, SATA power/data cable(s).

 

Case fans?

 

Mouse/keyboard?

 

Static?

[Ace T'Ken]

Cables have been mostly the same for the two USB cables I use. They are a backup drive and a cable for my phone, neither of which I leave plugged in unless actively transferring.

 

There is a single display coming out via HDMI to an Onkyo amp that then goes to my display. It's a new cable with a current spec, but the system has crashed while the display was off.

This was the same setup from the system PRIOR to the 9900k system that had no issues.

The power cable has been replaced twice.

 

The single SATA cable I use was swapped when I swapped the board as were the case fans. I've checked my temps and they're pretty solid for my parts.

 

The mouse was changed 4 months ago and has always been wireless. The keyboard has been the same throughout, but is also a wireless Logitech.

 

No static as it's on a hardwood floor is far away from anyone or anything touching it.

[nonspin]

Sicne i have the same board, same CPU .. same nvme's
.. i can narrow i down for you:

- IRQ_NOT_LESS_OR_EQUAL
Is Hardware related (usually Storage)

Are the  970 NVME's routed through the RSTe ?
Yes would be Default .. If you apply very new
RSTe-Drivers, v17.5+ the OPTANE-Mode (on or off)
sometimes goes nuts because of the NVME.
Try downgrading to v15's (no loss in speed).

Also, there are true NVME-Drivers
(stay away from the Magician)

... and de-select "NVME handled by RSTe"
 
 
  .. If you however run into
  CRITICAL_STRUCTURE_CORRUPTION
  -> that's clocking-related (RAM/CPU timing)
  If you overclock (even a little bit) on a
  9700K .. they just happen like every 2 weeks
 
  I get them too on my i9 9900K ... The only
  remedy is to clock-down the RAMs (no XMPs ..
  like WAY down.. 3600 -> 2166)