bsod PC riddled with different BSODs

[Sierrus]

I'll try to make my post short but also give as many details I can think of.

 

My PC specs:

• AM4 Gigabyte B550 AORUS ELITE AX V2

• AMD Ryzen 5900X

• Arctic Liquid Freezer 2 240 A-RGB

• 2x16GB Corsair LPX Vengeance DDR4 3600 C18

• ASUS ROG STRIX 1000G 1000 W

• Sapphire Nitro+ RX 6800XT

• Kingston A2000 M.2 NVME 250GB + Kingston A2000 M.2 NVME 500GB

• Toshiba P300 7200RPM 3TB HDD

• Windows 10 Pro 22H2

The PC was assembled last year around April, and I would say everything was fine except for 1 thing that I can recall. It randomly freezed up after the spinning mouse circle, usually when browsing the Internet but sometimes when opening random programs too. After spinning for a while, it freezes the whole PC taking the Explorer down with it and then coming back to life after about a minute. This occurred every once in a while at first, then more often after some months. Eventually BSODs popped up in some rare occasions until the BSODs became more frequent as well about maybe 3-4 months ago. That's when I started looking up ways to try and resolve the issue, since it's a self-made PC from parts all over the place.

The BSODs I've so far experienced are as follows, from most often occurring to the ones least occurring:

• PAGE_FAULT_IN_NONPAGED_AREA

• IRQL_NOT_LESS_OR_EQUAL

• SYSTEM_THREAD_EXCEPTION_NOT_HANDLED

• SYSTEM_SERVICE_EXCEPTION

• KMODE_EXCEPTION_NOT_HANDLED

• DRIVER_IRQL_NOT_LESS_OR_EQUAL

• CLOCK_WATCHDOG_TIMEOUT

• ATTEMPTED_WRITE_TO_READONLY_MEMORY

• CACHE_MANAGER

• DPC_WATCHDOG_VIOLATION

And these programs were only sometimes listed as the cause of the crash:

• fltmgr.sys

• ntfs.sys

• amdkmdag.sys

• ntkrnlmp.exe

What I've tried so far:

• Uninstall Windows about 3-4 times now, on both the 250GB NVME and most recently the 500GB one, which is where I'm currently at. The BSODs occurred even during some of the Windows installations earlier, so that didn't fix it.

• Removed and swapped around my RAM sticks and test them with Memtest86 and Windows Memory Diagnostic. The BSODs kept occurring and the tests came up with not a single issue with however many passes it took.

• Removed the HDD and NVMEs, BSODs kept occurring.

• Updated BIOS with whatever newest update is available on the MB site, BSODs kept occurring.

• Download every single bit of driver software I could scrounge up on both specific manufacturer and MB sites, BSODs kept occurring.

• I cleared CMOS on the MB and that came the closest to a fix I could notice, since BSODs were barely occurring shortly after I did this, until they started occurring again so this didn't work either.

The times the BSODs occur the most is when Windows is booting, either during the spinning circle on the BIOS Del/F12 logo screen, or the welcome page to Windows, or just shortly after Windows is started. But also sometimes randomly when installing something (most recently that I noticed it was with audio drivers and Gigabyte's RGB Fusion) or just opening any program on the PC. Now as far as Dump files go, the PC only creates 5 files and sometimes it created multiple so older ones get overridden, but I have 2 sets of them from 2 months ago when I first reinstalled Windows on the 250GB NVME and it BSOD a million times during that process (this one is on the 250GB NVME that I currently have removed, let me know if it's needed as well), and the most recent installation of it on the 500GB NVME. I just threw them all in and hopefully they show absolutely anything worthwhile (attached to post).

 

I'm going insane, because even as I still have warranty for all the parts to return for another 6 months or so, I have absolutely no idea what even is the issue to return only that 1 specific part in hopes that resolves it. If anyone has any clue what the problem is and what to do about it, please let me know since the only other last resort I have is taking it to a random repair shop but I have no idea what they'll do and how much worse it'll end up being. Lemme know if anything else is needed.

Minidumps.7z

[Bjoolz]

It looks like memory from the dump files. Memory doesn't have to mean RAM, but it's usually the main suspect. Windows puts low priority data from RAM into the page file and loads it back in when needed so storage can look like memory (And memory can look like storage). The memory controller is in the CPU and if this fails it will just look like memory.

 

When it's storage about half of the dumps will usually blame storage or storage drivers, which I don't see here, so it's likely not storage. You also tested multiple drives. 

 

If anything is overclocked or undervolted, remove it. Including automatic overclocking settings like Precision Boost Overdrive (PBO). 

 

To test the RAM, use the machine normally with one stick at a time. Use the second slot when counting from the CPU socket. If just one of the sticks cause crashes, faulty stick. If it crashes with either stick it's probably the CPU. Memory testers miss faulty RAM fairly often with DDR4 and newer so I much prefer this method. 

[Sierrus]

15 minutes ago, Bjoolz said:

It looks like memory from the dump files. Memory doesn't have to mean RAM, but it's usually the main suspect. Windows puts low priority data from RAM into the page file and loads it back in when needed so storage can look like memory (And memory can look like storage). The memory controller is in the CPU and if this fails it will just look like memory.

 

When it's storage about half of the dumps will usually blame storage or storage drivers, which I don't see here, so it's likely not storage. You also tested multiple drives. 

 

If anything is overclocked or undervolted, remove it. Including automatic overclocking settings like Precision Boost Overdrive (PBO). 

 

To test the RAM, use the machine normally with one stick at a time. Use the second slot when counting from the CPU socket. If just one of the sticks cause crashes, faulty stick. If it crashes with either stick it's probably the CPU. Memory testers miss faulty RAM fairly often with DDR4 and newer so I much prefer this method. 

Heya, thanks for looking into it.

 

I'm also leaning more and more towards it being RAM but it's just all over the place so I can't conclude it. Nothing is OC'd on the PC except the XMP being on from the PC's assembly to the first Windows reinstallation about 2 months ago. Now it's currently off and there's still sporadic BSODs. But the GPU MIGHT be pre-OC'd which I can't confirm since all the sellers of this specific one just had "OC" in its description.

 

I already both swapped the sticks to different DIMM slots and used them separately one at a time in the A2 channel, there were still BSODs. I can check this method out again later.

 

Right now I was trying to do a Sysnative file collection and it BSOD with PAGE_FAULT_IN_NONPAGED_AREA, so after backing up my stuff for worst case scenarios I'll do some more Memtests and see if anything comes up this time. Other than that, what I'd like to is have an additional RAM set to try the system with but I don't have one available so if every error is pointing straight to the RAM I might as well just RMA it and hope for the best they're the actual culprit.