Apparently all major games are experiencing crashes—seeking assistance with dump files.
Apparently all major games are experiencing crashes—seeking assistance with dump files.
I recently shared a concern about my RAM problems; clearly, my system was definitely faulty, as confirmed by memtest86. This issue appeared to be the second one to fail. I expected it to resolve after the fix, but it hasn’t. After running numerous RAM tests with memtest86—including an extended overnight session for about 12 hours—I reinstalled Windows 11 24H2 twice, updated to Windows 11 23H2 (my current release), and performed stress tests via 3D Mark and OCCT for CPU, RAM, GPU, VRAM, etc. Still, no defective component was found, nor did my problems vanish with a fresh Windows install. As someone who’s used dogfood builds of Windows Insider, I suspect the problem isn’t hardware-related. I’ve tried everything from page file adjustments to using Samsung Magician and Solidigm tools, but nothing resolved it. The page file issue was the main hurdle, which didn’t seem to affect other areas either.
I noticed no issues with NVME drives, as reported by Samsung Magician and Solidigm. The only recurring trouble came during gaming sessions. Initially, I thought it was tied to Easy Anti-Cheat in games, but recent Windows 11 updates and certain titles like For Honor, Far Cry 6, and Vicious Assault Llama Attack kept causing crashes. This prompted me to revert to Windows 23H2 using an existing image on my Ventoy drive—possibly to rule out that drive as the culprit.
The games I’ve struggled with most are The Finals, Fortnite, Sea of Thieves, and even titles like For Honor and Far Cry 6, which would crash or freeze. Every time a game triggered a system crash, it started with a BSOD followed by a reboot. After disabling automatic reboots on startup and recovery, the crashes persisted, even though I set the system to mini-dump mode. This suggested a deep-level issue with the page file.
I attempted various workarounds: moving the dump location to Solidigm, using a 58GB Optane drive with an Intel P1600X SSD, and even setting up a minimal page file on the C drive. Still, problems persisted. Eventually, I resorted to using my 58GB Optane drive with an Intel P1600X, manually configuring it for paging file support and directing dumps to that drive. The issue remains unresolved.
I have all the dump files ready, so if anyone here is willing, I’d appreciate any insights or help. If not, let me know where to send them—perhaps techsupport or Tom’s Hardware. I’ve read that missing a paging file can cause trouble, but even adding a tiny page file didn’t fix it. I’m also considering Malwarebytes again, though it seems the standalone version no longer works.
My system is now just showing a black screen instead of BSODs, though I did receive a valid dump file from this incident. Windows keeps reformatting the drive during reinstalls, which makes me think my storage devices (Samsung 990 Pro, SATA SSDs, HDDs) might be the culprit—or maybe there’s undetected malware interfering with Windows Defender. I’m at a loss and feel like I’m missing something critical. If you have experience with similar setups, any advice would be greatly appreciated.
Remove the hard drive with a warning. Even without use, a faulty HDD can disrupt Windows and lead to crashes just by being connected. If you've fully formatted and reinstalled Windows on clean drives, this may point to a hardware problem. The BSODs likely indicate a failing storage device or corrupted RAM. In some instances, a defective motherboard board could also be responsible. Since you've replaced the RAM already, I recommend checking the storage drives next. The one I recommended should be disconnected from the system and tested for crashes first. If they stop, the issue is likely with that drive; if not, the problem may lie with your motherboard.
I had a hunch about that, which is why I added the details—though I won’t use it. All the SATA drives are connected via AMD's Software RAID for some reason, so I’ll attempt to disable them first in software/BIOS, beginning with the one you referenced. This machine is quite large, so undoing things might need to wait physically, but it’s still a solid plan.
From the dump files it looks like a storage issue. One was a Kernel_Data_Inpage_Error which means that the page file corrupted (And the suberror shows "disk sub-system failure") and it's a big red flag for faulty storage. You had one Critical_Process_Died which isn't a strong indicator of storage by itself, but it's frequently seen with storage issues because it means that a Windows process crashed. And the three remaining crashes were because of something in the registry being corrupted. I know you have the SATA HDD with a caution, but the page file is almost never put on an HDD when you have SSDs available. It can be on any drive, or even multiple drives, but it's usually on the drive that has C:\. With the registry errors and the Critical_Process_Died, the OS drive would be the main suspect (Assuming it has the page file as well). I'm assuming that it the OS is on the 990 Pro. I know you checked it with CrystalDiskInfo, but SMART is completely useless with NVMe SSDs. If you look at the parameters in the bottom half of CDI, you'll probably notice that SATA drives have way more of them. With NVMe, they removed almost all of the useful ones. The only remaining useful ones are Media Integrity Error and Critical Warning. And in the hundreds of faulty NVMe drives I've seen, less than five have showed any counts in these fields. It's just not reliable. And if you are wondering what tool to use instead, we don't really have one. So if you still crash after removing the HDD with a caution, check which drive has the page file. This guide shows how to move the page file (Which you can do if it's on multiple drives or if it's not on the NVMe) and it's also useful for just checking where the page file is.
Just to be sure, everything should return to the CPU. That’s a good check. Make sure the pressure on your CPU cooling block isn’t excessive.
Appreciate the feedback that aligns with the BSOD patterns and Event Viewer results. It seems storage behavior was inconsistent, especially with both 990 Pro and Solidigm drives involved. I’ve already shifted the page file to the Intel Optane drive, which helped (explaining why I now have dumps). I’m curious if removing hidden pagefiles or adjusting settings on the other two drives could resolve the problem. Also, reviewing the manual would be useful to understand which NVMe and PCIe slots are chipset versus CPU-based—whether everything connected directly to the CPU works or if a different setup is needed. If everything is properly connected to the CPU, it might point to a chipset or motherboard issue.
I’m considering three drives: the old 990 Pro 2TB (used as boot drive), the new 990 Pro 4TB (new boot drive), and Solidigm 2TB (game drive). It’s possible I have three faulty drives, but the 4TB is brand new, so it should be possible to return it via RMA with Amazon or Samsung. The other two might still be under warranty, while the Asus motherboard isn’t covered. AM4 chipsets are also quite expensive, so I’m hoping the problem lies with the drives themselves.
Regarding the BSODs, they could stem from any running application on a specific drive—Windows itself, just the page file, or something else. If it’s only Windows, installing it on the Optane drive might help (it requires at least 64GB, and the current folder is close to that). If an app is causing it, testing other NVMe drives and their slots would be wise.
Additionally, I noticed swapfile.sys remains on the C drive, likely tied to UWP games or anti-cheat software. This could explain why some games work while others don’t. A guide on symbolically linking files to another drive might help temporarily, but it’s probably just a temporary fix for what looks like a hardware problem.
Connect one drive at a time. Problems with multiple drives might point to the motherboard or M.2 slot. Simply being connected can trigger crashes. For this set, Windows/page file issues are likely the main cause. Since you’ve moved it, page file errors are my concern. It raises questions about whether the problem started with storage or another factor. Not receiving dump files is usually a sign of storage trouble, but other causes can produce the same effect. Encrypted drives like BitLocker prevent writing dumps. Storage is nearly full—initial dumps can be as big as your RAM plus 3-4GB. You’ll need more than that available. If you changed the page file size on the old drive, it should match the size mentioned earlier.