The BSOD triggered by ntoskrnl.exe indicates a problem with the Windows kernel.
The BSOD triggered by ntoskrnl.exe indicates a problem with the Windows kernel.
Hi there. I'm just starting this forum and will stay straightforward.
I've been experiencing several BSODs for a while, starting at least in June. In June, I thought one of the 8GB RAM sticks might be faulty because it made clicking noises when pressed, so I swapped it out and reset my PC twice, but the issue persisted. I wasn't sure if the RAM was the cause or if there was a driver problem.
In July, while playing games on Steam (both VNs and FPS titles), I encountered a blue screen after about three hours. Nowadays, it randomly goes blue screen when I idle the PC. Last Sunday, I ran MemTest86 on both RAMs (though the test stopped early because I didn’t have time) and found no issues. After re-seating the RAM, the blue screens happened much more quickly when I left the PC idle. I turned off sleep mode and fast startup, updated my BIOS (Maxsun A520M), and refreshed my graphics driver (NVIDIA GTX 1660 Ti).
I suspect it might be overheating, but I don’t know why. Could it be related to the PC case? The case looks fine, and my hardware doesn’t seem dirty, though I’ll clean it next week if it’s really the problem.
My computer was a gift in 2023.
The original RAM stick led to BSODs because it was incompatible with the system, causing instability before the replacement.
Execute the ram at 3200 not 3600 faster, as 3200 isn't assured and may cause instability. Also, it might indicate a corrupt Windows setup.
That ntoskrnl.exe file is linked to a crucial missing file in Windows, which is commonly seen. The best solution is a complete reinstall; I recommend backing up your games and saves to a portable drive. Perform a full system reinstall of Windows. Additionally, keeping the RAM speed at 3600 can cause instability. The CPU officially supports up to 3200 MHz, and exceeding that often leads to inconsistent performance.
I examined multiple memory dumps. I would refresh the BIOS to the beta version in order to obtain the latest system fan configuration. (assuming the CPU is overheating)
amd chipset driver for a520 (updated)
(initial attempt)
https://www.amd.com/en/support/downloads.../a520.html
(my estimate is an overheated CPU or power issue, possibly due to outdated CPU drivers, CPU chipset, or old software)
I suspect the overheating might be caused by a CPU temperature problem or insufficient power delivery. It could also relate to outdated drivers or firmware.
Are you using rtcore64 to monitor GPU temperatures? (not intending to overclock the GPU)
My motherboard manufacturer hasn’t released updated CPU chipset drivers since 2023.
I might think about disabling the AMD Power Management driver. (blocking C states in BIOS or turning off AMD power management)
(You may want to try installing the AMD Ryzen Master driver.)
I would also look into downloading the AMD CPU chipset drivers directly from AMD.
(I could check the BIOS for power supply voltages to confirm they meet specifications)
Clean dust from CPU, power, and GPU fans.
This situation isn't completely correct. The ntoskrnl.exe file is actually the core of the Windows kernel, and that's not the source of the problem. Usually, the issue arises from a third-party driver that malfunctions, but the error only becomes apparent once control returns to the kernel. At that point, the kernel detects the driver's failure, yet it lacks information about why the driver failed, nor can it communicate with the driver to resolve the issue. Moreover, because the kernel doesn't understand whether the driver's malfunction will harm or erase user data, it must stop the system entirely through a BSOD. The module responsible at the time of the crash was indeed ntoskrnl.exe, which explains why the failing module appears in the dump. However, this isn't the actual culprit—we must trace back the dump to identify the problematic driver.
By the way, occasionally we encounter ntkrnlmp.exe instead, which is the Windows multi-processing kernel and is also not at fault.
It's strange since I've already reset Windows twice completely, but the issue keeps happening. I'll consider adjusting the RAM voltage settings in the BIOS next.