My gaming PC has become extremely unstable. Games and web browsers crash, applications sometimes fail to open, and the entire computer frequently blue-screens. The errors include MEMORY_MANAGEMENT (0x1A), KERNEL_MODE_HEAP_CORRUPTION, KERNEL_SECURITY_CHECK_FAILURE (0x139), SYSTEM_SERVICE_EXCEPTION, PAGE_FAULT_IN_NONPAGED_AREA, CRITICAL_PROCESS_DIED, and occasionally WATCHDOG_VIOLATION. MEMORY_MANAGEMENT is the most common.
The system uses a Ryzen 7 5800X3D, an RTX 5070, and Windows 11. CPU and GPU temperatures look normal. So far I have removed and reinstalled graphics drivers, rolled back or removed Windows updates, replaced my roughly ten-year-old power supply, tried different RAM kits and configurations, updated the BIOS, forced PCIe Gen 4, and disabled DOCP. I have also tried resetting or reinstalling Windows, but the computer crashes during the process too. What hardware or troubleshooting step should I investigate next?
4 Answers
The variety of stop codes points more toward unstable hardware or low-level drivers than one ordinary Windows problem. Start with the basics: load BIOS defaults, keep DOCP and any CPU or motherboard auto-overclocking disabled, disconnect unnecessary USB devices, and reseat the graphics card and memory. Test each RAM stick by itself in the recommended slot, then repeat with the other stick and another slot. A bad memory slot, motherboard, CPU memory controller, or unstable voltage can produce memory-related errors even when the RAM itself is good.
A clean Windows installation with completely default BIOS settings is a good dividing line. Back up important files first, because repeated crashes and disk changes can risk data. If the machine crashes while installing a clean copy of Windows, the problem is almost certainly hardware rather than the existing operating system. After installation, use only basic drivers at first and avoid motherboard or CPU utilities that apply automatic overclocks. If it remains stable, add software gradually.
I have tried resetting Windows several times, but the system crashes and restarts during the process. One attempt also produced a WATCHDOG_VIOLATION blue screen.
Check the storage drives with CrystalDiskInfo, including any secondary hard drive. Then run Windows Memory Diagnostic, while keeping DOCP disabled. The built-in test is useful, but a failure does not prove the RAM is the only problem because the motherboard, CPU, voltages, or memory controller can also cause errors. If the drives and memory look healthy, try repairing Windows from an elevated Command Prompt with SFC, then DISM, and run SFC again after restarting: sfc /scannow, DISM /Online /Cleanup-Image /RestoreHealth, then sfc /scannow.
MEMORY_MANAGEMENT appearing most often makes RAM or the memory subsystem the first suspect, even if one memory test passes. Try a thorough bootable memory test if possible, but also test each module and slot separately. If the computer fails with multiple known-good kits and still crashes during a clean installation, the motherboard or the 5800X3D's memory controller becomes more suspicious. Separate VRAM and system-memory tests in a tool such as OCCT can help identify whether the GPU memory or regular RAM is triggering the failures.
I have already tested one stick at a time, both sticks, and another RAM kit that works in a different system. I will continue testing the slots, but I am worried the CPU or GPU could be failing. Temperatures and gaming performance still look normal.