Diagnosing blue screens (BSOD)

Stop codes are clues, not verdicts. Read them properly, test memory and drivers methodically, and separate hardware from software causes.

1. 1

   Capture the stop code

   Photograph the blue screen's stop code (e.g., MEMORY_MANAGEMENT, WHEA_UNCORRECTABLE_ERROR, IRQL_NOT_LESS_OR_EQUAL). After reboot, Reliability Monitor (search 'reliability') timelines every crash with details.

2. 2

   Translate common codes

   MEMORY_MANAGEMENT / PAGE_FAULT → test RAM first. WHEA_UNCORRECTABLE → hardware, often CPU/power/overclock. IRQL / SYSTEM_SERVICE_EXCEPTION → usually drivers. VIDEO_TDR → GPU driver. DPC_WATCHDOG → drivers or storage firmware.

3. 3

   Undo overclocks and XMP

   First test: set BIOS to defaults (including disabling XMP/EXPO). Unstable memory profiles are a top BSOD cause even on 'working' systems. If crashes stop, re-enable XMP and test again to confirm.

   Tip: XMP instability often appears months later as sticks age or after a BIOS update changes training.

4. 4

   Test RAM properly

   Run MemTest86 from USB overnight (4+ passes), or Windows Memory Diagnostic as a quick check. ANY error = bad stick or unstable profile. Test sticks individually to isolate.

5. 5

   Attack the driver layer

   Update GPU + chipset drivers from vendor sites. Check Device Manager for warnings. For repeat offenders, the crash dump names the module: WinDbg's '!analyze -v' on C:\Windows\Minidump files, or NirSoft BlueScreenView for a quick look.

6. 6

   Hardware escalation

   Crashes only under load → run OCCT PSU test (transient power). Crashes at idle → RAM/board. Check drive health with CrystalDiskInfo. If WHEA errors persist at stock settings on good RAM, suspect CPU degradation or VRM/PSU issues.