MACHINE_CHECK_EXCEPTION is the CPU raising its hand and saying something is wrong at the hardware level. It sounds terminal, but in practice the cause is usually an overclocked or overheating processor, not a dead one.

The troubleshooting path runs from the easy software-adjacent fixes (kill the overclock, update the BIOS) toward genuine hardware testing. Follow it in order and you’ll know exactly where you stand.

Why this happens

  • Unstable CPU overclock. The top cause — marginal stability exposed under sustained load.
  • XMP/DOCP memory profiles. Technically a memory overclock, and a frequent machine-check trigger.
  • CPU overheating. Thermal errors reported directly by the processor.
  • Outdated BIOS. Old firmware with buggy microcode or power management.
  • Bad RAM. Memory errors surfacing through the CPU’s error reporting.
  • Voltage instability. Failing charger or power delivery causing droops under load.
  • Failing motherboard. Damaged power delivery or bus circuitry — rare, but real.
  • Genuinely failing CPU. The least common cause, diagnosed only by eliminating everything else.
  • BIOS settings corruption. A corrupted CMOS or bad BIOS setting (wrong voltages, disabled cores) producing errors no software change explains — resetting to defaults usually clears it.

How to fix it

Work through these in order, from Fix 1 onward, and stop as soon as the crashes stop.

Fix 1: Return everything to stock speeds

This resolves the majority of MACHINE_CHECK_EXCEPTION cases outright.

  1. Enter the BIOS/UEFI (F2 or Del during boot).
  2. Choose Load Optimized Defaults to clear any overclocking.
  3. Specifically set memory to stock — disable XMP, DOCP, or EXPO profiles.
  4. If you use Intel XTU, Ryzen Master, or ThrottleStop, reset them to defaults or uninstall them temporarily.
  5. Save, exit, and test under the workload that was crashing for several days.

Fix 2: Check CPU cooling and temperatures

A processor protecting itself from heat reports machine checks before it shuts down.

  1. Install HWMonitor or HWiNFO and log CPU temperatures during heavy use.
  2. Sustained temperatures hitting 95 to 100 degrees Celsius mean the CPU is thermal-throttling or worse.
  3. Clean all vents and fans with compressed air — dust is the number one cooling killer in laptops.
  4. Verify the cooling fans actually spin up under load; a dead fan is an emergency, not an inconvenience.
  5. Use the laptop on a hard, flat surface, and consider a cooling pad for sustained heavy workloads.

Fix 3: Update the BIOS and chipset drivers

BIOS updates deliver CPU microcode fixes that directly address machine check errors.

  1. Check your current BIOS version with msinfo32 under BIOS Version/Date.
  2. Visit your laptop maker’s support page and compare against the latest for your exact model.
  3. Read the release notes for microcode, stability, or CPU-related fixes.
  4. Flash with the charger plugged in and all programs closed — never interrupt a flash.
  5. Install the latest chipset drivers from the same page afterward, then retest.

Fix 4: Update low-level drivers

Chipset, storage, and bus drivers interact directly with the CPU’s error reporting.

  1. Download the full chipset driver package from your laptop maker’s site — don’t rely on Windows Update alone.
  2. Update storage controllers in Device Manager.
  3. Run Intel’s driver detection tool for Intel-based systems to catch stragglers.
  4. Restart after each major driver installation.
  5. Test with the crashing workload for a few days before moving on.

Fix 5: Stress-test and test the RAM

Determine whether the CPU or the memory is the weak link.

  1. Press Win + R, type mdsched.exe, and choose Restart now and check for problems.
  2. For conclusive results, run MemTest86 from a USB stick for at least two passes.
  3. If RAM passes, stress the CPU alone with Prime95 (small FFTs) for 30 minutes while watching temperatures.
  4. If Prime95 triggers the crash at stock settings with good cooling, the CPU or motherboard power delivery is suspect.
  5. Our memory diagnostic guide covers interpreting memory test results in detail.

Fix 6: Rule out power problems

Voltage droop under load causes machine checks that look exactly like CPU failure.

  1. Always test with the original manufacturer charger — third-party chargers cause subtle power issues.
  2. Check that the charging port connection is firm, not loose or wobbly.
  3. On desktops, test with a known-good power supply of adequate wattage.
  4. In Settings > System > Power & battery, set Power mode to Best performance while plugged in as a test.
  5. Try a different wall outlet to rule out bad mains power.

Fix 7: Clean boot to eliminate software interference

Some low-level utilities (hardware monitors, fan controllers) can destabilize the CPU.

  1. Press Win + R, type msconfig, and press Enter.
  2. On the Services tab, check Hide all Microsoft services, then Disable all.
  3. Disable all non-Microsoft entries in Task Manager > Startup apps.
  4. Restart and stress-test.
  5. If stable, re-enable in groups — pay special attention to overclocking, monitoring, and RGB utilities.

Fix 8: Reset Windows, then consider hardware service

Eliminate software completely before concluding it’s the silicon.

  1. Back up your files.
  2. Reset via Settings > System > Recovery > Reset PC with Keep my files and Cloud download.
  3. Install only essential drivers — no overclocking or monitoring tools — and stress-test.
  4. If machine checks persist at stock settings on clean Windows with good cooling, the CPU or motherboard is genuinely failing.
  5. Check warranty status and contact the manufacturer — CPU failure on a laptop means board-level service.

Closing

MACHINE_CHECK_EXCEPTION demands respect because it comes from the CPU itself, but the fix is usually unglamorous: stock clocks, clean cooling, current BIOS. Work from the easy answers toward the hard ones and you’ll rarely need to go past Fix 4. For the closely related hardware-error stop code, see the WHEA_UNCORRECTABLE_ERROR guide , and Microsoft’s hardware troubleshooting resources cover the rest.

Frequently asked questions

What does MACHINE_CHECK_EXCEPTION mean?
It’s the CPU itself reporting a hardware error to Windows — the processor detected something wrong in its own operation or in hardware it communicates with, like RAM or the system bus. Unlike most blue screens that blame drivers, this one originates in silicon. That said, the most common trigger is still an unstable overclock rather than a dying processor.
Is my CPU dying if I get this error?
Probably not. Genuine CPU failure is rare — far rarer than unstable overclocks, overheating, bad RAM, or outdated BIOS versions, which all produce the identical error. Only suspect the CPU itself after you’ve eliminated overclocks, heat, memory, and firmware, and the error persists at complete stock settings on a clean Windows install.
Can overclocking cause MACHINE_CHECK_EXCEPTION?
Absolutely — it’s the number one cause. An overclock that seems stable in light use can fail under sustained load when voltage droop or heat pushes it over the edge, and the CPU reports the failure as a machine check. This includes XMP memory profiles, which are technically memory overclocks. Returning everything to stock speeds is always the first fix.
Why does it happen mostly during gaming or heavy tasks?
Because that’s when the CPU runs hardest, hottest, and at its highest clocks — exactly the conditions that expose marginal stability. An overclock or cooling problem that’s invisible while browsing becomes a machine check exception twenty minutes into a game. If your crashes are load-dependent, heat and overclocks are your prime suspects.