Interactive Troubleshooter

Published: Sep 6, 2026

Interactive: PC Crashes in Games - Temp, PSU, or Driver?

Time: 15 mins Level: Beginner

By StepDiag Engineering Team

Reviewed by Ibraheem Taofeeq Opeyemi

An interactive root-cause diagnostic tree to resolve random game crashes, black screens, and system hard reboots. Systematically isolate thermal limits, power supply transient spikes, and graphics driver faults.

Full Diagnostic Tree & Step-by-Step Overview

What exact physical symptom or system behavior occurs when your computer crashes during a gaming session?

The game freezes instantly or drops to desktop with a graphics driver error message or Unity/Unreal crash log.
The entire computer instantly cuts to black and reboots automatically without displaying a Blue Screen of Death.
The computer displays a Blue Screen of Death (BSOD) containing memory or kernel error codes.
The game stutters severely, frame rates plummet to single digits, and fans ramp to maximum speed before crashing.

What specific error state or behavior accompanies the application crash to desktop?

Windows displays a notification stating 'Application has been blocked from accessing Graphics Hardware' or TDR timeout.
The game crashes only when Ray Tracing, DLSS, or high-preset texture streaming is enabled.
DirectX error code popups appear citing device hung, lost device, or bad shader compilation.

Perform Clean Graphics Driver Reinstallation Using Display Driver Uninstaller (DDU)

Solution:

Root Cause: Corrupted Display Driver Registry Keys and TDR Timeout Trigger

When a graphics driver fails to respond to GPU scheduling commands within a specific time window (Timeout Detection and Recovery - TDR), Windows terminates the driver context. If previous driver updates leave behind stale registry entries or conflicting shader caches, recurring TDR timeouts crash games directly to the desktop.

Step-by-Step Fix:

1Download Display Driver Uninstaller (DDU):

Download the latest DDU utility package from the Wagnardsoft Official Downloads Portal.

2Boot Windows into Safe Mode:

Open Windows Settings (Win + I) > System > Recovery > Advanced startup > Click Restart now.

Navigate to Troubleshoot > Advanced options > Startup Settings > Click Restart, then press 4 or F4 to enter Safe Mode.

3Execute Driver Purge:

Run DDU, select device type GPU, and choose your vendor (NVIDIA, AMD, or Intel).

Click Clean and restart. This completely scrubs all driver files and registry upper/lower filters.

4Install Fresh Driver:

Upon rebooting into normal Windows, install the freshly downloaded stable WHQL graphics driver package.

Disable Overclocking, Downclock GPU Core, and Verify VRAM Stability

Solution:

Root Cause: GPU Factory Overclock Instability or Excessive VRAM Temperature

Many modern aftermarket graphics cards ship with factory overclocks that push silicon voltage limits to the edge. When running intensive ray tracing workloads or aggressive texture streaming, micro-voltage drops trigger calculation errors in CUDA/Stream processors, causing immediate DirectX crashes.

Step-by-Step Fix:

1Reset GPU Clocks Using MSI Afterburner:

Download and open MSI Afterburner.

Set Core Clock (MHz) offset to -50 and Memory Clock (MHz) offset to -100 to test for stability.

Lower the Power Limit (%) slider to 90% to reduce transient current spikes.

2Monitor VRAM and Junction Temperatures:

Configure Afterburner On-Screen Display (OSD) to monitor GPU Temperature, GPU Hot Spot, and VRAM Temperature.

Ensure Hot Spot temperatures do not exceed 105°C and VRAM remains under 95°C.

3Disable Hardware-Accelerated GPU Scheduling (HAGS):

Open Windows Settings > System > Display > Graphics > Default graphics settings.

Toggle Hardware-accelerated GPU scheduling to Off and restart the PC.

Clear DirectX Shader Cache and Repair Corrupted Game Files

Solution:

Root Cause: Corrupted DirectX Pipeline Shader Cache in Disk Storage

Modern games compile complex shader pipelines and store them in local DirectX disk caches. If storage read/write errors or abrupt game closes corrupt these binary cache files, subsequent game launches attempt to parse malformed compiled instructions, resulting in fatal device-hung crashes.

Step-by-Step Fix:

1Clear Windows DirectX Shader Cache via Disk Cleanup:

Press Win + R, type cleanmgr, and select your primary system drive (C:).

Check the box for DirectX Shader Cache and click OK / Delete Files.

2Purge NVIDIA/AMD Driver Shader Cache manually:

For NVIDIA: Delete contents inside C:\Users\<Username>\AppData\Local\NVIDIA\DXCache.

For AMD: Open AMD Software Adrenalin > Gaming > Graphics > Click Reset Shader Cache.

3Verify Game Files via Launcher:

Open Steam, Epic Games Launcher, or Xbox app, right-click the crashing game, select Properties / Manage, and click Verify integrity of game files.

What does Windows Event Viewer log immediately prior to the instant hard system reboot?

Event ID 41 (Kernel-Power) with BugcheckCode set to 0 indicating sudden power interruption or PSU trip.
Event ID 18 (WHEA-Logger) indicating a corrected or uncorrected hardware processor machine check exception.
The system reboots cleanly without any critical pre-crash error entries except for Event ID 6008 (Unexpected shutdown).

Resolve PSU Over-Current Protection (OCP) Trips and Transient Load Spikes

Solution:

Root Cause: Power Supply Unit (PSU) Over-Current Protection (OCP) Trip from GPU Transient Spikes

Modern high-performance GPUs (such as RTX 30/40 or RX 6000/7000 series) draw microsecond-long electrical current spikes that can reach up to 200% to 300% of their rated TDP. If the Power Supply Unit is rated below required wattage or uses single-rail/multi-rail protection thresholds that trigger prematurely, the PSU cuts power instantly to protect components, causing a hard system restart (Event 41).

Step-by-Step Fix:

1Use Dedicated PCIe Power Cables (No Daisy-Chaining):

Ensure your graphics card is powered by separate, independent PCIe cables running directly from the PSU to each connector on the GPU. Never use a single 'piggyback' (daisy-chained) cable for high-draw cards.

2Undervolt the GPU to Reduce Transient Spikes:

Open MSI Afterburner, open the Voltage/Frequency Curve Editor (Ctrl + F), and flatten the voltage curve at a safe maximum voltage point (e.g., locking 925mV at 2650MHz) to limit peak amp draws.

3Verify PSU Capacity and Rails:

If using a multi-rail PSU, ensure heavy power loads are distributed evenly across 12V rails, or upgrade to an ATX 3.0 / PCIe 5.0 compliant power supply designed to handle 2x total power transient spikes.

Diagnose WHEA-Logger CPU Core Voltage Instability and Curve Optimizer Settings

Solution:

Root Cause: Processor Core Voltage Starvation under Low-Load / C-State Transitions

When Windows Event Viewer logs WHEA-Logger Event ID 18 (Machine Check Exception), the CPU has detected internal hardware telemetry failure. This frequently occurs when CPU overclocking or aggressive AMD Ryzen Curve Optimizer / Intel Adaptive Boost settings undervolt individual cores too aggressively, causing crashes during sudden gaming load transitions.

Step-by-Step Fix:

1Revert BIOS Overclocks to Stock Defaults:

Reboot your PC, enter UEFI/BIOS setup (pressing Del or F2), and select Load Optimized Defaults.

2Adjust Curve Optimizer / PBO Settings:

If running AMD Precision Boost Overdrive (PBO) with negative Curve Optimizer offsets, reduce the magnitude of the negative offset (e.g., change from -30 to -15) on unstable cores.

3Increase CPU Core Voltage (vCore):

Add a slight positive offset to CPU load-line calibration (LLC) or vCore voltage in BIOS to prevent voltage sagging during high-frequency execution spikes.

Inspect Motherboard Power Delivery (VRM) Overheating and Power Switch Wiring

Solution:

Root Cause: Motherboard VRM Thermal Throttling or Faulty Chassis Power Switch Cabling

If a system reboots without logging pre-crash hardware errors, the fault often points to physical motherboard power delivery failure. Overheated Voltage Regulator Modules (VRMs) surrounding the CPU socket will trip internal thermal protection circuits and cut system power instantly. Alternatively, a damaged chassis reset/power switch button wire can short-circuit when vibrating during heavy game audio bass.

Step-by-Step Fix:

1Monitor VRM Temperatures:

Use HWMonitor or HWiNFO64 to monitor Motherboard / VRM temperature sensors during gaming.

If VRM temperatures exceed 115°C, improve case airflow or direct a fan toward the CPU socket area.

2Test Chassis Switch Isolation:

Disconnect the front panel Power SW and Reset SW jumper cables from the motherboard header.

Short the power pins momentarily with a screwdriver to boot the system and test if gaming triggers spontaneous reboots. If the issue stops, replace the faulty chassis power button cable.

What specific stop code or module name is referenced in the Blue Screen of Death (BSOD) memory dump?

Stop codes like MEMORY_MANAGEMENT, CRITICAL_STRUCTURE_CORRUPTION, or PAGE_FAULT_IN_NONPAGED_AREA.
Stop codes referencing KMODE_EXCEPTION_NOT_HANDLED or DRIVER_IRQL_NOT_LESS_OR_EQUAL with a driver filename.
Stop codes citing CLOCK_WATCHDOG_TIMEOUT or WHEA_UNCORRECTABLE_ERROR.

Diagnose RAM Instability, XMP/EXPO Profile Faults, and Memory Controller Overclocking

Solution:

Root Cause: Unstable RAM XMP/EXPO Profile or Memory Controller Overclocking

Games load large asset packages directly into system RAM and VRAM at high speeds. If your system memory runs on an unstable XMP (Intel) or EXPO (AMD) overclock profile, or if memory sticks are seated in incorrect motherboard DIMM slots, data corruption occurs inside memory pages, resulting in instant BSOD crashes.

Step-by-Step Fix:

1Verify Physical DIMM Seating:

Power down the PC, unplug power, and reseat your RAM modules into slots A2 and B2 (second and fourth slots from the CPU, standard for dual-channel motherboards).

2Test and Adjust XMP/EXPO Profiles:

Enter UEFI/BIOS and temporarily Disable XMP/EXPO (run RAM at base JEDEC speeds, e.g., 2133MHz/4800MHz). Test if games crash.

If stable at base speed, re-enable XMP but manually step up memory voltage by +0.05V (e.g., from 1.35V to 1.40V) or lower the memory frequency by one step.

3Run Windows Memory Diagnostic:

Press Win + R, type mdsched.exe, select Restart now and check for problems, and let the tool complete a full scan.

Isolate Faulty Third-Party Kernel Drivers Using Driver Verifier and Minidumps

Solution:

Root Cause: Rogue Third-Party Kernel Driver Causing Memory Access Violations

Kernel-mode BSOD crashes triggered by gaming sessions are frequently caused by outdated anti-cheat software drivers (Easy Anti-Cheat, BattlEye, Vanguard), legacy audio drivers, or RGB lighting controllers accessing restricted physical memory addresses improperly.

Step-by-Step Fix:

1Analyze Minidump Files:

Download BlueScreenView or WinDbg from Microsoft Store to inspect the .dmp files located in C:\Windows\Minidump.

Identify the offending driver filename listed in the crash module (e.g., EasyAntiCheat.sys or nvlddmkm.sys).

2Update or Clean Uninstall Anti-Cheat and Peripheral Software:

Completely uninstall game anti-cheat engines, Razer Synapse, Corsair iCUE, or Logitech G Hub, which frequently install faulty kernel filter drivers.

Reinstall the game to force a clean anti-cheat driver re-initialization.

Fix Processor Thermal Throttling, Socket Pressure, and Uncorrectable Hardware Errors

Solution:

Root Cause: CPU Thermal Overload or Excessive Cooler Mounting Pressure

WHEA_UNCORRECTABLE_ERROR and CLOCK_WATCHDOG_TIMEOUT indicate that hardware components failed to report back within designated timing intervals. On modern Intel and AMD processors, this is frequently caused by a CPU cooler mounting bracket tightened unevenly, warping the socket pins, or thermal paste dry-out causing immediate 100°C throttling.

Step-by-Step Fix:

1Check CPU Cooler Mounting Pressure:

Power down, remove the CPU cooler block/heatsink, and inspect mounting screws. Uneven pressure across LGA/AM5 sockets can cause memory channel failure and WHEA crashes.

Re-torque cooler screws in a diagonal star pattern until snug (do not overtighten).

2Replace Thermal Paste and Clear Heatsink Dust:

Clean old thermal paste using 99% isopropyl alcohol and apply a fresh pea-sized drop of high-performance thermal compound.

3Update Motherboard UEFI/BIOS:

Flash your motherboard to the latest BIOS revision to resolve CPU microcode voltage bugs.

What physical or thermal conditions occur inside the PC case when game frame rates plummet?

GPU or CPU temperatures exceed 90°C–100°C, causing extreme thermal throttling and frequency downclocking.
System RAM or VRAM capacity is completely exhausted, causing severe paging to storage drives.
Background thermal accumulation builds up because case fans remain stuck at minimum RPM profiles.

Clean Heatsink Fins, Repaste Thermal Interfaces, and Optimize Fan Curves

Solution:

Root Cause: Severe Thermal Throttling from Dust Obstruction or Dried Thermal Paste

When semiconductor junction temperatures cross critical thermal thresholds (typically 95°C for CPUs and 105°C for GPUs), the hardware automatically invokes thermal throttling—slashing core clock speeds by 50% or more to prevent permanent silicon melting. This manifests as sudden catastrophic stuttering followed by crashes.

Step-by-Step Fix:

1Physically Clean Dust Filters and Radiators:

Use compressed air cans to clear accumulated dust bunnies from CPU cooler fin stacks, GPU radiator fans, and intake mesh filters.

2Repaste GPU and CPU Thermal Paste:

For older graphics cards, replacing dried stock paste and degraded thermal pads on VRAM/VRMs typically drops temperatures by 10°C–15°C.

3Configure Aggressive Custom Fan Curves in BIOS / Software:

Enter UEFI/BIOS and set CPU fan profiles to Performance Mode.

Create a custom fan curve in MSI Afterburner where fan speed scales linearly to 100% once GPU temperature reaches 75°C.

Increase Windows Virtual Memory (Pagefile) and Close Background Memory Hogs

Solution:

Root Cause: Physical RAM Exhaustion and Pagefile Allocation Failure

Modern AAA games require 16GB to 32GB of system memory. If physical RAM is entirely filled while running background web browsers, Discord, and recording software, the OS must swap memory pages to storage. If the Windows pagefile is disabled or set too small, games run out of memory allocations and crash abruptly.

Step-by-Step Fix:

1Configure Automatic Windows Pagefile Management:

Press Win + R, type sysdm.cpl, and hit Enter to open System Properties.

Go to the Advanced tab, click Settings under Performance, and switch to the Advanced tab in Performance Options.

Click Change under Virtual memory.

Check Automatically manage paging file size for all drives and click Set.

2Limit Background Browser Memory Usage:

Disable hardware acceleration in web browsers and close memory-intensive background applications prior to launching demanding titles.

Reconfigure Chassis Intake/Exhaust Airflow Balance and Fan Header Control

Solution:

Root Cause: Negative Airflow Pressure and Thermal Heat Soak Inside Enclosed Cases

If case exhaust fans outpace intake fans, negative pressure pulls dust through crevices and causes stagnant hot air pockets to pool around the motherboard VRMs, M.2 SSDs, and backplate of the GPU. This ambient heat soak prevents coolers from dissipating heat effectively during prolonged gaming.

Step-by-Step Fix:

1Balance Case Fan Configuration:

Ensure you have more intake fans pulling cool air into the case than exhaust fans pushing air out (positive pressure setup).

2Connect Fans to Correct Motherboard Headers:

Connect AIO pump headers to dedicated AIO_PUMP motherboard headers running at 100% DC speed, and chassis fans to CHA_FAN headers mapped to CPU temperature sensors.

3Remove Restrictive Front Glass Panels:

If your case features a solid glass front panel with minimal side intake slits, temporarily test gaming with the front panel removed to check if ambient airflow starvation caused the crash.

Related Diagnostic Guides

Windows Update Error 0x80070002 Missing Files Diagnostic Guide

System Restore Failed 0x80070005 Antivirus Conflict Tree

Windows Photo Viewer Not Showing Thumbnails: Codec Diagnostic

Safe Mode Password Not Working With Microsoft Account Fix

Clipboard History Not Working Windows 11: Sync Service Tree

Subscribe to StepDiag

Embed Guide on Your Website