Windows 11 freezing randomly — where the cursor stops moving, keyboard input stops, and the screen either stays frozen or goes black, requiring a hard restart — is a different problem from Windows freezing during specific activities. Random freezes have random-looking patterns precisely because they’re caused by hardware-level issues: failing RAM, temperature-induced throttling, or storage errors that can occur during any operation. For a broader walkthrough, our Complete Guide to Fixing Windows, Browser, and Software Errors is a good next read.
The strategic question before any fix: when does it freeze? The timing is the most useful diagnostic clue available.
- Freezes within minutes of startup, regardless of what you’re doing — RAM instability or temperature issue
- Freezes only during graphics-heavy tasks (games, video, 3D) — GPU temperature or GPU driver
- Freezes after the machine has been running for hours — Thermal throttling as components heat up over time
- Freezes that produce a blue screen before the restart — Note the stop code. That code identifies the component. See our BSOD guide.
- Freezes specifically after Windows updates — Driver regression
If freeze timing doesn’t match any of these, check temperatures first — thermal causes are underdiagnosed compared to how common they actually are.
Fix 1: Check Temperatures Immediately
Download HWiNFO64 (free, no install required) and run it during a task that triggers the freeze — or just during normal use while watching the sensor panel. The values to watch:
- CPU Temperature: Should stay under 90°C under load. Consistently over 95°C = thermal throttling or shutdown
- GPU Temperature: Under 90°C under load. Above 95°C = GPU thermal protection activating
- NVMe SSD Temperature: Under 70°C. Above this and the drive throttles its throughput dramatically
If temperatures are consistently high: compressed air through the laptop vents (or desktop heatsink fans) takes 5 minutes and often resolves the issue entirely. Dust accumulation reducing airflow is the most common hardware cause of overheating on machines that have been in use for a year or more. After cleaning, the same tasks complete without freezing because the hardware is maintaining thermal limits.
Fix 2: Test the RAM
Random freezes with no consistent trigger — freezes while typing in a document, while loading a page, with no GPU activity — point at RAM errors. Memory faults occur at any memory access, producing truly random-looking freeze patterns.
Test with Windows Memory Diagnostic: search in Start → “Windows Memory Diagnostic” → Restart now and check for problems. The test runs on next startup and results appear in Event Viewer (Windows Logs → System → MemoryDiagnostics-Results) after Windows boots back in.
For a more thorough test: MemTest86 from a USB drive. Run at least two full passes. Any errors at all confirm RAM problems. If errors appear: first try reseating the RAM sticks (remove, clean contacts, firmly reseat). If errors persist after reseating, the RAM needs replacement.
Also check whether XMP or EXPO is enabled in BIOS — these profiles overclock the RAM to its rated speed, which occasionally causes instability on specific hardware. Disabling XMP and running at default speeds is a useful test even when RAM diagnostics show no errors.
Fix 3: Driver Update and Rollback
A recently updated driver — particularly GPU, audio, or network adapters — running at kernel level can cause system hangs when it encounters a condition it doesn’t handle correctly. The freeze is the system waiting indefinitely for a driver operation that never completes.
Check Reliability Monitor (search “reliability history” in Start) — it plots software changes and system failures on a calendar, making it easy to see whether freezes correlate with a driver installation. If they do: Device Manager → right-click the relevant device → Properties → Driver → Roll Back Driver. If the timeline shows freezes preceded by a Windows Update: Settings → Windows Update → Update history → Uninstall updates → remove the most recent update → test stability.
Fix 4: Check Storage Health
A failing storage drive produces freezes when Windows tries to read from a sector that can’t be accessed — the operation hangs for 30–120 seconds waiting for a read that never completes, then either errors out or requires a hard restart. These storage-caused freezes often have a pattern: they occur when specific applications launch (loading from a problematic sector), or when Windows performs maintenance tasks.
CrystalDiskInfo shows S.M.A.R.T. health data: look for non-zero values in “Reallocated Sectors Count,” “Pending Sectors,” or “Uncorrectable Sector Count.” Any of these indicate physical drive problems. A drive showing these values needs data backup immediately — don’t wait for it to fail completely before securing your data.
Also run CHKDSK: administrator Command Prompt → chkdsk C: /f /r → confirm to run on next restart → restart. CHKDSK scans the filesystem and marks bad sectors to prevent future reads from them. This doesn’t fix a failing drive, but it reduces freeze frequency by steering Windows away from known-bad sectors.
Fix 5: SFC and DISM
Corrupted Windows system files can cause hangs when Windows tries to execute a damaged component. SFC and DISM repair the damage:
DISM /Online /Cleanup-Image /RestoreHealthThen:
sfc /scannowAdministrator Command Prompt, restart after both complete. SFC reporting “found and repaired corrupt files” means system file corruption was contributing to the freezes.
Fix 6: Disable Fast Startup
Windows Fast Startup saves the kernel session during shutdown and resumes from it on the next boot. When the saved state becomes inconsistent with the current hardware configuration (after driver updates, hardware changes, or certain Windows updates), Fast Startup resumes from an invalid state — and freezes can result from the inconsistent state being used as the starting point for the session.
Control Panel → Power Options → Choose what the power buttons do → uncheck “Turn on fast startup.” Shutdowns now perform a complete hardware reinitialisation on the next boot. Freezes caused by stale Fast Startup state typically stop immediately after this change, and boot time increases slightly — usually by 10–30 seconds.
Overclocking Instability
If the CPU, GPU, or RAM is overclocked: instability from the overclock produces exactly the “random” freeze pattern described here. An overclock that appears stable under typical loads can fail under specific computational patterns, producing freezes that seem unrelated to any particular activity.
Reset everything to stock in BIOS (Load Optimised Defaults) and run the machine for a day. If freezes stop, the overclock was unstable. Reduce the overclock incrementally to find the actual stable ceiling, or leave it at stock if the performance difference isn’t critical for the current use case.
Our guide on Windows 11 keeps freezing covers the freeze scenarios with more detail on the minidump analysis approach for identifying the specific driver or hardware component responsible. For thermal issues specifically on laptops where cleaning isn’t sufficient, our Windows 11 performance guide covers the power plan and throttling settings that reduce heat generation at the cost of some performance. Microsoft’s Windows 11 stability documentation covers the Reliability Monitor interpretation and the Windows Performance Analyzer for detailed CPU sampling that identifies what the system was executing at the moment of a freeze.
Third-Party Software Causing System Hangs
Applications that install kernel-mode drivers — antivirus software, virtual machine software, VPN clients, certain system utilities — run code at the highest privilege level in Windows. A bug in a kernel-mode driver produces a system hang rather than an application crash, because there’s nothing higher-level that can kill the frozen driver code. The freeze is complete and requires a hard restart.
Identify kernel-mode drivers: in Reliability Monitor, look at the “days” with critical failures and check what software was installed around those dates. Also test stability in Safe Mode (hold Shift + Restart → Troubleshoot → Advanced options → Startup Settings → Safe Mode with Networking) — Safe Mode loads only Microsoft-signed kernel drivers. If the machine is stable in Safe Mode but freezes normally, a third-party kernel driver is the cause. Common culprits: recently installed antivirus software (kernel-mode realtime scanning driver), VPN clients (network filter drivers), and system tuning software that hooks into the Windows kernel.
Power Supply Issues (Desktop Machines)
Desktop machines with aging power supplies sometimes deliver unstable voltages under load. When the 12V rail supplying the CPU or GPU dips below specification under peak load, the processor or GPU triggers protective shutdown — which Windows experiences as an instant freeze. This is more likely in systems with high-power components (gaming rigs, workstations) where the PSU is running near its rated capacity.
OCCT (free) includes a power supply stress test that monitors voltage stability during load. Voltage readings below 11.4V on the 12V rail under load indicate PSU instability. Event Viewer → Windows Logs → System → “Kernel-Power Event ID 41” (unexpected restart without clean shutdown) immediately after freeze events confirms the freeze was power-related rather than software. A PSU that’s failing its rated load capacity needs replacement — voltage stabilisation can’t be fixed in software.
USB Devices Causing System Freezes
Faulty USB devices or USB devices with buggy drivers can cause system-wide freezes by sending incorrect data through the USB stack that hangs the USB controller driver. The freeze pattern: happens specifically when inserting a USB device, or randomly while a specific USB device is connected but not actively being used.
Test: disconnect all non-essential USB devices (leave only keyboard and mouse if they’re USB). Run the machine without any other USB devices connected. If freezes stop, reconnect devices one at a time and observe. The device that triggers a freeze when connected is the cause. Try the device in a different USB port — some freezes are USB port-specific rather than device-specific. Updating or reinstalling the device’s driver from the manufacturer’s website resolves buggy-driver-caused freezes without needing to permanently remove the device.
Event Viewer After a Freeze
After a freeze that requires a hard restart, Windows logs what happened before the crash in Event Viewer. Windows Logs → System → look for entries around the last shutdown time with sources like:
- “Kernel-Power Event ID 41” — Unexpected shutdown (always appears after hard restart; the “reason” code within this event gives more detail)
- “LiveKernelEvent 141” — GPU driver timeout (display driver stopped responding before the freeze)
- “SYSTEM_THREAD_EXCEPTION_NOT_HANDLED” with a named driver — that driver caused the hang
These entries, combined with the freeze timing and the Reliability Monitor timeline, give enough information to identify the probable cause for most freeze scenarios without needing to run every possible hardware diagnostic. The named driver in a system thread exception event is particularly valuable — it immediately identifies which specific driver needs to be updated or rolled back.
When Nothing Works: Hardware Swap Testing
For persistent random freezes that survive all software fixes — driver updates, SFC, temperature management, RAM testing showing no errors — the issue is most likely a hardware component failure that the software diagnostics aren’t catching. Possible causes that diagnostics miss: failing RAM that only produces errors under specific data patterns, storage drive controller issues that appear intermittently, and GPU hardware faults that manifest under specific rendering workloads.
The definitive approach: swap hardware components one at a time if spare components are available. Try different RAM sticks (borrow from another machine, or swap between slots). Try a different storage drive (reinstall Windows on a clean drive and run without the original drive). Try a different GPU. The component swap that eliminates the freezes is the failing component. This is time-consuming but provides certainty that no amount of software diagnostic can deliver when hardware is intermittently failing.
Memory Integrity and Windows Security Features
Windows 11 enables Memory Integrity (also called Core Isolation) by default — it runs Windows in a hypervisor to protect core system processes. Some older drivers are incompatible with this virtualization layer and can cause system instability when Memory Integrity is enabled. The freezes often started after upgrading to Windows 11 or after enabling this feature.
Check: Settings → Privacy and security → Windows Security → Device Security → Core isolation → Memory integrity. If enabled: try temporarily disabling it (requires restart) and observe freeze frequency. Windows will warn about drivers that are incompatible — these driver incompatibilities are the root cause and should be addressed by updating those specific drivers. After updating the incompatible drivers, re-enable Memory Integrity for continued system security.
One final diagnostic worth knowing: Windows can be configured to write a complete memory dump on freeze — capturing the full system memory state at the moment of the crash. Settings → System → About → Advanced system settings → Startup and Recovery → System failure → set “Write debugging information” to “Complete memory dump.” The next freeze generates a large dump file that tools like WinDbg can analyse in detail, identifying the exact memory location and code path that caused the hang. This is the most thorough freeze diagnosis available and is what Microsoft support engineers use for persistent, unexplained freezes that resist all standard troubleshooting. See also Slow Upload Speed on Windows 11 for a related case.
