Computer Keeps Freezing? Diagnose It by Freeze Type (2026)

By LK Wood IV · 2026-07-05 · ~9 min read · St. Louis County, MO

Some retailer links below are Amazon affiliate links; if you buy through them TechFuel HQ may earn a small commission at no extra cost to you. As an Amazon Associate I earn from qualifying purchases. Commissions never influence what gets recommended — see our disclosure.

Most freezing guides — and the AI answer at the top of the results — hand you the same generic list: run sfc /scannow, clear the dust, update Windows. Sometimes that works. Often it doesn’t, because a computer freezes for very different reasons, and the fix for one type does nothing for another. The move that actually saves time is to sort the freeze first, then read what Windows already logged about it — and only then fix the specific thing.

Step 1: Sort the freeze type (this decides everything)

Before you change a single setting, figure out which of these you have. The type points straight at the likely cause:

If you’re not sure which you have, watch for one tell: does the mouse cursor still move during the freeze? A cursor that moves while everything else is stuck is a soft hang (software). A cursor frozen solid with everything else is a hard freeze (hardware). That single observation rules out half the causes.

Here’s the whole method on one grid. Find your freeze type on the left, and the bright cells are where to start — the dim ones are worth a look only if the bright ones come up clean. This is why sorting the type first saves hours: it crosses out most of the columns before you touch a screwdriver.

Step 2: Read what Windows logged (the step everyone skips)

Windows keeps a record of crashes and freezes. Reading it is faster than trying random fixes, and it’s the part almost every guide leaves out.

Start with the friendly view:

1. Press Windows Key + R, type perfmon /rel, and press Enter to open Reliability Monitor.
2. You’ll see a timeline with red X marks on the days things went wrong. Click the day of a freeze to read a plain-English summary of what failed.

Then get the detail:

1. Press Windows Key + R, type eventvwr.msc, and press Enter.
2. Go to Windows Logs → System, and look at the events stamped at the time of your freeze.

Three events tell you most of what you need:

• Kernel-Power, Event ID 41 — Windows lost power or was hard-reset without a clean shutdown. This is the fingerprint a hard freeze leaves when you had to hold the power button. It’s a symptom, not the cause — but combined with “it only happens under load,” it points at power or heat.
• WHEA-Logger (Event ID 17, 18, or 19) — the hardware itself reported an error. This is a strong signal of CPU, RAM, or memory-controller instability (often an unstable memory overclock/EXPO profile, or a genuinely failing part).
• “Display driver stopped responding and has recovered” / LiveKernelEvent 141 — the graphics driver timed out and Windows reset it (a “TDR” event). Your fix lives in Step 5.

Match the event to the step below. If Event Viewer is empty at the freeze time, that itself is a clue: a freeze that logs nothing is usually a dead-hard hardware or power cut.

Here’s how to read the three fingerprints at a glance — the event name is the label Windows writes; the verdict is what it actually means; the arrow is where to go next.

Step 3: Hard freezes → check the hardware, in order

If it’s a total lock (and especially if you saw Kernel-Power 41 or WHEA-Logger), work these in order:

• RAM. Faulty or unstable memory is the classic cause of sudden hard freezes. Run the built-in Windows Memory Diagnostic (Windows Key + R → mdsched.exe → “Restart now and check for problems”), or MemTest86 from a USB stick for a thorough overnight pass. If you run an EXPO/XMP profile, set memory back to its default JEDEC speed and test — an overclock the kit or controller can’t hold freezes a system that’s otherwise healthy. The same bad or unstable memory often blue-screens instead of hanging, most often as PAGE_FAULT_IN_NONPAGED_AREA, which is worked in the same RAM-first order.
• Heat. A CPU or GPU parked at its thermal limit can stall the whole system. Check load temps with HWiNFO; our safe CPU temperature guide covers what’s actually too hot (Ryzen to ~95 °C, X3D ~89 °C, Intel ~100 °C). Clear dust from fans and heatsinks with a can of compressed air, and if the paste is a few years old, a fresh tube of Arctic MX-4 often drops load temps several degrees.
• Storage. A failing or nearly-full system drive causes hangs and hard freezes. Check the drive’s SMART health with CrystalDiskInfo (anything but “Good” is your answer), and make sure the system drive has at least 15–20 GB free — Windows freezes when it can’t page or write.
• Power. If freezes are load-related and log nothing, suspect the PSU. Reseat every power cable (including all pins on a 12VHPWR/12V-2×6 GPU connector), and if a hungry GPU sits on an old supply, that’s a common cause. This overlaps heavily with reboots — if it sometimes restarts instead of freezing, work our PC keeps restarting guide alongside this one.

Step 4: Soft hangs → free the stuck resource

If the cursor still moves and apps go “Not responding,” you’re waiting on a resource. Open Task Manager (Ctrl+Shift+Esc) before the next hang, or right after, and watch the columns:

• Disk pinned at 100%. The usual culprits are SysMain (SuperFetch) and Windows Search indexing thrashing the drive — common on systems with a hard drive or an aging SSD. Open services.msc, find SysMain, set Startup type to Disabled, and stop it; do the same for Windows Search as a test. If the hangs stop, that was it.
• Memory near 100%. Too little free RAM for what’s open forces constant paging and stutter. Close heavy background apps and browser tabs; if it’s chronic, add memory. On a 8 GB machine in 2026, a browser plus a game will do this on its own. (If the game hitches rather than locking up the whole PC, that’s frame-pacing, not a freeze — see why your PC stutters in games.)
• A specific process spiking. Sort by CPU or Disk to catch a misbehaving background app or a stuck update. A stuck Windows Update or a runaway antivirus scan freezes weak systems until it finishes.

A special case — brief freezes every few seconds, in a rhythm. If the machine hitches for a second or two on a repeating beat (the mouse catches up, then stalls again), that pattern points somewhere specific, not at general resource starvation. Check two things in order. First, the drive: a disk with failing or pending sectors stalls the whole system while it retries a read, and the stalls come in waves — run CrystalDiskInfo (look for a “Caution” SMART status or reallocated/pending sector counts) and, on a mechanical drive, listen for repeated clicking. Second, driver latency (DPC): a badly-behaved driver — most often network (Wi-Fi/Ethernet), audio, or a Bluetooth/USB device — can hog the CPU in short bursts and freeze the desktop periodically even when Task Manager looks idle. The free tool LatencyMon names the offending driver directly; a network or audio driver at the top of its list is your lead. Update or roll back that specific driver rather than reinstalling Windows.

Step 5: Load / gaming freezes → the graphics driver

If it only freezes under a game or GPU load — or Event Viewer showed a “display driver stopped responding” / TDR event — the graphics driver is the prime suspect:

• Clean-reinstall the GPU driver. A plain reinstall often isn’t enough; do a full clean removal with DDU and a fresh driver. The complete process is in our GPU driver crash fix guide.
• Confirm it’s not heat or power in disguise. Load freezes are also how a marginal PSU or an overheating GPU shows up — check GPU temps and hotspot in HWiNFO while you play. If temps are fine and a clean driver doesn’t fix it, treat it as power (Step 3).

Step 6: Idle or after-sleep freezes → power states

A PC that’s fine while you use it but freezes on idle, or locks up right after waking from sleep, is almost always a power-state or driver-power problem:

1. Turn off Fast Startup. Control Panel → Power Options → “Choose what the power buttons do” → “Change settings that are currently unavailable” → uncheck “Turn on fast startup.” Fast Startup half-hibernates the PC and is a frequent cause of post-wake freezes and cold-boot lockups.
2. Set the power plan to Balanced or High performance (not a third-party or aggressive custom plan), and in the plan’s advanced settings, turn PCI Express Link State Power Management to Off and disable USB selective suspend if a USB device wakes the system badly.
3. Update chipset and storage drivers from your motherboard maker — idle freezes often trace to a power-management bug in an old chipset or NVMe driver.

Step 7: The basics worth doing anyway

Regardless of type, these are cheap and occasionally the whole fix:

• Run sfc /scannow, then DISM /Online /Cleanup-Image /RestoreHealth, to repair corrupt system files.
• Install pending Windows updates (and optional driver/firmware updates), then reboot.
• Scan for malware — a hidden miner or bad process freezes a system under its own load.
• Keep 15–20 GB free on the system drive.

The fix order, in one list

1. Sort the freeze type — hard lock (hardware) vs soft hang (software) vs load vs idle. Cursor still moving = soft.
2. Read the logs — perfmon /rel then Event Viewer → System. Kernel-Power 41 (power/heat), WHEA-Logger (CPU/RAM), display-driver-stopped (GPU).
3. Hard freeze → RAM (mdsched/MemTest86, test at JEDEC), temps, drive SMART + free space, PSU cables.
4. Soft hang → Task Manager: 100% disk (disable SysMain/Search), low RAM (close/add), a spiking process.
5. Load/gaming → clean-reinstall the GPU driver (DDU), recheck GPU temps/power.
6. Idle/after sleep → turn off Fast Startup, fix the power plan, update chipset/NVMe drivers.
7. Basics → sfc/DISM, updates, malware scan, free space.

Work it in that order and the freeze type plus the logged event usually points you straight to the right step — no guessing, no parts-cannon.

Sources

• The diagnostics and fixes use Microsoft’s built-in tools — Reliability Monitor (perfmon /rel), Event Viewer (eventvwr.msc), Windows Memory Diagnostic (mdsched.exe), services.msc, Power Options, and sfc /scannow + DISM /Online /Cleanup-Image /RestoreHealth — all standard and self-verifiable on any Windows 10/11 PC. The event signatures (Kernel-Power 41 = unclean shutdown/hard-reset; WHEA-Logger 17/18/19 = hardware-reported error; LiveKernelEvent 141 / “display driver stopped responding” = a graphics TDR event) are documented by Microsoft and cross-corroborated across community troubleshooting threads (Microsoft Q&A, Tom’s Hardware forums). A few widely used third-party diagnostics are named where the built-in tools don’t cover the job: CrystalDiskInfo and MemTest86 for drive-SMART and memory testing, HWiNFO for load temperatures, and LatencyMon for pinpointing the driver behind rhythmic, every-few-seconds micro-freezes (a DPC-latency pattern the generic “close some tabs” advice never catches).
• Thermal limits and what “too hot” means: our safe CPU temperature guide (first-party). Graphics-driver clean reinstall: our GPU driver crash fix guide (first-party). When it reboots instead of freezing: our PC keeps restarting guide (first-party). When it’s the network dropping rather than the whole PC locking up: our WiFi keeps disconnecting guide (first-party). When a game closes to the desktop instead of the whole PC freezing: our games keep crashing on PC guide (first-party).

Steps are for Windows 10 and 11 in 2026; menu names vary slightly by version. Found an error? Email hello@techfuelhq.com with the article URL.