CLOCK_WATCHDOG_TIMEOUT (bug check 0x101) means one processor core stopped responding. Windows expects each active core to handle a regular clock interrupt, a steady tick it uses for timekeeping and scheduling. On a multi-core PC, when one core misses that tick for longer than Windows allows, the system stops. Microsoft’s description is blunt: the processor isn’t processing interrupts, typically because it’s unresponsive or deadlocked.
That’s a different failure from most stop codes, which catch a driver doing something wrong. Here a core has simply gone silent, so the usual suspects sit closer to the hardware: an overclock or undervolt, a memory profile, outdated BIOS microcode, heat or unstable power. Drivers and virtualization software are possible, just less likely. The stop code and fixes are the same on Windows 11 and Windows 10; on Windows 11 version 24H2 and later the stop screen may be black instead of blue.
The fastest tell is history. If someone tuned the BIOS or installed an overclocking tool, undo that first. If the PC runs at stock settings, update the BIOS next, especially on an Intel 13th or 14th Gen Core desktop. If it crashes only under sustained heavy load, look at cooling.
Try these first
- Undo any CPU overclock, undervolt or tuning profile, in BIOS setup or in an app such as Intel XTU or AMD Ryzen Master.
- Turn off the XMP or EXPO memory profile in BIOS setup as a test.
- Check CPU temperatures under load; make sure fans spin and vents are clear.
- Update the BIOS from the PC or motherboard maker, after a backup and with BitLocker suspended.
- Install the chipset drivers for your model from the PC maker (or AMD or Intel on a custom-built PC).
- Note whether a virtual machine, emulator or other virtualization software was running at the time.
Clock watchdog vs DPC watchdog
With clock watchdog timeout vs DPC watchdog violation, the names sound alike and both involve a timer, but they catch opposite problems.
| CLOCK_WATCHDOG_TIMEOUT (0x101) | DPC_WATCHDOG_VIOLATION (0x133) | |
|---|---|---|
| What stopped | A core stopped handling clock interrupts at all | A core spent too long at high priority, in one driver routine or many in a row |
| Is the core alive? | Effectively no: hung or deadlocked | Yes, but tied up |
| Typical causes | Overclock or undervolt, memory profile, BIOS microcode, heat, power delivery, a failing CPU | Storage, SSD firmware, Bluetooth audio, dock and graphics drivers |
| What the small dump shows | Often only the core that noticed the problem, not the one that hung | Often the driver that ran too long, though less reliably when parameter 1 is 1 |
| First move | Back to stock settings, then BIOS update | Find the driver, then update it and its firmware |
If your stop code is 0x133, our DPC watchdog violation guide has the driver-first approach. If 0x101 alternates with other hardware-flavored codes, such as UNEXPECTED_KERNEL_MODE_TRAP or WHEA_UNCORRECTABLE_ERROR, see our unexpected kernel mode trap guide as well.
The parameters
Parameter 1 is the timeout interval in clock ticks, parameter 3 is the address of the hung processor’s control block, and parameter 4 is the index of the hung processor, counted from 0. The blue screen doesn’t show them, but Reliability Monitor does: run perfmon /rel, select the crash and choose View technical details.
Parameter 4 counts logical processors, so with Hyper-Threading or SMT two numbers can share one physical core. The same number or pair crash after crash points at one core, not the whole system.
Step 1: put the processor and memory back to stock
On home-built and gaming PCs, the first suspect is any setting that pushes the processor past what it can sustain. Even a tune that passed a stress test months ago can turn unstable as the chip ages or the room warms.
- Reset CPU and memory settings to default. First suspend BitLocker and have the recovery key ready, because a TPM or Secure Boot change can trigger a recovery prompt. Then change back the individual settings you or the builder tuned, or load BIOS defaults, often labeled Load Optimized Defaults or Load Setup Defaults. A full reset can also change the storage mode and stop Windows starting with
INACCESSIBLE_BOOT_DEVICE, so note the storage mode (AHCI, RAID or VMD), boot order, Secure Boot, TPM and virtualization settings before you do. On Intel 13th and 14th Gen desktops, an older BIOS’s defaults can exceed Intel’s power guidance; the BIOS update and Intel Default Settings in Step 2 return those to stock. - Turn off memory profiles. XMP (Intel) and EXPO (AMD) run memory faster than the default speed, which also stresses the processor’s memory controller. Test without them for a few days; if the crashes stop, leave them off.
- Remove automatic overclocking. Many motherboards add their own boost features, and AMD’s Precision Boost Overdrive and Curve Optimizer raise power limits or shift the voltage curve. Set them to default or disabled.
- Remove undervolts and tuning apps. An undervolt saves heat by lowering voltage, and too little voltage is a classic way to hang a core. Reset Intel XTU, AMD Ryzen Master, ThrottleStop or the PC maker’s performance app, and uninstall it if it reapplies settings at startup.
Business desktops and laptops from Dell, HP or Lenovo are rarely overclocked, so if that’s what you have, go straight to the BIOS step. Remote tools run inside Windows, so BIOS setup changes need someone at the keyboard.
If it crashes before Windows finishes starting
BIOS setup runs before Windows loads, so you can still undo overclocks and memory profiles there. If 0x101 still hits during startup, have your BitLocker recovery key ready and start in Safe Mode with the steps in our blue screen of death guide. Safe Mode loads only a minimal set of drivers, so a crash there points at hardware rather than software.
Step 2: update the BIOS and processor microcode
Microcode is the processor’s own low-level firmware, and fixes for it reach your PC mainly through BIOS updates from the PC or motherboard maker, so check the BIOS release notes for your exact model.
Intel 13th and 14th Gen Core desktop processors
A clock watchdog timeout on a 13th or 14th Gen Core desktop has one more suspect. Intel has confirmed an instability issue in some 13th and 14th Gen Core desktop processors, which it calls Vmin Shift Instability. In its root-cause update, Intel traced it to a clock circuit inside the processor cores that degrades faster under high voltage and temperature. Intel says its 13th and 14th Gen laptop processors aren’t affected. To check, open Settings > System > About and read Processor under Device specifications; a desktop model number starting with 13 or 14, such as i7-13700K or i9-14900K, may be affected.
As of October 2026, Intel’s guidance is the latest BIOS with microcode 0x12F or later, used with the Intel Default Settings profile (labels vary by board maker). 0x12F followed the consolidated 0x12B fix and targeted crashes on systems running for days at light load; release notes often state the microcode version. Microcode is preventive, so a processor that’s already unstable may need replacing; Intel has extended the warranty on eligible chips, so check its support article for current terms. If yours is one and crashes with 0x101 or a mix of codes, update the BIOS before anything else.
Do the update safely
A BIOS update that’s interrupted can leave the PC unable to start. Back up your files first, plug a laptop into its charger, and never restart or power off while it runs. Suspend BitLocker beforehand and have your BitLocker recovery key ready, because a firmware change can trigger a recovery prompt at the next start; afterward, check that protection is on again.
A BIOS update often resets settings, so recheck that memory profiles and overclocks didn’t come back.
In an office, 0x101 on several identical PCs soon after a BIOS rollout points at the release, not each machine. Test each BIOS on one PC for a week before rolling it out with the maker’s fleet tool, and keep every PC of a model on the same version.
Step 3: heat and power delivery
A processor that overheats throttles first, but cooling that’s barely adequate can still leave a core unstable under sustained load. Watch temperatures with the PC maker’s utility or a hardware monitoring app while the PC is busy (the BIOS hardware monitor shows only readings at rest), and look for crashes that cluster after long renders, compiles, games or warm afternoons.
- Desktops: check that the CPU cooler is seated firmly and its fan spins, and that the processor’s power cable, usually near the top edge of the motherboard, is fully plugged in. An old or undersized power supply can sag under load.
- Laptops: test on the original charger, on a hard surface, with vents clear. A laptop that crashes only on battery or a smaller charger points at power delivery or power management, which BIOS and chipset updates address, more than at an ordinary driver.
Heat isn’t the only pattern: on an affected Intel desktop, a crash at idle still points to the microcode step, not cooling.
Rather have a technician sort it out?
With your approval, a technician can review your crash dumps and BIOS and chipset versions, update the software side and tell you plainly whether the pattern points to hardware that needs hands-on testing. A GoProIT technician can connect remotely, with your permission, and fix it while you watch. $59 per fix, Mon–Fri 8am–8pm ET.
Step 4: chipset drivers and virtualization software
With hardware settings ruled out, look at software that works closest to the processor.
- Chipset drivers. On AMD systems, the chipset package includes power-management components that control how cores park, idle and wake; Intel’s package mainly installs INF files that let Windows identify the chipset’s devices. Install the version for your model from the PC maker, or from AMD or Intel on a custom-built PC. Windows’ own processor power drivers,
intelppm.sysandamdppm.sys, manage core idle states and sometimes appear in 0x101 dumps. Update the chipset drivers and BIOS rather than disabling them. - Virtualization software. Hyper-V and features built on it, such as Windows Sandbox and WSL 2, plus VMware Workstation, VirtualBox and Android emulators, use the processor’s virtualization features directly. If 0x101 started after you installed or updated one, or turned on a Windows virtualization feature, update it or remove it for a few days to test. Don’t turn off Memory integrity or other security features as a test, and on a company-managed PC, ask whoever manages it before removing Hyper-V. If the blue screen appears only inside a virtual machine’s window, the guest Windows crashed: update the virtualization app and the VM’s guest tools.
- Windows updates. Install pending updates, including optional driver updates from the PC maker.
A driver that disables interrupts and never releases a core can cause 0x101 too, which is where a kernel memory dump (next section) earns its keep: it can show what the hung core was running, though a fully frozen core sometimes leaves little to read. If it names a third-party driver, update or remove that driver.
If it keeps crashing: save a kernel memory dump
A small memory dump records only the processor and thread that raised the crash, and Microsoft notes that problems not caused by that thread may not show up in it. With 0x101, that’s a healthy core noticing a hung one, so the minidump often blames Windows itself.
A kernel memory dump captures all the memory in use by the kernel, which often lets a technician look at the hung processor as well. To keep one:
- Search for Advanced system settings and, under Startup and Recovery, select Settings.
- Set Write debugging information to Automatic memory dump or Kernel memory dump.
- Select Disable automatic deletion of memory dumps when disk space is low. Without it, a PC that isn’t on a domain can delete
MEMORY.DMPwhen the system drive has less than about 25 GB free. Also keep a page file on the system drive.
The dump is saved as C:\Windows\MEMORY.DMP and the next crash overwrites it, so copy it somewhere safe. In WinDbg, a technician can switch to the hung processor with the commands Microsoft lists for this bug check, such as !prcb; opening dumps is covered in the blue screen of death guide linked in Step 1.
When it’s the processor or motherboard
Treat it as hardware when:
- Crashes continue at default BIOS settings, with the latest BIOS, normal temperatures and no virtualization software.
- The Intel Processor Diagnostic Tool, or the PC maker’s hardware diagnostics, reports a processor failure.
- 0x101 alternates with other hardware stop codes, or crashes happen in Safe Mode too.
- The PC also freezes or restarts in BIOS setup or the maker’s pre-boot diagnostics.
- Parameter 4 shows the same number, or the same pair, every time, even after the steps above.
The processor and motherboard can be hard to tell apart without swapping parts. On a pre-built PC, run the maker’s diagnostics and open a warranty case with the PC maker, which handles the processor inside it, even an affected Intel 13th or 14th Gen chip. On a custom-built PC, a boxed Intel processor goes through Intel’s warranty and a tray processor through the place of purchase. Replacing parts is hands-on work, and parts aren’t included in a remote fix. An on-site visit can be arranged through vetted local partner technicians as a separate paid add-on, where available, or you can use the maker’s warranty service.
The software side, from dump review to chipset drivers and Windows updates, is remote work. If you’d rather hand it over, it’s a typical $59 remote fix as long as the PC stays up long enough to connect. Every session needs your approval, and you can end it at any time; here’s how our remote support works.
Common questions
Is CLOCK_WATCHDOG_TIMEOUT a CPU problem?
Often, but not always a faulty CPU. The stop code means a core stopped responding, and common reasons are settings that push the processor too hard, outdated BIOS microcode or heat. A failing processor or motherboard becomes the likely answer only when crashes continue at default settings with current firmware and normal temperatures, or when diagnostics report a fault.
Can RAM cause clock watchdog timeout?
Indirectly, yes. Running memory faster than its default speed with an XMP or EXPO profile also stresses the memory controller, which is part of the processor, and that can hang a core. Turn the profile off as a test. If crashes continue, run Windows Memory Diagnostic, and if it reports errors, test one memory module at a time.
Can a graphics driver cause CLOCK_WATCHDOG_TIMEOUT?
It’s possible but less common than with other stop codes. A driver that keeps a core busy with interrupts disabled can stop that core from answering the clock, and graphics drivers are among the most complex on a PC. If a kernel memory dump shows a graphics driver on the hung core, reinstall the version your PC maker supports and retest.
Does CLOCK_WATCHDOG_TIMEOUT damage my computer?
Rarely, from the crash itself. Like any sudden stop, it loses unsaved work and can occasionally corrupt a file or update that was being written, so keep backups current. The underlying cause can be another matter: on affected Intel 13th and 14th Gen desktop processors, Intel says the instability comes from a circuit that degrades under high voltage and temperature, which is why the BIOS update shouldn’t wait.
Why does clock watchdog timeout happen when the PC is idle?
Idle isn’t always gentle on a processor. Cores constantly drop into low-power states and wake back up, and a marginal voltage setting, undervolt or BIOS bug can hang a core during those changes. On affected Intel desktops, Intel’s 0x12F microcode specifically targeted crashes on systems running at light load for days. Update the BIOS and chipset drivers and remove any undervolt.
Will reinstalling Windows fix CLOCK_WATCHDOG_TIMEOUT?
Rarely. The usual causes, such as overclock and memory settings, outdated BIOS microcode, heat and a degraded processor, sit below Windows, so a clean install leaves them in place. A reinstall helps only when a driver or virtualization tool can’t be removed any other way. If the PC shows 0x101 while Windows is installing from a USB drive, that’s strong evidence of a hardware problem, so test the processor and memory instead.