by The Computing Australia Group | PC won’t start or keeps
restarting
The Central Processing Unit (CPU) is often called the “brain” of your computer for a reason: it coordinates instructions, processes calculations, and keeps everything – from web browsing to video editing – moving. When the CPU is healthy, your PC feels responsive and stable. When the CPU is struggling (or the conditions around it are failing), you’ll often notice symptoms long before a total breakdown.
The tricky part is that many “CPU failure” symptoms can also be caused by other components – RAM, storage, the power supply, motherboard VRMs, GPU, drivers, or even overheating from dust buildup. That’s why the best approach is to learn the most common warning signs, understand what they usually mean, and follow a structured troubleshooting process before you replace anything.
In this guide, we’ll modernise and expand the typical symptom list and show you how to diagnose CPU – related issues in a practical, low-risk way – whether you’re supporting a business workstation, a gaming PC, or a laptop that’s started misbehaving.
Quick reality check: Is it really the CPU?
True CPU failure is less common than problems around the CPU. Many CPUs last for years (often a decade) without “dying,” but the CPU can still appear to fail when something else prevents it from operating correctly – such as:
- Thermal issues (cooler not seated, dried thermal paste, failing fan, dust blockage)
- Power delivery issues (PSU instability, motherboard VRM degradation)
- Memory faults (bad RAM, unstable XMP/EXPO profiles)
- BIOS/UEFI misconfiguration (overclocking, undervolting, wrong power limits)
- Driver or firmware conflicts (chipset drivers, BIOS updates, microcode updates)
- Motherboard socket damage (bent pins, debris in socket, cracked solder joints)
So rather than jumping straight to “the CPU is dead,” it’s best to treat the CPU as part of a system where heat + power + stability must all line up.
Common CPU failure symptoms (and what they usually indicate)
1) Random computer freezes (especially soon after login)
What it looks like
- PC boots, you log in, then it locks up without warning
- Mouse/keyboard stop responding
- You may hear system beeps or audio stutter
- Restarting temporarily “fixes” it - until it happens again
Why it happens
Freezing can occur when the CPU is:
- overheating and throttling hard
- encountering hardware-level errors
- starved of stable power
- affected by unstable RAM settings that look like CPU failure
What to check first
- CPU temperatures at idle and under light load
- Background processes spiking CPU usage
- BIOS overclocking settings (including “auto OC” features)
- Windows Event Viewer (WHEA errors are a major clue - more below)
2) Booting issues (blank screen, no POST, no beep, no peripherals)
What it looks like
- Fans spin, lights come on, but there’s no display
- Keyboard doesn’t light up, Caps Lock doesn’t toggle
- You don’t hear the usual POST beep (if your system has a speaker)
Why it happens
“No POST” can happen with CPU failure, but it’s also common with:
- motherboard failure
- RAM not seated properly / RAM failure
- GPU failure (on systems without integrated graphics)
- PSU failure
- BIOS corruption or incompatible BIOS version (especially after upgrades)
Many newer boards have debug LEDs or a Q-Code display. If the CPU LED stays lit or the code consistently points to CPU init, that’s meaningful. If you’re comfortable opening the case, these indicators can save hours.
3) Sudden shutdowns (especially under load)
What it looks like
- PC turns off abruptly (no graceful shutdown)
- Often happens during gaming, exporting video, compiling code, or running heavy apps
- You don’t hear the usual POST beep (if your system has a speaker)
Why it happens
The most common culprit is temperature or power:
- CPUs are designed to protect themselves by throttling first, and shutting down if temperatures are unsafe; this is a normal protection behaviour in modern systems.
- Unstable PSU or motherboard VRMs can also trigger shutdowns when the CPU draws more power.
What to check first
- CPU cooler mounting pressure and thermal paste condition
- Dust in heatsink fins and case filters
- CPU temperature under load
- Power supply quality/age, and whether power cables are seated properly
4) Blue Screen errors (BSOD) that point to hardware-level faults
What to check first
- Windows displays a stop code like:
- WHEA_UNCORRECTABLE_ERROR
- CLOCK_WATCHDOG_TIMEOUT
- The system may reboot immediately or loop on startup
Why it happens
Some stop codes are strongly associated with CPU/hardware problems (though not exclusively):
- WHEA_UNCORRECTABLE_ERROR indicates Windows detected a fatal hardware error. Microsoft Support +1 In debugging terms, it’s commonly linked to bug check 0x124.
- CLOCK_WATCHDOG_TIMEOUT indicates an expected clock interrupt wasn’t received on a secondary processor core in time, which can indicate a hung core or low-level system instability (often hardware/firmware/driver related).
What to do next
- If BSODs happen frequently, you’re past “minor glitch” territory.
- Disable overclocks (including XMP/EXPO temporarily), update BIOS/chipset drivers, and test stability step-by-step.
5) Beep alerts during startup (POST beep codes)
What it looks like
- A pattern of beeps when you power on
- No display output, or the system halts
What it looks like
Beep codes are not universal – they vary by manufacturer and BIOS/UEFI implementation.
What to do
- Identify your motherboard or laptop brand/model
- Look up the beep code chart for that specific vendor
- Combine that info with other evidence (debug LEDs, logs, behaviour under load)
6) Visible signs of damage near the CPU socket
What it looks like
- Burn marks on/around the socket area
- Discoloured board near VRMs
- Smell of burnt electronics
- Bent pins (common on some socket types) or debris in the socket
What it usually means
- Sustained overheating
- Power delivery fault
- Liquid damage
- Physical damage during an upgrade (CPU installation, cooler mounting)
In these scenarios, the issue may be CPU or motherboard (often both), and “trial and error” powering on repeatedly can worsen damage.
Modern troubleshooting workflow (safe, structured, and practical)
Step 1: Protect your data first
If the system is still booting sometimes:
- Back up critical files immediately (cloud + external drive)
- If shutdowns/freezes are severe, consider pulling the drive and backing up from another machine
Step 2: Check thermals (the most common cause)
- Clean dust from filters, fans, and heatsinks
- Confirm all fans spin and airflow is correct
- Re-seat the CPU cooler and apply fresh thermal paste if temps are suspicious
- Monitor temperature at:
- idle
- light use (browser + email)
- sustained load (a stress test)
Rule of thumb: If the system shuts down under load and temps are spiking rapidly, treat cooling as your top suspect before you blame the CPU.
Step 3: Remove instability multipliers
In BIOS/UEFI:
- Disable overclocking
- Reset to defaults
- Temporarily disable XMP/EXPO (RAM “one-click” overclocks)
- Save and retest
This step alone resolves a surprising number of “CPU failure” cases.
Step 4: Look for hardware error signals in Windows
If you can boot into Windows:
- Check Event Viewer for WHEA-related hardware events
- Note the exact BSOD stop code text
- If you’re analysing dumps, remember:
- Bug check 0x124 maps to WHEA hardware error conditions
- Bug check 0x101 maps to clock watchdog timeout conditions
Step 5: Isolate components (the “minimum boot” test)
For desktops (if you’re comfortable):
- Boot with only:
- CPU + cooler
- one RAM stick
- boot drive (optional for POST testing)
- integrated graphics (if available) or GPU
- Remove non-essential peripherals and extra drives
- Swap RAM sticks/slots one at a time
For desktops (if you’re comfortable):
- Ensure vents are clear
- Use manufacturer diagnostics if available
- Consider professional servicing if thermal paste replacement is needed (laptops are easier to damage during disassembly)
Step 6: Decide whether it’s repairable or replaceable
A CPU rarely “half-fails” in a clean way – many issues are environmental (cooling/power/BIOS). But replacement becomes likely when:
- You’ve verified stable cooling and power delivery
- BIOS defaults still crash
- Known-good RAM doesn’t help
- The system repeatedly throws hardware-level stop codes
- You see physical damage around socket/VRMs
What typically causes CPU-related failures (and how to prevent them)
Overheating (and repeated thermal cycling)
Even with modern protections like throttling and shutdown safeguards, chronic overheating stresses the whole platform.
Prevention: clean dust, maintain airflow, replace thermal paste periodically, and avoid running laptops on soft surfaces that block vents.
Power instability
Cheap or aging power supplies, unstable power rails, and weak motherboard VRMs can cause crashes that masquerade as “CPU failure.”
Prevention: use quality PSUs, avoid cheap adapters, and ensure correct power connectors are used (especially EPS/CPU power).
Overclocking and “auto OC” features
Overclocks can be stable for months and then become unstable after BIOS updates, heat changes, or component aging.
Prevention: keep conservative settings, stress test after changes, and document your known-good BIOS profile.
BIOS/UEFI and chipset driver mismatches
Platform firmware matters more than most people realise – especially on newer CPUs.
Prevention: keep BIOS and chipset drivers current, but update carefully (read release notes, avoid interrupting updates).
Understanding what the symptoms of CPU failure are can help to prevent extensive damage. However, computers can sometimes throw symptoms at you that need trained eyes and ears to be decoded. If you ever find yourself in such a situation, you can reach us at helpdesk@computingaustralia.group for 24-hour helpdesk service.
Jargon Buster
Motherboard – is a printed circuit board that connects all key components and allows communication between them for smooth system operation.
Error codes – are codes displayed on a blue screen during unsuccessful booting. Each error code signifies a particular problem.
POST test – Power On Self Test is a series of test run by a computer at the start of the booting process to check if all hardware is working properly.
FAQ
Can a PC freeze because of RAM, not the CPU?
Yes. Unstable or failing RAM can cause freezes, BSODs, failed boots, and app crashes – often indistinguishable without testing.
Does WHEA_UNCORRECTABLE_ERROR always mean the CPU is dead?
No. It indicates a fatal hardware error detected by Windows, but the underlying cause can include CPU, RAM, motherboard, storage, power delivery, or drivers/firmware.
What does CLOCK_WATCHDOG_TIMEOUT usually point to?
It indicates an expected clock interrupt wasn’t received in time (bug check 0x101), often linked to low-level instability that can involve the CPU, BIOS, drivers, or hardware configuration.
Why does my PC shut down to “save itself”?
Modern processors include thermal protection behaviours like throttling and shutdown when temperature limits are exceeded.
Why does my PC freeze randomly after logging in?
This can be caused by CPU throttling, unstable power, driver issues, storage problems, or unstable RAM settings. If it happens consistently, check temperatures first, then test RAM stability, then review Event Viewer for hardware errors.
