Is My CPU Overheating? Signs Causes and How to Fix It

CPU overheating occurs when a processor exceeds its safe thermal threshold, typically 90–100°C depending on the model, triggering thermal throttling, system instability, or emergency shutdown to prevent permanent hardware damage.

Last updated: May 2026

Your PC ran fine for months. Now it’s throttling mid-game, crashing to desktop, or shutting off without warning. Something changed, and that something is almost always diagnosable and fixable without buying a whole new system.

This guide covers three situations you might be dealing with: overheating that started suddenly out of nowhere, overheating that seems to happen for no obvious reason, and overheating that persists even with a decent AIO cooler installed. All three have different root causes, and all three have clear fixes. Work through this systematically and you’ll know exactly what’s wrong before you finish reading.

What Temperature Is Too Hot for a CPU?

Before you can fix anything, you need to know what “overheating” actually means for your specific chip. Saying “80°C is too hot” is too vague, it ignores the fact that different processors have completely different thermal budgets.

Safe vs. Dangerous CPU Temperature Ranges

Condition	Safe Range	Caution Zone	Critical / Throttle Point
Idle (desktop)	30–45°C (86–113°F)	46–60°C	60°C+
Light load (browsing, video)	45–65°C	66–75°C	76°C+
Heavy load (gaming, rendering)	65–85°C	86–94°C	95°C+
Stress test (Prime95, Cinebench)	75–90°C	91–99°C	100°C / TjMax

Temperature Limits by CPU Generation and Model

TjMax (Thermal Junction Maximum) is the highest temperature a CPU can reach before it automatically throttles or shuts down. It’s baked into the chip’s design, not a soft setting you can change. Here’s how major current CPUs compare:

CPU Model	TjMax	Typical Idle	Typical Gaming Load
Intel Core i9-14900K	100°C	40–55°C	85–100°C
Intel Core i9-13900K	100°C	35–50°C	80–95°C
Intel Core i7-12700K	100°C	30–45°C	75–88°C
Intel Core i5-12400	100°C	28–40°C	60–75°C
AMD Ryzen 9 7950X	95°C	40–55°C	75–90°C
AMD Ryzen 7 7700X	95°C	35–50°C	70–85°C
AMD Ryzen 5 5600X	95°C	30–45°C	60–75°C

One important thing about AMD Ryzen 7000 series chips: AMD’s official documentation explicitly states that these processors are designed to operate up to 95°C during normal workloads. The chip’s Precision Boost algorithm actively targets that thermal ceiling to extract maximum performance. If your Ryzen 7700X is hitting 90°C under gaming load and you’re wondering why your CPU is overheating “for no reason”, it probably isn’t. That’s by design.

Is 120°F hot for a CPU? Short answer: no. 120°F converts to roughly 49°C, which sits comfortably in the normal range for light-to-moderate load. You can find a full breakdown of normal temperature ranges by workload in our guide to what is a good CPU temp for idle, gaming, and workloads.

Signs Your CPU Is Overheating

Symptoms vary a lot depending on how hot things actually get. Here’s how to read what your system is telling you.

Early Warning Signs (Don’t Ignore These)

• Fans suddenly ramping to full speed, the system is trying to dump heat fast.
• CPU clock speeds dropping mid-task, thermal throttling. In Task Manager, you’ll see CPU usage sitting at 100% but actual clock speeds dropping well below base clock. Performance tanks as a result.
• Sluggish performance despite low CPU usage %, throttled clocks can make a CPU look idle while actually struggling.
• Stuttering in games that ran fine before, frame pacing suffers when the CPU can’t hold boost clocks.

Severe Overheating Symptoms

• BSOD with WHEA_UNCORRECTABLE_ERROR, the CPU is producing corrupt data due to thermal stress.
• Sudden, instant power-off, no warning, no shutdown sequence. The board cut power to protect the chip.
• Crashes on startup or immediately after POST, the CPU is already too hot from a previous session and can’t stabilize.
• Boots then shuts down within seconds, thermal protection firing almost immediately. Serious.

How to Check Your CPU Temperature Right Now

Don’t guess. Get a number. Here are the best tools:

• HWiNFO64 (recommended), most detailed, shows per-core temps, TjMax distance, pump RPMs, and fan speeds all in one view. Free at hwinfo.com.
• Core Temp, lightweight, straightforward, great for a quick check.
• Ryzen Master, AMD only. Official tool with per-core voltage and temp data.
• Intel XTU, Intel only. Shows power limits and thermal data in one place.
• BIOS/UEFI hardware monitor, works without any OS install. Good when the system won’t stay stable long enough to run software.

To use HWiNFO64: install and launch it, click “Sensors Only,” then scroll to the CPU section. Look for the “CPU Package” temperature row. The “Distance to TjMax” column shows how many degrees you are from the throttle point, higher is better.

Why Is My CPU Overheating? 8 Most Common Causes

This is where most guides fall short. They list two or three causes and call it done. There are eight distinct failure points worth understanding, and some of them are genuinely surprising.

1. Dried Out or Incorrectly Applied Thermal Paste

Thermal paste fills the microscopic gaps between your CPU’s integrated heat spreader and the cooler’s contact plate. Over time it dries, cracks, and loses conductivity. Standard compounds like Arctic MX-4 typically last 2–5 years before degrading. Premium options like liquid metal can hold up 8 years or more.

Dried paste looks cracked, powdery, or chalky when you remove the cooler. But bad paste application matters too. Too much paste causes it to squeeze over the CPU edges and potentially contact pins. Too little leaves air gaps. Both hurt temps significantly. A pea-size dot centered on the IHS is the correct amount for most CPUs, the cooler pressure spreads it.

If you’re asking why your CPU is overheating after applying thermal paste, the most common culprits are: spreading it manually with a card (introduces air bubbles), applying too thick a layer, or not cleaning off the old compound completely before reapplying. You need 90%+ isopropyl alcohol and a lint-free cloth for proper removal.

2. Cooler Not Properly Seated or Mounted

This is the single most common cause of CPU overheating, full stop. Intel’s push-pin mechanism degrades over time, and if a cooler has been removed and reinstalled, the pins often don’t re-engage correctly, leaving one or more corners slightly raised and causing uneven contact.

For AMD AM5 builds, the correct mounting torque is 0.6 N⋅m per screw, applied in a cross pattern. Over-tightening can actually warp the cooler base. Under-tightening gives poor contact. The tell for bad mounting is consistent: temps spike to 90°C+ within 10–15 seconds of any real load, then stabilize or keep climbing. Good mounting gives a gradual, controlled rise.

Before you buy anything, remove your cooler and check. Always check for a plastic protective film on the cooler’s cold plate, it’s a surprisingly common oversight that completely blocks heat transfer.

3. Dust and Debris Buildup

Dust is slow but relentless. In an average home environment, meaningful buildup accumulates in 6–12 months. Heavy pet ownership or carpeted floors speed that up considerably. Dust-clogged heatsink fins can add 15–25°C to your CPU temps by preventing airflow through the cooler stack.

Check everywhere: heatsink fins (use a flashlight, it’s often completely blocked), AIO radiator fins, case intake filters, and fan blades. Clean every 3–6 months. Tools you need: a can of compressed air and a soft brush. Hold fans stationary while blasting them, letting them spin from compressed air can damage the bearings.

4. Poor Case Airflow

Even a great cooler can’t dissipate heat if the hot air has nowhere to go. The goal is moving cool air in and hot air out in a consistent direction. Front and bottom as intake, rear and top as exhaust is the standard configuration that works. Mixing that up creates turbulence and hot spots.

Case placement matters more than most people realize. The system needs at least 6 inches of clearance from walls. Sitting directly on carpet restricts bottom intake fans entirely. For laptops specifically, blocking the bottom vents even partially can push CPU temps up 10–20°C, always use a hard, flat surface or a dedicated stand.

5. Inadequate Cooler for the CPU’s TDP

Intel’s included stock cooler is rated for roughly 65W TDP. It handles a locked Core i5 fine. Put it on a Core i9-14900K that can draw 250W under Cinebench and it will fail spectacularly. This is the classic “overheating for no reason” situation, the reason is that the cooler was never matched to the CPU’s actual power draw.

If you recently upgraded your CPU without upgrading your cooler, this is almost certainly your problem. A cooler rated for 65W cannot keep a 125W+ chip in its safe thermal range under sustained load.

6. AIO Cooler Problems, Pump Failure, Air Bubbles, or Wrong Radiator Orientation

This one trips people up because they assume “I have an AIO, so cooling can’t be the issue.” Wrong. AIOs have failure modes that air coolers don’t.

Pump failure is the most serious. A dead or dying pump means liquid isn’t circulating, the CPU heats up within seconds of load, often hitting 90°C+ in 10–15 seconds. To diagnose it, open HWiNFO64 and look for the pump RPM sensor in your AIO’s readings. A reading of 0 RPM or consistently below 1,000 RPM means the pump isn’t working properly. AIO pumps typically last 5–7 years before degradation becomes noticeable.

Air bubbles in the loop reduce cooling efficiency and cause a gurgling sound when the pump runs. You can sometimes resolve this by slowly tilting and rotating the case to move the bubble toward the radiator.

Radiator orientation is underappreciated. Top-mounting the radiator as an exhaust is optimal, hot air exits the case directly. Front-mounted radiators pulling warm air from inside the case back through the system can actually increase CPU temps.

7. Software and OS-Level Causes

If you’re asking why your CPU is overheating in Windows 11 specifically, this is where to look. Windows 11 ships with the “Best Performance” power plan enabled in some configurations. That plan removes limits on sustained boost clocks, meaning the CPU runs hotter and faster than it would under the “Balanced” plan, even at idle.

Background processes are another underdiagnosed cause. Open Task Manager and sort by CPU column. Windows Update, Search Indexer, and antivirus scans can peg a CPU at 30–50% for extended periods without obvious notification. Cryptominer malware is worth checking for, it runs CPU utilization near 100% silently in the background. Run a full malware scan if you can’t identify the CPU-hogging process.

Outdated chipset drivers and BIOS firmware can also cause irregular power delivery behavior that manifests as unexpected thermal spikes.

8. Overclocking or Aggressive Power Limits Without Adequate Cooling

Intel’s Adaptive Boost and Thermal Velocity Boost features on i9-series chips can push power draw to 250W+ under sustained load. Many motherboard manufacturers, especially in the enthusiast ASUS and MSI segments, ship their boards with power limit unlocked by default, meaning the CPU ignores Intel’s official TDP spec and draws as much as the VRM will deliver. Your CPU isn’t overclocked in the traditional sense, but it’s running outside its rated power envelope.

AMD’s Precision Boost Overdrive (PBO) does something similar on Ryzen, it allows the CPU to boost higher and longer than stock settings permit, pushing thermals toward TjMax. If you enabled PBO in the BIOS and your temps jumped, that’s the cause.

Why Is My CPU Overheating All of a Sudden?

Sudden-onset overheating, where nothing obviously changed, usually has one of a handful of causes. Here’s a quick diagnostic flow:

• Did it start right after a Windows update? Check your power plan in Settings. A major update can reset it to “Best Performance.” Roll back the update if needed or just switch the power plan back to “Balanced.”
• Is the system 3+ years old? Repaste first. Thermal paste degradation follows a curve, it doesn’t fail gradually, it often seems fine until it suddenly isn’t.
• Do you have an AIO? Check pump RPMs in HWiNFO64. Pump failure often happens suddenly rather than gradually.
• Did you recently move or transport the PC? Physical movement can loosen push-pin mounts or shift an AIO pump head. Check cooler seating.
• Did a new game or app trigger this? Some titles push sustained CPU loads that expose marginal cooling that was “good enough” for lighter workloads. The cooling wasn’t adequate before, you just weren’t stressing it enough to notice.
• Is there a new dust accumulation threshold? Buildup is gradual, but the performance impact can feel sudden once it crosses a critical point.

Not a hardware issue in every case. Check the software side before disassembling anything.

How to Fix an Overheating CPU, Step-by-Step

Fix 1, Clean Your System (Start Here)

This costs nothing and should always be step one. Power off completely and unplug. Ground yourself by touching the case metal before handling components. Hold each fan stationary, don’t let compressed air spin the fans freely, as this can damage bearings. Blast the heatsink fins, radiator fins (if AIO), intake filters, and fan blades from multiple angles.

Expected temperature improvement: 5–20°C depending on how clogged things were. Takes about 20–30 minutes done properly.

Fix 2, Reapply Thermal Paste

Remove the old compound with 90%+ isopropyl alcohol and a lint-free cloth or coffee filter. Apply 0.3–0.5mL (pea-size) centered on the IHS. Don’t spread it manually. The cooler pressure distributes it correctly.

Good options that are genuinely worth using: Thermal Grizzly Kryonaut for high-performance builds (our Thermal Grizzly Kryonaut review covers exactly what temperature gains you can expect), Noctua NT-H1 for everyday builds, Arctic MX-6 for a budget-friendly choice that still performs well.

Expected improvement after a proper repaste: 5–15°C. Worth doing before spending anything on new hardware.

Fix 3, Reseat or Upgrade Your CPU Cooler

For Intel push-pin coolers: remove the cooler completely, rotate each push-pin counter-clockwise to reset it, then re-insert with firm, even pressure on all four corners simultaneously. For AMD systems: follow the X-pattern torque sequence, applying 0.6 N⋅m per screw.

If temps consistently exceed 90°C on a non-overclocked CPU despite clean mounting and fresh paste, you need a better cooler. General guidance:

• Up to 65W TDP: Cooler Master Hyper 212 (~$35) handles this easily.
• 65–125W TDP: Noctua NH-U12S Redux (~$50) or be quiet! Dark Rock 4.
• 125W+ TDP: Noctua NH-D15, or a 240mm+ AIO. No exceptions for sustained workloads.

Fix 4, Improve Case Airflow

Rearrange fans to create a consistent front-to-rear or bottom-to-top airflow path. Tuck cables out of the airflow path, a rat’s nest of loose cables in front of the CPU cooler genuinely affects temps. If you’re on a small or restricted case and temps don’t drop 5°C+ after cleaning and fan optimization, the case itself may be the limiting factor.

Fix 5, Fix AIO-Specific Issues

If pump RPMs look normal but temps are still high, try burping air bubbles: with the system off, slowly tilt and rotate the case in different orientations while listening for the gurgle to change or stop. Remount the radiator at the top as an exhaust if it’s currently front-mounted.

If the pump RPMs read 0 or sub-1,000 in HWiNFO64, the AIO needs replacing. If it’s over 6 years old, this is expected, consider switching to a quality air cooler at that point, since AIO lifespans are finite.

Fix 6, Adjust Power Limits in BIOS

For Intel systems, set PL1 equal to your CPU’s rated TDP and PL2 equal to TDP as well in the BIOS power limit settings. This stops the motherboard from letting the CPU boost past its thermal design indefinitely. Intel’s ARK database lists official TDP values for every chip.

For AMD systems, disable PBO or set the PBO scalar to 1x if your cooling is marginal. In Windows, switch from “Best Performance” to “Balanced” in power settings. This alone can drop sustained temps by 10–15°C on systems that have never had power plans configured.

Fix 7, Check for Software Causes

Open Task Manager, click the CPU column header to sort by usage, and watch for anything consistently consuming 20%+ that shouldn’t be. Windows Update Service, Search Indexer, and Antimalware Service Executable are common culprits. Run a full scan with Windows Defender or Malwarebytes to rule out cryptominer malware, this is a legitimately underdiagnosed cause of unexplained high CPU load and heat.

CPU Overheating on a Laptop? Here’s What’s Different

Laptop cooling is fundamentally more limited than desktop cooling. The thermal solution is thinner, the fans are smaller, and airflow is constrained by chassis design. Expecting desktop-level thermal performance from a laptop is unrealistic.

That said, there are real fixes available:

• Clean the vents: Laptop vents clog faster than desktop cases. Use compressed air to clear bottom and side vents every 3–6 months.
• Elevate on a stand: Bottom vents need clearance. A laptop cooling stand improves airflow and typically reduces temps by 3–10°C on average.
• Repaste (advanced): Laptop thermal paste degrades like desktop paste. This is worth doing on a 3+ year old laptop but requires disassembly, check your model’s iFixit guide first.
• Undervolting: Intel XTU or Throttlestop can reduce CPU voltage without reducing performance, lowering heat output. Works on most Intel mobile chips.

Warning signs a laptop cooler is actually failing vs. normal laptop thermals: temps that hit 100°C+ under any load, sudden shutdowns within minutes of starting a game, or a fan that makes grinding or rattling sounds.

When to Replace Rather Than Fix

Sometimes the fix is a new part. Here’s when to stop troubleshooting and start shopping:

• AIO older than 6–7 years with confirmed pump failure: Replace it. A high-quality air cooler is often a better choice at this stage than another AIO.
• Heatsink fins physically bent or damaged: Airflow through a damaged heatsink is severely compromised. Replace the cooler.
• Persistent WHEA errors after cooling is resolved: Test the CPU in another board. Repeated thermal stress can degrade the processor itself over time.
• Budget stock cooler on a 125W+ chip: There’s no fix that doesn’t involve a better cooler. Cleaning and repasting buys time, nothing more.

Frequently Asked Questions

Is 120°F hot for a CPU?

120°F equals approximately 49°C. That’s completely normal for a CPU under light-to-moderate load like browsing or streaming. Idle temps of 30–45°C and gaming temps up to 85°C are acceptable for most modern processors. 120°F gives you nothing to worry about.

Why is my CPU overheating all of a sudden when nothing changed?

Sudden overheating most commonly points to one of three things: thermal paste that has finally dried out after 3–5 years of use, an AIO pump that has failed (often without warning), or a Windows update that reset your power plan to “Best Performance.” If your system is more than three years old, start by reapplying thermal paste before anything else.

Why is my CPU overheating with an AIO cooler installed?

An AIO doesn’t guarantee good cooling. Check pump RPMs in HWiNFO64, a reading of 0 RPM or under 1,000 RPM confirms pump failure. If RPMs are normal, check radiator orientation (top-mount exhaust is optimal), inspect for air bubbles causing reduced flow, and verify the cold plate is properly seated with good thermal paste contact. AIOs have a 5–7 year average lifespan before performance degrades.

Why is my CPU running hot at idle?

Idle temps above 50–60°C typically mean one of three things: poor mounting contact between the cooler and CPU, dried or incorrectly applied thermal paste, or a background process pegging the CPU at significant load. Open Task Manager and sort by CPU usage before assuming you have a hardware problem. Search Indexer and Windows Update are frequent culprits that can make a CPU look “idle” in overall usage while actually running hot.

Can CPU overheating cause permanent damage?

Modern CPUs use thermal throttling and automatic emergency shutdown to protect themselves before permanent damage occurs. However, sustained operation near TjMax degrades the processor over time through electromigration and thermal stress cycling. Consistent temps above 95°C under normal workloads aren’t immediately destructive, but they shorten the chip’s lifespan meaningfully. Fix it.

The Bottom Line

Most CPU overheating problems come down to one of three things: a thermal paste issue, a mounting problem, or a mismatch between cooler capacity and CPU power draw. Start with the free fixes, clean the system, check Task Manager, verify power plan settings. If that doesn’t move the needle, repaste. If temps are still high after a fresh paste job, check cooler mounting and then evaluate whether your cooler is actually rated for your CPU’s TDP.

The good news: most overheating situations are fixable for under $30 with thermal paste and a can of compressed air. If you’ve worked through every step here and temps are still dangerously high under normal loads, it’s time to look at a proper cooler upgrade matched to your chip’s actual power requirements. The investment is worth it, running a CPU hot constantly isn’t just a performance problem. It’s a longevity problem.