Booted a PC Without a CPU Cooler? Intel and AMD Thermal Protection, Shutdowns and What to Do Next

The short answer​

With no cooler, the CPU heats up almost immediately. A modern chip should throttle, cutting power and clock speed. If that isn't enough, the system should shut itself down.

Intel's documentation describes this two-stage protection. AMD is blunter: proper cooling is mandatory to avoid damaging the processor, and damage can occur very quickly if it is powered up without a heatsink present.

So the safeguards are real, and the warning is also real. If you boot coolerless once by accident and the machine shuts off, you probably haven't ruined anything. Nobody guarantees that, though, and you shouldn't treat it as a test you can repeat.

What the protections do​

Intel's temperature guidance says a CPU shutting off doesn't mean it is safe to turn the computer on again. BGR makes that point too. Intel's support documentation describes the mechanism in more detail:

• Throttling: when a core passes its throttle temperature, the processor reduces power to hold a safe level. Intel notes that the throttle temperature can vary by processor and BIOS settings.
• Automatic shutdown: if throttling can't hold a safe temperature, the processor shuts down to prevent permanent damage.
• Performance cost: activating thermal control usually means lower frequency and power, so you lose speed.

Intel's own wording is that overheating damage is "unlikely" thanks to these safeguards. That is a statement about design intent. It isn't a promise for every motherboard, BIOS setting or failure scenario.

Throttling is also a warning signal. Intel describes it as a way to protect the chip and to tell the user there is an overheating problem to look into. Later in this piece there is a checklist for what to do when you see it.

How hot is too hot?​

This is where popular advice gets sloppy. BGR says Intel CPUs generally top out around 100–110°C, and that 90°C under load or 80–90°C at idle deserves investigation.

Intel's documentation supports part of that and cautions against the rest:

• Intel says the maximum junction temperature (Tjunction max) varies by product and is usually between 100°C and 110°C. For exact numbers, check the processor's page on Intel's ARK specification site.
• Intel doesn't publish typical operating ranges for each processor, because temperatures depend on system design and workload.
• Hitting maximum temperature under load isn't necessarily a problem. Intel points to Turbo Boost, which lets chips run at very high frequency briefly. Temperatures can climb rapidly and sustained workloads may sit near the limit.
• Intel also says it is hard to tell from temperature alone whether a cooling solution is inadequate, since normal operation can reach the maximum value.

So treat BGR's 80°C and 90°C figures as rough prompts to look closer, not as universal fault thresholds. The right comparison is your exact CPU's published limit, your workload, and your cooler. A reading that is alarming on one chip can be routine on another. That's general industry reasoning built on Intel's own caveats.

To find your CPU's limits on an Intel part:

1. Identify your processor model.
2. Open Intel's ARK product specification page and search the processor number.
3. Open the processor's page and go to Package Specifications.
4. Look for the Max Operating Temperature, Tjunction or Tcase field.

For AMD, the vendor points users to its own product specifications.

A cooler missing from the box is not a cooler you can skip​

Many builders end up in this mess because their CPU came without a cooler. Intel says all Core and Core Ultra desktop processors with K, KF, KS, XE or X in the model number are not shipped with Intel's thermal solution (fan heatsink). Examples Intel lists include the i7-14700K, i5-13600KF and Core Ultra 285K. Many enthusiast-grade AMD Ryzen parts also ship without one.

"Not included" means you have to buy one. It doesn't mean the chip is meant to run bare.

How to choose a cooler​

Intel's baseline guidance for a self-built system with a boxed processor has two parts:

1. Socket compatibility: check that the cooler supports the processor's socket (for example LGA) and your motherboard.
2. Heat rating: look up the Processor Base Power (previously called TDP) on the ARK page. The cooler's rated wattage should be equal to or higher than that figure.

For heavy sustained use such as gaming, Intel says to meet or better exceed that requirement. AMD gives matching advice for processors bought without a cooler: the heatsink's rated heat dissipation should meet or exceed the processor's power consumption. AMD also says a good thermal interface material should be applied in a uniform layer over the processor lid.

If you already booted without a cooler​

Follow this order:

1. Power off and unplug. Don't try again with no cooler. Neither Intel nor AMD publishes a safe duration for running without cooling.
2. Install a compatible cooler properly. Check socket support and the heat rating as above, and follow the cooler maker's installation instructions. Intel points to its boxed-processor installation videos for stock coolers.
3. Apply the right amount of thermal paste. Intel's troubleshooting steps list this explicitly. Too little or badly spread paste is a common cause of poor contact.
4. Check that the fan is connected and spinning. A cooler fan plugged into the wrong header, or not at all, is the classic "I thought I installed it" failure. This also applies to liquid coolers, whose pump and fans need power.
5. Check airflow. Intel's checklist includes system fans and ventilation. Blocked intakes, dust and a crowded case all matter.
6. Boot into BIOS or UEFI first and watch the temperature. This is my own practical suggestion, not Intel or AMD guidance. A healthy cooler should show sensible idle readings soon after boot. If temperatures climb quickly again, shut down and recheck the mount.

What success looks like: the system boots normally, idle temperatures settle instead of climbing steadily, and you see no throttling during ordinary work.

Pre-built desktops and laptops are a different story. Intel says laptop makers set the power and current limits, so throttling there depends on the OEM's design. For a complete system from a vendor, Intel advises contacting the vendor for troubleshooting.

Common mistakes​

• Cooler plastic film left on: many coolers ship with a protective film on the contact plate. Leaving it on is a classic way to build a PC that "has a cooler" but overheats badly. This is general building knowledge rather than something drawn from the vendor pages above.
• Cooler not clamped evenly: both vendors stress correct installation and a secure fit.
• Fan plugged into the wrong header: the cooler is present, but it doesn't spin.
• Treating the shutdown as the end of the problem: the shutdown is the symptom, and the cause is still there until you fix it.

What we can't say​

The reporting doesn't support a few claims you'll see elsewhere:

• An exact time to damage: neither Intel nor AMD provides one. Web articles that quote precise seconds should be treated with suspicion unless they show methodology.
• A guarantee of safety after a shutdown: the protections are described as safeguards, not as a guarantee of damage-free operation in every case.
• One temperature rule for all CPUs: your own chip's specification is the right reference.

The bottom line​

A coolerless boot probably won't kill a modern Intel or AMD processor in one brief, accidental attempt, because throttling and automatic shutdown exist for this kind of failure. AMD's warning that damage can occur very quickly is the better guide to behavior. Don't do it deliberately, don't repeat it after an accident, and fix the cooling before you power the machine on again. A cooler costs far less than a new CPU, and it doesn't leave you wondering how much of your processor's lifespan went in a few seconds.