A gaming PC works its processor (CPU) and graphics card (GPU) harder than almost anything else you can run. When the cooling can't keep up, you'll notice it in a few typical ways: fans that ramp up to a roar, frame rates that are fine at first and then drop after a while, stutters in the middle of a match, or a PC that switches itself off or restarts during a game.
Here's how to tell whether heat is really the cause, what's normal for a gaming PC under load, and the fixes that bring temperatures back down.
Why heat costs you frame rate
Modern processors and graphics cards protect themselves. When they reach their temperature limit they lower their clock speed, which is called thermal throttling. The PC keeps running, but slower, and in a game that shows up as falling or uneven frame rates. If temperatures keep climbing past that point, the hardware shuts the PC down to avoid damage.
That's why an overheating gaming PC often feels fine for the first ten or twenty minutes. The heat builds up until the parts hit their limit and start pulling back.
How to check your temperatures under load
Checking temperatures at the desktop tells you very little. You need to see them while the PC is working hard:
- Install a monitoring tool such as HWiNFO, or use the overlay in your graphics card software.
- Play a demanding game for at least 20 to 30 minutes, the kind of session where you normally see problems.
- Note the highest CPU temperature, GPU temperature and, if shown, the GPU hot spot. Also watch whether clock speeds drop as the temperatures rise.
What temperatures are normal for a gaming PC?
There's no single number, because every chip has its own limit. Many current desktop processors are designed to run up to around 95 to 100°C before they throttle, and some will sit near that under heavy load by design. Graphics cards have their own limits, and the hot spot reading normally runs higher than the main GPU temperature. Check the maker's specification for your exact CPU and card.
More useful than any single number are these warning signs:
- Temperatures that sit right at the limit while clock speeds drop.
- Temperatures that are much higher than when the PC was new, for the same game and settings.
- A big gap between the GPU temperature and its hot spot that has grown over time, which can point to dried-out thermal paste on the card.
- The PC switching off under load but never at the desktop.
The usual causes
Poor airflow through the case
A gaming case needs cool air coming in and warm air going out. Typically that means front or bottom fans pulling air in, and rear or top fans pushing it out. Fans facing the wrong way, a solid front panel with few intake fans, cables bunched in front of the fans, or a case pushed into a cupboard or against a wall all trap hot air around the graphics card.
Dried-out thermal paste
Thermal paste carries heat from the chip to its cooler. After a few years of heavy gaming it can dry out and stop doing that job well, on both the CPU and the graphics card. Temperatures creep up even though nothing else has changed.
A cooler that's not up to the job
A basic cooler that suited the original processor may struggle after a CPU upgrade. All-in-one liquid coolers can also lose performance as they age, or have a pump that's slowing down or has failed.
Fan settings and failing fans
Fan curves set too quiet in the BIOS or software, or a fan with worn bearings that no longer spins at full speed, can leave the cooling short just when the game needs it.
Shuts off with no warning? A PC that cuts out instantly under gaming load can also be a power supply problem rather than heat. If temperatures look normal right up to the shutdown, read our guide to power supplies and graphics cards.
Fixes, from easiest to most involved
- Give the case room to breathe. Out of cupboards, a little away from the wall, and off carpet if the power supply draws air from underneath.
- Check fan direction and fan curves. Make sure intake and exhaust fans work together, and that fan curves ramp up properly under load.
- Tidy the cables. Cables routed behind the motherboard tray keep the air path to the graphics card clear.
- Clean and repaste. Fresh thermal paste on the CPU and graphics card, and new thermal pads where needed, often brings temperatures back to where they were when the PC was new.
- Upgrade the cooling. A better CPU cooler, an extra intake fan or a case with a more open front panel can make a real difference for a hot-running build.
Opening a graphics card to repaste it is fiddly and, depending on the card, may affect its warranty, so check before you start or leave that part to us.
Let us sort out the heat
Overheating is one of the most common gaming PC faults we see in our Midland workshop. A clean and repaste is one of our standard desktop jobs, and we look at airflow, fan curves and the cooler at the same time so the fix lasts. We stress-test the hardware afterwards to make sure it holds up under load. If heat turns out not to be the problem, our bottleneck checker can point at the part that's holding your frame rate back, and our gaming PC repairs page covers the other faults we fix.