PC Overheating and Thermal Throttling — A Real Diagnostic Order
Frame rate that degrades after ten minutes is thermal, not a driver problem. Here's how to identify which component and what actually fixes it.
The short answer
Thermal throttling shows as a frame rate that is fine for the first five to ten minutes and then degrades, distinguishing it from other performance problems which are present from the start. Identify the component by logging CPU and GPU temperatures separately during a session; CPU throttling typically appears above 95 degrees and GPU throttling above 83 degrees on the hotspot sensor. Case airflow and dust are the usual causes, not a failing component.
Verified and last updated August 12, 2026.
- Time needed
- 40 min
- Difficulty
- Intermediate
- Read time
- 4 min
There is one symptom that identifies a thermal problem with near certainty, and it is a timing pattern rather than a temperature reading.
The first five to ten minutes are fine. Then it gets worse.
Software and driver problems do not behave like that — they are present from the moment you load in. If your session degrades over time and recovers after the machine sits idle, you have a heat problem.
Temperature numbers are less useful than trends
Modern CPUs are designed to sit near 95 °C and will throttle rather than damage themselves. Hitting that is not an emergency. What matters is whether clock speeds are falling — a chip at 90 °C holding full clocks is fine, and a chip at 85 °C dropping clocks is not.
Step 1: Confirm the time signature
Step 2: Identify which component is throttling
Log both during a fifteen-minute session. A thirty-second test proves nothing because heat takes time to build.
Log CPU package temperature and clock speed 15 min
Throttling typically begins around 95 °C on modern processors. The tell is the clock speed dropping below its sustained boost figure while load is still high.
Log GPU core and hotspot temperature and clock 15 min
Core throttling generally starts around 83 °C. Hotspot readings run considerably higher and that is normal — a very large gap between core and hotspot, however, suggests a mounting or thermal paste problem.
Note which one falls first —
This tells you where to spend effort. In small cases both may throttle together because they are heating each other, in which case case airflow is the answer rather than either component's cooler.
Step 3: Airflow and dust first
This is the cause in most cases and it is free.
Dust in the fins. Not on the surfaces — in the fins of the CPU cooler, GPU heatsink and any radiator. That is where it blocks airflow most effectively and where it is least visible. A machine that ran well two years ago and throttles now has almost certainly not developed a fault; it has developed a blanket in the heatsink.
Intake versus exhaust balance. Slightly more intake than exhaust creates positive pressure, which reduces dust entering through unfiltered gaps. Very unbalanced setups in either direction cool poorly.
Cable management. A bundle of cables across the intake path is a real and very common restriction.
Front panel restriction. Cases with solid glass or heavily restricted front panels starve intake regardless of how many fans you fit. Check whether air can actually get in.
Where the machine sits. A tower in a closed desk cubby recirculates its own exhaust. This is more common than you would think and it is the easiest fix on this list.
Step 4: Thermal paste on older machines
Thermal interface material dries out over years. On a machine three or more years old with clean fins and good airflow that still throttles, repasting is the next step and it frequently produces a large improvement.
On graphics cards, the thermal pads on memory and power delivery components degrade as well, and a card that throttles with a normal core temperature but a very high hotspot reading is often a pad problem rather than a paste problem.
Laptops are different
Laptop throttling is largely by design — the cooling and power budgets are small and the components share a heat path. Elevating the machine, cleaning the vents and undervolting are the meaningful interventions. A laptop GPU running well below its rated clock under sustained load is normal, not broken.
What does not fix it
What works
What doesn't
If the machine crashes or shuts down entirely rather than slowing, that is a different diagnostic path — see PC game crash diagnosis.
Key takeaways
Frequently asked questions
What temperature is too hot for gaming?
Modern CPUs are designed to run near 95 degrees under load and will throttle rather than damage themselves, so hitting that number is not an emergency — it just means you are leaving performance on the table. GPUs are comfortable to around 83 degrees on the core, with hotspot readings running considerably higher and still being within specification. The concerning readings are sustained maximums with degrading clocks, not any specific number in isolation.
How do I know it is thermal and not a driver problem?
The time signature. Thermal problems are absent for the first several minutes and then appear as the machine heats, and they clear after the machine has been idle. Driver and software problems are present from the moment you load in. If your first ten minutes are consistently fine and the eleventh is not, you have a thermal problem and no driver reinstall will change it.
Is dust really that big a factor?
Yes, and it is the single most common cause. Dust accumulates in the fins of coolers and radiators, where it blocks the airflow path most effectively. A machine that ran fine two years ago and now throttles has almost certainly not developed a hardware fault — it has developed a blanket of dust in the heatsink fins that a two-minute clean removes.
Do I need liquid cooling?
Rarely. A good air cooler handles the vast majority of consumer processors, and case airflow matters more than cooler type in most builds. Liquid cooling helps with very high-power processors and with moving heat out of a constrained space, but a well-ventilated case with a competent air cooler outperforms a poorly ventilated case with a radiator. Fix airflow before considering it.
Why do laptops throttle so much more?
Because the cooling and power budgets are physically small and the components sit close together, so CPU heat raises GPU temperature and vice versa. A laptop GPU running twenty per cent below its rated clock under sustained load is normal behaviour rather than a fault. Elevating the machine for airflow, cleaning the vents and undervolting are the meaningful interventions.
About the author
PC Hardware & Performance Editor
Marcus has been building, breaking and repairing gaming PCs since 2011, first as a weekend favour for friends and then for eight years as a bench technician at an independent repair shop in Portland, where roughly a third of the job was diagnosing machines that "ran fine yesterday".
Found something wrong? Games patch and fixes go stale. If a step here no longer works, tell us at hello@spyld.com and we will retest it. Read how we test and our editorial policy.