How to Fix Thermal Throttling on Your Computer
Your computer runs fine for light tasks but slows to a crawl under load — gaming, video editing, compiling code, or even running multiple browser tabs during a video call. The CPU or GPU hits a temperature limit and deliberately reduces its clock speed to avoid damage. This is thermal throttling, and it is the most common cause of performance that degrades over time in both desktops and laptops. A computer that was fast when new but now stutters under the same workload is almost certainly throttling.
What thermal throttling actually is
Every CPU and GPU has a maximum safe operating temperature set by the manufacturer (called Tjunction max for Intel CPUs, Tmax for AMD). Modern processors typically have limits between 95°C and 105°C. When the chip reaches this temperature, the processor's built-in protection reduces its clock speed — sometimes dramatically — to lower heat output. The system stays functional but performance drops, often by 30 to 50 percent or more.
Throttling is a safety mechanism, not a fault. Without it, the processor would overheat and sustain permanent damage. But when throttling happens during normal use, it means the cooling system is not removing heat fast enough to let the processor run at its intended speed.
How to check if your computer is throttling
Before fixing anything, confirm that thermal throttling is actually the problem:
Check temperatures under load
- Download HWiNFO (free, widely used by technicians and overclockers). Open it in Sensors-only mode
- Find your CPU package temperature and GPU temperature in the sensor list
- Run the task that causes slowdowns — play a game, start a render, open the heavy workload
- Watch the temperatures. If the CPU hits 95-100°C or the GPU exceeds 90°C, thermal throttling is occurring or imminent
HWiNFO also shows a Thermal Throttling flag for most processors — if it reads Yes during your workload, you have confirmed the problem.
What healthy temperatures look like
- Desktop CPU under load: 60-80°C with a stock cooler, 50-70°C with an aftermarket tower cooler, 45-65°C with a 240mm or larger AIO liquid cooler
- Laptop CPU under load: 75-90°C is typical for thin laptops, 65-80°C for thicker gaming laptops with better cooling
- Desktop GPU under load: 65-80°C for most models, 80-90°C for compact or blower-style cards
- Laptop GPU under load: 75-87°C depending on the laptop's thermal design
Temperatures above these ranges under sustained load indicate a cooling problem that needs attention.
Fix 1: Clean dust from the cooling system
Dust is the single most common cause of thermal throttling that develops over time. Dust accumulates on heatsink fins, fan blades, and intake vents, progressively reducing airflow:
- Desktops: Open the side panel. Use compressed air (canned air or an electric air duster) to blow dust out of the CPU heatsink, GPU heatsink, case fans, and all intake/exhaust filters. Hold fan blades still while blowing air through them to prevent the bearings from spinning beyond their rated speed. Clean the intake dust filters if your case has them
- Laptops: Use compressed air to blow into the exhaust vents (the hot air output, usually on the side or back) and intake vents (usually on the bottom). Short controlled bursts prevent moisture from the compressed air can from entering the system. For a thorough cleaning, remove the bottom panel (most laptops have Phillips screws holding it on) and blow dust directly off the fan and heatsink assembly
A computer that has not been cleaned in 1-2 years can have significant dust buildup. Cleaning alone often drops temperatures by 10-20°C and eliminates throttling entirely.
Fix 2: Improve case airflow (desktops)
Even with clean components, poor airflow causes heat to recirculate inside the case:
- Check fan orientation. Fans should create a front-to-back or bottom-to-top airflow path. Front and bottom fans intake, rear and top fans exhaust. Randomly oriented fans cause hot air to pool around components
- Add fans if slots are empty. A case with only one rear exhaust fan creates negative pressure. Adding one or two front intake fans dramatically improves temperatures
- Cable management. Route cables behind the motherboard tray to keep the airflow path clear between intake and exhaust
- Remove side panel as a test. If temperatures drop more than 5°C with the panel off, the case has an airflow problem that needs addressing
Fix 3: Replace thermal paste
Thermal paste (thermal compound or thermal interface material) fills the microscopic gaps between the CPU or GPU die and the heatsink. Over time — typically 3 to 5 years — thermal paste dries out, hardens, and loses its ability to transfer heat efficiently:
CPU thermal paste replacement
- Remove the CPU cooler. For tower coolers, this means unscrewing the mounting bracket. For stock coolers, twist and pull (AMD) or unlock the lever (some Intel). For AIOs, unscrew the cold plate from the CPU mounting bracket
- Clean the old paste from both the CPU and the cooler's contact surface using isopropyl alcohol (90% or higher) and a lint-free cloth or coffee filter. The surfaces should be clean and shiny
- Apply new thermal paste to the center of the CPU — a pea-sized dot for most desktop CPUs, a slightly larger dot or thin X pattern for larger chips (AMD Ryzen 7000 series, Intel 13th/14th gen). Do not spread it manually; the mounting pressure of the cooler spreads it evenly
- Remount the cooler. Tighten the mounting screws in a cross pattern (diagonal corners alternating) to ensure even pressure
Laptop thermal paste replacement
Laptop thermal paste replacement follows the same cleaning and application process but requires more disassembly — the bottom panel comes off, then the heatsink assembly (which usually covers both the CPU and GPU on a shared heatpipe). This is more involved than a desktop repaste because:
- Multiple mounting points must be removed and reinstalled in the correct order
- Ribbon cables near the heatsink are fragile
- Laptops use thermal pads on VRM components alongside paste — replace degraded pads at the same time
A fresh thermal paste application typically drops CPU temperatures by 5-15°C compared to paste that is 3+ years old.
Fix 4: Adjust power and performance settings
If the cooling system is clean and thermal paste is fresh but throttling still occurs (common in thin laptops with powerful CPUs), reducing the heat output of the processor is the remaining option:
- Windows power plan. Switch from High Performance to Balanced. The High Performance plan keeps the CPU at maximum clock speed even when idle, generating unnecessary heat. Balanced lets the CPU downclock when not needed
- CPU undervolting. Reducing the voltage supplied to the CPU lowers heat output while maintaining the same clock speed. Tools like ThrottleStop (Intel) or Ryzen Master (AMD) allow per-core voltage offsets. A typical undervolt of -50mV to -100mV reduces temperatures by 5-10°C with no performance loss. Go too far and the system becomes unstable (crashes or blue screens) — start small and test with a stress test after each adjustment
- Manufacturer thermal profiles. Many gaming laptops have thermal management software (Lenovo Vantage, ASUS Armoury Crate, MSI Center, Dell Alienware Command Center) with profiles like Performance, Balanced, and Quiet. These adjust fan speed curves, power limits, and clock speeds. The Performance mode runs fans louder but allows higher clocks; Quiet mode reduces power to keep temperatures and noise low
- GPU power limit. In MSI Afterburner or similar tools, reducing the GPU power limit by 10-15 percent often drops temperatures significantly with only a small FPS loss (5-8 percent). The efficiency curve of GPUs means the last 10 percent of power delivers diminishing performance returns
Fix 5: Upgrade the cooler (desktops)
Stock CPU coolers — especially Intel's included cooler — are designed for the base clock speed at stock settings. If you are running sustained workloads, an aftermarket cooler makes a significant difference:
- Tower air coolers ($25-60). A tower cooler like a Thermalright Peerless Assassin or DeepCool AK400 dramatically outperforms stock coolers and mounts without removing the motherboard
- AIO liquid coolers ($60-150). A 240mm or 280mm all-in-one liquid cooler provides the best cooling for high-power CPUs (Intel i7/i9, AMD Ryzen 7/9). The radiator mounts to the case in a fan slot location and the cold plate mounts to the CPU like a standard cooler
An aftermarket cooler is the most effective single upgrade for a desktop that throttles under load with a stock cooler.
When throttling indicates a hardware problem
If temperatures remain high after cleaning, repasting, and verifying airflow:
- Failed fan. A CPU or GPU fan that spins slowly or not at all cannot move heat. Listen for the fan and watch it spin under load. Replace fans that are noisy, grinding, or stopped
- Detached heatsink. If the cooler mounting has loosened (a push-pin popped out, a screw backed out), the heatsink is not making full contact with the CPU and temperatures spike immediately under load
- Degraded thermal pads on VRMs or memory. Laptops use thermal pads on components surrounding the CPU and GPU. If these pads have degraded, those components overheat and the system throttles even if the CPU temperature appears acceptable
- Clogged laptop heatpipe. Rare but possible — if the internal heatpipes have lost their working fluid (due to a manufacturing defect or physical damage), the heatsink cannot transfer heat from the chip to the fins regardless of fan speed or thermal paste condition. The heatsink assembly needs replacement
Thermal throttling diagnosis in Marion
LuxuriousComputers Repair diagnoses thermal throttling on desktops and laptops — temperature monitoring under load, dust cleaning, thermal paste replacement, fan testing, and cooler upgrades. We verify temperatures return to safe operating ranges under the workload that was causing throttling before returning the system.