Hardware Temperature Safety Lookup
Safe max temps for CPU, GPU, SSD and motherboard so you catch throttling early.
Local & private by design
| Idle range | 30–50°C |
|---|---|
| Safe sustained | < 85°C |
| Throttle point | 90–95°C |
| Shutdown point | 105–115°C |
Watch hotspot (Tctl/Tdie), not just core average.
"Is my PC running too hot?"
This is the question behind half the support threads on every hardware forum. The problem is that "hot" is different for every component — a CPU at 85°C is fine, an SSD at 85°C is throttling, and a GPU hotspot at 85°C means you have tons of headroom. Without a reference, you are guessing.
The Hardware Temperature Safety Lookup gives you the reference: idle, safe, throttle and shutdown ranges for CPUs, GPUs, NVMe SSDs, motherboard VRMs and GPU VRAM. Enter your reading and it tells you whether you are in the clear or about to throttle.
How the tool works
Pick a component and the tool shows its temperature bands. The numbers come from vendor thermal protection behavior:
- CPUs throttle around 90–95°C and shut down at 105–115°C. Watch the hotspot (Tctl/Tdie), not the core average.
- GPUs throttle around 83–88°C core, but the hotspot can run 10–20°C hotter and is the real trigger. A hotspot-to-core gap over 25°C means a repaste is overdue.
- NVMe SSDs throttle at 70°C — sustained writes heat them fast, and a heatsink is cheap insurance.
- VRMs throttle the CPU indirectly around 105°C; poor airflow over the I/O area is the usual cause.
Enter your measured reading and the tool flags whether you are safe, getting hot, throttling, or in shutdown territory.
Step-by-step usage
- Pick the component you are worried about.
- Optionally enter your reading from HWiNFO, HWMonitor or your vendor app.
- Read the bands and the verdict on your reading.
- Act on the note — each component has specific guidance (repaste, add airflow, add a heatsink, improve VRM cooling).
Common hardware problems this tool solves
- "My CPU hits 95°C in games" — you are thermal-throttling. Check cooler mounting, repaste, and case airflow. See the CPU fix hub.
- "GPU core is 70°C but hotspot is 95°C" — a 25°C+ gap means poor TIM contact. Repaste the GPU.
- "NVMe slows to a crawl during big file transfers" — it is thermal-throttling at 70°C. Add a heatsink.
- "PC is stable until a long gaming session, then stutters" — VRM or GPU throttling as heat soaks in. Improve sustained cooling, not peak cooling.
Pro tips
- Watch hotspot, not average. A 75°C average with a 100°C hotspot is a problem; a 75°C average with an 80°C hotspot is fine.
- Repaste every 2–3 years on GPUs. Pump-out and dry TIM are the silent killers of high-end cards.
- NVMe heatsinks are cheap and effective. A $5 stick-on heatsink keeps a Gen4 drive under throttle during sustained writes.
- VRM airflow matters more than CPU fan RPM. Point a case fan at the I/O / VRM area if you run heavy all-core loads.
Temperature is a symptom, not a fault. The lookup tells you whether the symptom is normal or a warning — then you fix the cause.