Quick answer: Event ID 37 from Kernel-Processor-Power is a warning that the PC’s firmware is holding the processor below its full speed, usually because of a heat, power or current limit. Intel describes it as part of normal power management, not a hardware fault, so occasional short entries can be ignored. Look into it only if the PC feels slow, runs hot, or the entries report long periods: check the power mode, cooling and charger, then reset or update the BIOS.
In this guide
What event 37 means
You’ll find it in Event Viewer under Windows Logs > System, with the source Kernel-Processor-Power and the level Warning. The text follows this pattern:
The speed of processor 3 in group 0 is being limited by system firmware. The processor has been in this reduced performance state for 71 seconds since the last report.
The processor number, group and seconds change from entry to entry. Microsoft’s support article on the event explains that it is logged when the hardware platform decides Windows can’t use part of the frequency range the processor supports, and that it tells you the core can’t run at full speed. In other words, Windows asked for more speed and the firmware (the BIOS or UEFI) said no.
Intel’s support pages give the reasons the firmware does that: a thermal limit, a power limit or a current limit. Intel also says the message does not indicate a hardware issue. It’s the same mechanism that lets a laptop slow down to stay cool or save battery.
When you can ignore it
| What you see | What it suggests |
|---|---|
| A few entries, each reporting a few seconds, and the PC feels normal | Routine power management; nothing to fix |
| Many entries during games or heavy work, with long durations | The CPU is hitting a heat or power limit; work through the steps |
| Entries alongside a slow PC at light load | A power setting, charger or firmware limit is holding it down |
| Entries near unexpected shutdowns | Check cooling first; the shutdown itself is logged by a different event |
The seconds field is the most useful clue. It counts how long the processor stayed limited since the previous report, so a steady run of large numbers means the limit is lasting, not just flickering.
Step 1: Check whether your CPU is really slowed
Open Task Manager with Ctrl + Shift + Esc, go to Performance and select CPU. Compare the live Speed with the Base speed shown below the graph while you run something demanding. If the speed climbs to or above the base speed under load, the limit the event reports isn’t costing you much. If it stays well below base speed even when the CPU is busy, keep going. Our guide to a CPU not running at full speed covers that symptom in more depth.
Step 2: Check the power mode and plan
On Windows 11, open Settings > System > Power & battery and look at Power mode. Microsoft lists three choices: Best power efficiency, Balanced and Best performance, and on a laptop you can set one for battery and one for when it’s plugged in. Best power efficiency deliberately trades speed for battery life. Microsoft notes that if you can’t change the mode, a custom power plan may be active; selecting Balanced in Control Panel > Power Options brings the slider back.
Also check the processor limits inside the plan. In Power Options, click Change plan settings next to the active plan, then Change advanced power settings, and expand Processor power management. Maximum processor state should be 100%. If it has been lowered, Windows itself is capping the CPU.
Intel’s troubleshooting page for its NUC mini PCs goes further and suggests setting both the minimum and maximum processor state to 100%. That keeps the CPU at full speed even when idle, which means more heat, more fan noise and shorter battery life, so treat it as a test, not a permanent setting. To see what your plan and drivers are doing, open Command Prompt as administrator and run:
powercfg /energyMicrosoft says this analyses the system for common energy-efficiency problems for 60 seconds and saves an HTML report in the current folder. Run it with no programs open, then look for warnings about processor power management.
Step 3: Rule out heat and power supply
Intel’s support staff and PC makers’ community moderators often point to heat first. Clean dust out of the vents and fans, make sure a laptop isn’t sitting on a bed or cushion, and on a desktop check that the CPU cooler is seated properly and its fan spins. If temperatures stay very high under load, the firmware will keep throttling whatever Windows asks for.
Power can do the same. If you’re on a laptop, test with the original charger plugged in. Some users report the entries stopped after they replaced a lower-wattage or failing adapter; this isn’t documented by Microsoft or Intel, but it’s quick to rule out. If high CPU use itself is pushing temperatures up, System interrupts high CPU may help find the cause.
Step 4: Reset or update the BIOS
Both Microsoft and Intel point at the firmware. Microsoft’s resolution is to check for any power-capping policy from the hardware maker and for system options that hand power control to the firmware. Intel’s advice is to load the BIOS defaults, since some motherboard defaults are aggressive about power and thermal limits, and to update to the latest BIOS from the PC or motherboard maker.
- Restart and enter the BIOS or UEFI setup. Our explainer on Secure Boot and UEFI shows how to reach it from Windows.
- Choose the option to load default or optimized settings, save and restart.
- If the warnings continue, check the maker’s support page for your exact model and install a newer BIOS using their instructions. Keep a laptop on its charger throughout.
Some guides suggest disabling Intel SpeedStep or uninstalling Intel’s thermal framework driver. Those features exist to manage heat and power, and removing them can make the PC run hotter or drain the battery faster without addressing why the firmware is limiting the CPU. Leave them as a last resort and follow your PC maker’s guidance.
Events 55, 35 and 41
You’ll often see event 55 from the same source at every start-up. It’s an information entry listing each processor’s power management capabilities and is normal. Event 35, an error saying performance power management features are disabled due to a firmware problem, is more serious and is another reason to update the BIOS. Event 41 comes from a different source, Kernel-Power, and records an unexpected shutdown. Event 37 doesn’t cause it, but the two sometimes appear together when a PC overheats, so if you see both, focus on cooling first.
Related guides
- CPU not running at full speed in Windows 11/10: how to fix it
- System interrupts high CPU in Windows 11 and 10: what it is and how to fix it
- How to open Command Prompt as administrator in Windows 11 and 10
- Secure Boot state: how to check it and what On, Off and Unsupported mean
Leave a Reply