How reliable is the program lm-sensors in Linux? When I run watch sensors, I see large skips of temperature, as high as 10 degrees Celsius within the refresh interval of 2 seconds. The two readings below show this, with two screenshots taken 2 seconds apart, with Core 1 jumping 8 degrees down between those.
Reading 1:
Every 2.0s: sensors Thu Nov 19 14:02:41 2020
iwlwifi_1-virtual-0
Adapter: Virtual device
temp1: +66.0°C
acpitz-virtual-0
Adapter: Virtual device
temp1: +59.0°C
coretemp-isa-0000
Adapter: ISA adapter
Package id 0: +67.0°C (high = +100.0°C, crit = +100.0°C)
Core 0: +61.0°C (high = +100.0°C, crit = +100.0°C)
Core 1: +67.0°C (high = +100.0°C, crit = +100.0°C)
Core 2: +60.0°C (high = +100.0°C, crit = +100.0°C)
Core 3: +59.0°C (high = +100.0°C, crit = +100.0°C)
pch_skylake-virtual-0
Adapter: Virtual device
temp1: +55.5°C
Reading 2, taken 2 seconds later:
Every 2.0s: sensors Thu Nov 19 14:02:43 2020
iwlwifi_1-virtual-0
Adapter: Virtual device
temp1: +65.0°C
acpitz-virtual-0
Adapter: Virtual device
temp1: +61.0°C
coretemp-isa-0000
Adapter: ISA adapter
Package id 0: +59.0°C (high = +100.0°C, crit = +100.0°C)
Core 0: +59.0°C (high = +100.0°C, crit = +100.0°C)
Core 1: +59.0°C (high = +100.0°C, crit = +100.0°C)
Core 2: +58.0°C (high = +100.0°C, crit = +100.0°C)
Core 3: +57.0°C (high = +100.0°C, crit = +100.0°C)
pch_skylake-virtual-0
Adapter: Virtual device
temp1: +55.5°C
答案1
It's totally expected for new AMD and Intel CPUs due to their heat density due to very small transistors where billions of them are crammed into very small space.
If you don't want to see such high temperatures or temperature fluctuations, disable turbo boost:
echo 0 | sudo tee /sys/devices/system/cpu/cpufreq/boost
It might be different for Intel CPUs, please consult with https://askubuntu.com/a/620114