Jump to content

[Benchmarking Tip] Testing your CPU coolers performance UNIX style

for i in $(seq $(getconf _NPROCESSORS_ONLN)); do yes > /dev/null & done

What this little command/script does is cause your cpu to endlessly spam "yes" to STDOUT (your terminal) peaking your cpu at 100% usage on every core (also sidesteps systemd-oomd if your distro uses it)

you can ctrl+c to kill it or use

pkill yes

Bear in mind if your going to compare this to a typical benchmark you also want to track your ambient air temperature as well depending on your rooms environment.

 

This kind of script is actually pretty common practice in the systems engineering field for testing a new server deployment. Use it wisely.

 

To get a hold of your cpu clock speed in a nice log file you can add this script

while true
  do
    cat /proc/cpuinfo | grep MHz >> stress-bench 
  sleep 1
done

and for tracking temperature (this one requires modification btw since your motherboard cpu combination can't vary the device) requires lm_sensors

while true
  do
    sensors | grep insertrelevantchiphere >> stress-temps 
  sleep 1
done

 

Link to comment
Share on other sites

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×