CPU Nice Value & Share Calculator
Convert a Linux nice value into its scheduler weight and see the CPU-time share it gets against competing tasks.
Nice value to kernel weight
The scheduler turns each nice value into a weight; CPU time is divided in proportion to those weights. This is the full table for nice −20 to 19.
| Nice | Weight | vs nice 0 | Share* |
|---|
*The share column assumes two tasks on one CPU: the listed nice value and one nice 0 task, both always runnable. Nice 0 has weight 1024, each step of one nice value changes the weight by about 1.25 times, and the table is the kernel's prio_to_weight set used by the fair scheduler (CFS and its EEVDF successor). Real CPU time also depends on how often tasks sleep, how many CPUs are available, cgroup cpu.weight and cpu.max limits, and scheduling policies such as batch or idle, so read the result as the proportional-share target rather than a guarantee. Everything is calculated in your browser; nothing is sent or stored.