Stats is the app I recommended to people for years. It is free, it is open source, it is maintained, and it does more than most paid monitors. If you are looking for an alternative, it is almost certainly not because of a missing feature.
It is because you looked at Activity Monitor and found it near the top.
The one number that matters here
On the same MacBook Pro, sampled by the same command in the same window, Stats idled at 1.931% CPU. AirStats idled at 0.046%.
That is the whole pitch, so I will not belabour it. If you want the method, the sample counts and the places the result does not flatter me, it is all in the benchmark.
The part worth saying here is that the gap is not because Stats does more work. It is a design difference, and I could only partly prove which one. Stats reads processes by shelling out to a command line tool on a timer, according to its own source, and spawning a process every second costs enormously more than reading the same thing through a syscall. That module was not enabled in the install I measured, so I can tell you the gap is real and I cannot fully tell you why.
Where Stats is better
- It runs on Intel Macs. AirStats does not. Apple silicon only, macOS 14 or later.
- More modules and more widget styles. Stats covers more ground than AirStats does, with more ways to draw each thing.
- Years of releases behind it. AirStats is new, with the bug reports still ahead of it rather than behind it.
- A large community. If you hit something odd in Stats, someone has already filed it.
Where AirStats is better
Idle cost. Covered above. It also shows up as memory, 11 MB against 145 MB on a fresh launch, and as energy impact, where the gap tracked the CPU result almost exactly.
One process, nothing installed around it. No helper, no launch agent. Quitting it leaves nothing running.
Sampling that follows the screen. A collector nothing is displaying is not sampled slowly, it is not sampled at all, and it releases what it was holding. That design is most of the difference and it is the part I would steal if I were maintaining any monitor.
Both are free and open source, so just try it
This is not a purchase decision. Both apps are MIT licensed and cost nothing, so the only thing you are spending is two minutes.
Install AirStats, put the same metrics in your menu bar that you have in Stats now, and run both for an afternoon. Then check them yourself:
top -l 2 -s 1 -stats command,cpu,mem -o cpu | grep -iE 'airstats|stats'Read the second sample, not the first. The first sample of any top run reports CPU cumulative since launch rather than over the interval, which flatters whichever app has been running longer. More on measuring this properly.
If your numbers come out different from mine I would genuinely like to see them. The whole point of publishing the method was so that it could be argued with.