Seems to have some control loop deficiencies. If CPU utilization is ~50% then it will enable turbo and utilization will fall below 50% and then it will disable turbo again.
Absolutely. I don’t know what this tool is about, but the README statement is wrong; I can’t think of any distribution dumb enough to use performance governor by default. Most Intel-based platforms use Intel-specific governors.
I developed an energy-aware DVFS controller for HPC clusters a few years back, and there was no such thing as performance by default even on compute nodes.
The Arch Wiki suggests that both powersave and performance have dynamic scaling, and running 'watch -p -n 0.25 grep \"cpu MHz\" /proc/cpuinfo' indicates that both of them do, with powersave mostly sitting at 1.2GHz with occasional increases on a handful of cores while powersave fluctuates from 2~ to 4.2GHz across all cores with minimal load.
The linked docs from Clear Linux say that power draw isn't entirely dependant upon cpu frequency when there's no load so there's no issue with keeping it on performance, and I'm not about to doubt Intel here, but I'd be surprised if powersave didn't save power if only by clamping down how many resources programs can use. Anecdotally I've noticed that the powersave governor doesn't really work too well when doing things like running virtual machines and will keep the frequency very low, as if the scaler is blind to the resources the VM is using, while having it on the performance governor will pin all my cores to a far more appropriate 4.2GHz.
Without touching anything, it's using powersave on my box. It seems to work well enough, both on performance as well as battery life, but that's just my personal experience.
The question that this tool completely fails to consider is that clocking the CPU lower isn't automatically the most battery efficient thing to do, cf. race to idle mentioned above. To me, this sounds a lot like someone on the trough of the Dunning-Kruger curve looked at a problem and wrenched at it without understanding enough of it.
> Since kernel 3.9 the new scaling driver intel_pstate is available and enabled by default on Intel Sandy Bridge (or newer) hardware. intel_pstate supports the governors powersave (recommended default) and performance only, ondemand is gone.
This tool is a great idea. On recent computers your only choices are: powersave or performance.
That looks stupid compared to automatic CPU throttle that we had before.
So, similar to the author of this tool, I also waste time to switch my computer from powersave to performance depending on my tasks and if keeping battery is critical or no. Like when traveling without power plug.
Also, maybe I use it badly, but with cpufreq in have to manually change all the state of all cores individually and can do a single command to change all of them!
I recently installed Ubuntu 19.10 on my Thinkpad and one of the first things I noticed was how bad battery life was. If this tool can help, it'll be a godsend.
You'll probably get a lot more benefit from just pulling up powertop and fixing whatever dumb thing is eating your battery. Common issues that are never right by default are PCIe ASPM, SATA ALPM, audio codecs that are always powered, USB devices that are always active, bluetooth spamming away when you're not using it, etc.
They try to do the same things in different implementations. They conflict with each other and you shouldn't install both. Tlp seems to be the modern default replacing laptop-mode, but I don't have details around why.