PowerTOP Still Worthwhile For Extending Linux Battery Life In 2018

First up is a look at the battery power draw when the two laptops were idling within Antergos 18.7 at the GNOME Shell 3.28.3 desktop within the X.Org Server 1.20 session.

Enabling all of the power-savings features -- which mostly came down to chipset, storage, and sound related tunables -- the maximum power draw was now substantially lower during this desktop idle process... The minimum power draw was also 700 Milliwatts lower for the Broadwell-era X1 Carbon and knocked down the average power use by a significant 6 Watts. For the ASUS XU301LAA Haswell laptop, the peak power draw on battery was also much lower though not as much of a difference on the bottom side. The average power use of this ASUS Zenbook was three Watts lower while idling, which is still quite significant.

Now a look at any power differences when these two laptops are under various loads...

First up was a look at the Python performance, which PowerTOP did not hurt the raw performance...

While the Python compute performance was the same, the X1 Carbon laptop saw about a 0.5 Watts saving under this benchmark. But the ASUS UX301LAA didn't see any real difference in the results when running in the PowerTOP-optimized state.

Under the test profile timing the code compilation performance, the speed was about the same while the X1 Carbon continued saving just over a half Watt on average under load. The UX301LAA actually saw about a 100 Milliwatt higher power draw on average, though would basically be scratched off as about the same power use given the very slim difference.