Fedora 18 is conservative wrt battery life

[lsatenstein]

On my netbook, I had 4 hours of battery life with Fedora 17

With Fedora 18 (RC4), I now have 4 hrs 20 minutes of battery life.

F18 is more efficient than Fedora 17 (gnome vs gnome)

F18 with KDE gives me 4 hours 7 minutes battery time.

[AdamW]

All of those results would be well within the margin of error for non-scientific testing, I think. Different sessions are always going to vary slightly wrt power usage. I wouldn't draw many conclusions from such relatively small differences.

[hmaarrfk]

Yea.... from my experience, battery life is calculated on the total consumption of power in the near past.

For example, if you are booting up, then you are loading many new things into memory from your harddrive and your processor is working pretty hard consistently getting everything up and running. If you check your battery life RIGHT after that, compared to waiting 5 minutes and then checking again, you will find that your battery life is actually higher (maybe by about 20% or so) if the estimation is done 5 minutes after boot.

Also, if your screen auto dimmed itself for 5 minutes and you check your battery life right after that, your computer will think it is using less power than usual and report a longer battery life than if you were "lightly" using your computer for 5 minutes.

The exact algorithm depends from platform to platform, the 5 minutes I quoted was simply arbitrary and maybe one of the simplest ways you can estimate battery life.

The other factor might be "trickle charging". It is explained here for the case of iProducts but the same might very well apply to other products (iProducts simply have pretty good battery life so people notice a difference between "10 hours remaining" and "11 hours remaining")
http://www.macrumors.com/2012/03/27/...battery-level/

TLDR, you can in fact charge a battery past what your OS would report as "100%" and in doing so notice that the battery life is larger than in other cases.