On 5/28/13, Thomas Pfeiffer <colo...@autistici.org> wrote: > On 28.05.2013 12:04, Àlex Fiestas wrote: >> For what is worth (and from the Solid side of things). >> >> Batteries have improved a lot since last time we discussed this issue, >> back on the days a high CPU round of 10min would drain huge percentage >> of the power in your battery, hence the estimation was really bad. >> Additionally the estimation was done in most of the cases in the >> software side so calculation was always bogus. Now days the situation >> has changed though. >> >> Batteries are way smarter than they used to be, even the stupid ones are >> kinda smart though. In most laptops sold in the last years you can find >> batteries that implement SBS (Smart Battery System) or similar (there >> is another one I can't recall). >> Additionally battery capacity has grown a lot while cpu power >> requirements have decreased (specially since Intel Sandy Bridge) so a >> CPU pike of 10mins will not affect that much the estimation time (and >> systems like SBS are smart enough to prevent that). >> >> Maybe usability wise showing the remaining time is not recommendable or >> it is confusing, but technically the situation has clearly changed and >> the remaining time can now be shown accurately. > > I cannot see any usability disadvantage of showing a remaining time if > it is accurate. If we can get the time with an accuracy of e.g. 10 > minutes, it is useful from a usability perspective to see the remaining > time with a precision of 10 minutes. > > Do we know whether a "smart" battery (i.e. managed by SBS or similar) is > present? If we do and the presence of SBS or similar does indeed offer > us accurate remaining time estimation even with fluctuating consumption, > I see no reason not to show it in this case.
You guys have heard about "moving average" right? Quoting from wiki[1]: "A moving average is commonly used with time series data to smooth out short-term fluctuations and highlight longer-term trends or cycles" - ie. no random jumping from hours to minutes and back. [1] - https://en.wikipedia.org/wiki/Moving_average Cheers -- Martin Klapetek | KDE Developer _______________________________________________ Plasma-devel mailing list Plasma-devel@kde.org https://mail.kde.org/mailman/listinfo/plasma-devel