Rockbox mail archiveSubject: Re: 3.0 - What MUST be done?
Re: 3.0 - What MUST be done?
From: Mike Holden <rockbox_at_mikeholden.org>
Date: Thu, 22 Jun 2006 10:07:27 +0100 (BST)
Keith Mosher said:
> On 6/21/06, Mike Holden <rockbox_at_mikeholden.org> wrote:
>>It could be of course that maybe the runtime prediction is well off?
> Bingo! The old runtime prediction is based off the old battery life
> expectancy. The formula will need to be changed to reflect the new
> durations. Until then you can trick it by upping the mAh rating of your
> battery in General Settings->System->Battery->Battery Capacity.
Indeed. After using it since yesterday, I would estimate that runtime is
increased by around 100%, which is excellent. I now expect to see playinmg
time of around 16-18 hours, compared to the 8 or less I got before.
The estimated runtime is about half what I am expecting to see. After 3
hours of playback of 128kbit mp3 files, the runtime has dropped by 1.5
hours when it should have dropped by 3.
Maybe my recollection of the code is off, or maybe it has changed in the
last year or so since I was actively programming Rockbox, but I thought it
worked on percentage drop in battery voltage from full, and compared this
over time to give an estimate of runtime?
As an example, if the battery charge drops 10% from full in 1 hour, then
we estimate that we have 9 hours left. We don't need to know that we get
10 hours from full. Instead, we work it out on the fly from voltage drop
over time, with a known maximum and minimum voltage.
We calibrate it by measuring actual voltage drop as a percentage, so real
time isn't involved in the calibration.
-- Mike HoldenReceived on 2006-06-22