Rockbox.org home
release
dev builds
extras
themes manual
wiki
device status forums
mailing lists
IRC bugs
patches
dev guide



Rockbox mail archive

Subject: Re[2]: charging algorithm averaging

Re[2]: charging algorithm averaging

From: Uwe Freese <mail_at_uwe-freese.de>
Date: Tue, 13 Aug 2002 11:59:53 +0200

Hello Heikki,

Heikki Hannikainen wrote on Tuesday, August 13, 2002, 9:11:08 AM:

>> instead of delta = (s6 - s1) / 6
>> calculate if by delta = ((s4 + s5 + s6) - (s1 + s2 + s3)) / 6

HH> Yes, this is a good idea, thanks.

HH> I implemented it and am trying it out now, with a small modification
HH> though: the above would produce much smaller values for the delta, so at

I now think it should be:

calculate if by delta = ((s4 + s5 + s6) - (s1 + s2 + s3)) / 3

because (s4 + s5 + s6) makes an average value at s5 and (s1 + s2 + s3)
at s2. s5 and s2 are 3 samples away, so is should be "/ 3".

HH> first I'll try with only two samples at both ends of the range.

Then it should be "/ 5". But try also the one with the average out of
three values! With dividing by 3 it should work better (with the old
delta of -50).

Bye, Uwe.
Received on 2002-08-13

Page was last modified "Sat May 23 08:12:40 2020" The Rockbox Crew