Closed GoogleCodeExporter closed 8 years ago
Why? Because storing the last 10 raw samples and averaging only from those is
both wasteful of SRAM and inferior to a running average.
It will be important to start the initial voltage sample at something very
close to actual (say, an average of three quick-fire samples after the first
non-zero reading) so that the running average doesn't start at zero and slowly
creep up to actual voltage.
Original comment by gru...@gmail.com
on 15 Nov 2010 at 11:46
Done. A 7-cell virtual history, running average is now standard and seems to be
as stable as the previous, wasteful, 10-cell actual history system.
Original comment by gru...@gmail.com
on 20 Dec 2010 at 8:46
Original comment by gru...@gmail.com
on 23 Sep 2011 at 1:50
Original issue reported on code.google.com by
gru...@gmail.com
on 12 Nov 2010 at 7:24