Does Nanonote give any signal when battery is near shutdown voltage?

Rubén Berenguel ruben at
Tue Apr 20 03:13:29 EDT 2010

I think the "correct" one to know about batery status is cat
/sys/class/power_supply/battery/capacity. Not sure, but it gives a
number I guess it is between (0) and 100 with the percentage (I guess,
again) of remaining battery power. Yesterday I started a minute cron
job outputting voltage_now, status and capacity, and will leave it
running until battery depletes to see which of these numbers is more
meaningful (and estimate times).

After that I'll probably write something for every 5 minutes checking
that value to be able to generate nice plots (in my main computer, not
in the nano sadly) of power usage. Geek power :) Then it will be able
to notify, somehow (what about it playing some sound file?) or
whatever. By the way, I solved the problem with the mic: it was turned
off and had to be turned on... Probably the same problem I had before.
Thanks Marc!

Btw, I don't have my cross-compiling computer here and can't check: is
"at" (i.e. at 06:00 04212010 echo "Good morning!") in the instalable
packages via menuconfig? I prefer at than cron, yesterday I got mad at
it :/


On Sun, Apr 18, 2010 at 07:19, Delbert Franz <ddf at> wrote:
> On Tuesday 13 April 2010, Ron K. Jeffries wrote:
>> Don't forget that Nanonote has Lua, a very capable yet lightweight
>> scripting language. My understanding is Lua has capability
>> to interface with GTK
>> ---
>> Ron K. Jeffries
>> >
>> > When I get time, I want to investigate adding numpy to the python
>> > already there--then I can do some interesting things on this little
>> > jewel-even with its current limits:) Would also like the python-gtk
>> > package as well.  Then this would be a really sweet little gizmo!
>> > (I'm assuming we can reach the DirectFB from python-gtk.)
>> >
> Thanks for the tip about Lua.  I did some research and then
> translated a simple adaptive integration routine from Python to
> Lua (rather simple but one has to declare all "local" variables as
> local in the recursive functions in Lua-the are local by default
> in Python).  Th syntatic differences for my simple case was small.
> Some simple timing tests on my Acer Aspire One showed about a factor
> of two advantage for Lua.  However, the test was short so it might
> not be valid.  Can't do a valid test on the Nanonote because Lua
> is compiled there with single precision floating point whereas Python
> only comes in double precision:)  When I set the tolerance to the same
> value and run the Lua routine on both the Acer and the Nanonote I see
> about a factor of 10 difference in time:  The Acer computes the results
> in double precision in 1/10 the time that the Nanonote computes the
> result in single precision.  The number of function evaluations agrees
> exactly.  Not bad considering the Acer using IEEE hardware floating point
> and the Nanonote uses some "unknown" OpenWRT software floating point:)
> Over the next few weeks I'll do some more simple numerical tests to see
> how the Nanonote does.  I did not expect it to be fast but for many possible
> applications it is fast enough-and that's all we need:)  In fact, considering
> the electrical power drawn by the CPU, it does quite well on speed.
> There are bindings for Lua and gtk but I don't think they exist in the
> OpenWRT list of packages.  The Python gtk bindings appear to be there
> but are not in the Nanonote image--yet.  However, the Nanonote software
> is in its infancy--we should see rather rapid changes over the next
> few months.
>                             Delbert Franz
> _______________________________________________
> discussion mailing list
> discussion at

* Rubén Berenguel
* Dep. de Mat. Aplicada i Anàlisi
Gran Via de les Corts Catalanes 585
08007 Barcelona, Spain

More information about the discussion mailing list