>     I've seen it on a table on a web page that I can't find anymore...

I think the rated capacity relates to discharge at 10mA down to
1V or something like that. So yes you will get much longer life
at very small drainage. If you can turn everything off when not
needed then that lifetime will approach the shelf life, or beyond
that (assuming they call a cell expired at 1V) if you use the LF.
A reservoir electrolytic (eg 100uF) will help supply IR pulse currents
as the cell nears its end-of-life voltage and increasing internal R

> > (btw, the frequency of the micro is irrelevant when it's asleep)
>
>     Oops, sure. :)

But the mA vs Hz does matter when the PIC's awake.  If you're
writing your own IR codes you could use a slower clock speed
(eg 32kHz) than you'd need for standard pulse widths, but that
would mean going without a ~40kHz carrier unless you use an
external ~40kHz  generator, eg this

http://home.clear.net.nz/pages/joecolquitt/txless.html

The 555's Vcc can be supplied by a PIC pin or two, depending on
what drive you need for the IR transmitter. I recommend a Zetex for
the output transistor - it switches pulse currents better than most
other common types

--
http://www.piclist.com hint: To leave the PICList
mailto:piclist-unsubscribe-request@mitvma.mit.edu