[prev in list] [next in list] [prev in thread] [next in thread] 

List:       familiar
Subject:    Re: [Familiar] Battery lifetimes & CPU scaling
From:       "James ''Wez'' Weatherall" <jnw22 () cam ! ac ! uk>
Date:       2002-01-30 14:30:42
Message-ID: 007501c1a99a$b27b3040$1062a981 () eng ! cam ! ac ! uk
[Download RAW message or body]

> As for lower CPU speed holding the hardware running for longer, if it's a
> CPU-bound task, its not likely going to be using other hardware all that
> much, is it?

Switching things on and off also takes power so you may leave the radio on
while you process received data, before sending a reply.  That means the
radio is burning power even though the CPU is in full use, so a slower CPU
costs you power.

> I wonder how much all of this really matters on modern computer/CPU
> designs, however. Does the CPU not automatically turn off parts it is not
> using? Turn to power-saving mode if it's inactive, as processors tend to
> be most of the time?

Yes, but the most power-consuming single part of the CPU is the oscillator
for the clock in most CPUs.

Anyway, the point of the post I made was that drawing high loads from
Lithium ion batteries can cause significant divergence from ideal battery
behaviour.  It may not be a problem on the iPAQ, because it's peak load is
really pretty low, but it's worth considering.

Cheers,

---
James "Wez" Weatherall
          "The path to enlightenment is /usr/bin/enlightenment"
Laboratory for Communications Engineering, Cambridge - Tel : 766513
AT&T Labs Cambridge, UK                              - Tel : 343000



[prev in list] [next in list] [prev in thread] [next in thread]