[Community-news] OLPC News (2008-02-09)

[david at lang.hm]

On Mon, 11 Feb 2008, Hal Murray wrote:

>> what's the power draw of the WLAN chip if the transmitter is turned off?
>
> Most of the power goes into the receiver.  Yes, that might seem
> counter-intuitive.
>
> To first order, the transmitter doesn't take any power unless it's actually
> transmitting something.  After that, it's linear with how much you transmit.
> There may be significant overhead on each packet.

agreed, although there is probably some level of control in how strong to 
transmit.

I'm familiar with two-way transmitter power requirements as I am a Ham and 
also do a fair amount of emergancy services communication work. in my 
experiance there, the transmitter can take so much more power then the 
receiver that even a small amount of transmitting can match the power 
needed to run the receiver for a significant amount of time (I'll point 
out that those radios did transmit _far_ more powerful signals than the 
WLAN does, so the balance may be different)

> So if you are using transmit power you are probably doing useful work.
> (That's assuming that you call maintaining the mesh topology useful and
> things like that.)

almost, what I am thinking is that there may be value in listening to the 
mesh without transmitting to gather the state of the mesh and decide if 
you need to participate and transmit.

if the mesh is extremely chatty (which I am gathering from other posts) 
there may be noticable savings in just listening, but not participating in 
maintaining the mesh, unless there aren't enough other machines doing so.

David Lang