On Mon, Jul 07, 2003 at 10:38:18AM -0500, David Young wrote:
> >>From this paper:
> 
> On Mon, Jul 07, 2003 at 07:51:52AM -0700, Wolfgang S. Rupprecht wrote:
> >         http://black.csl.uiuc.edu/~prkumar/ps_files/compow_ewc_2002.pdf
> 
> "When we estimated the latency of a power change at the network layer
> by monitoring ping traffic on the network, it was close to 100ms." =(

So not a real problem for a host [1], but not usable for an AP or in
ad-hoc mode.

	David

[1] Reduce power until you fail to get a response, then increase it.
Reduce rate at which you drop it once you seem to have found
a working level.

-- 
David Laight: david@l8s.co.uk