The problem with HA

["accidental plumber" <aplayerinla@xxxxxxxxx>]

For the high bandwidth stuff, AV, the good things are locked up by the
software content providers.  You pay top dollars for media server but
the quality is 2nd best at best.  Otherwise, all AV units should be
Ethernet (most likely wi-fi) appliances.  How simple.  With under $40
for a full wifi router available at hardware stores, there's not much
price difference adding wi-fi capability to most audio (7.1 sound) and
video (plasma) equipments.

With the low price example of wifi webcams, and wifi surround speakers,
the only reason the other AV are not wifi is because the providers
don't want to.  Either you have to pay up for say multiple satellite
boxes in multiple rooms, or you redigitalize the analog signal.  But if
I have that sort of large house, and spend that sort of money of AV
home automation, I would 1st find a house in those experimental 100M or
200M cable areas.  (1M for DSL/cable, 5-10M for high speed cable).

It's not going to happen for your toaster to become an internet
appliance, for a long long time.  So a low bandwidth home network is
required.  But the major companies aren't interested in low tech stuff
that help save your trouble and money.  They are looking for a Trojan
horse into your home, so that one day everything with an electrical
cord has a XXX home network adapter.

Zeebee, an open standard, looks exactly like one.  Spread spectrum at
2.4 GHz to co-exist with wifi.  It's a full mesh network like, allowing
multiple controllers at the same time.  Interestingly, spread spectrum
isn't associated with power saving for battery operation.  You always
have longer battery life with narrow band modulation.  And any sensible
designer will think about a power supply inside a wall switch, perhaps
charging a backup battery, rather than requiring all the batteries be
changed every year or so.

Zwave is another full sort of network.  It's say open, but I doubt if
the standard belongs to the public domain.  Battery in responders
should have longer life if only batteries can be used.  But narrowband
signals should have more propagation problems inside homes.  If you
have few devices to control, scattered in a farm, I doubt if the
technology is any better than simpler systems.  And if you have lots of
things, I think at some point you need to find a network troubleshooter
instead of an electrician.

Insteon is promoted as a dual carrier, power line and RF, network.  But
in reality, it's x10 in disguise, it's compatible too.  Forget about
the claim that every device repeats power line signal and RF signal as
in a mesh network.  Most switch will be dumb, listening (and repeating)
for power line signal only.  The RF devices is sort of like the X10
transceivers.  It's not a network - you can't have two controllers in
two room controlling the media center without affecting each other.
One command is propagated to the whole house and repeated several
times.  No big network or chip players will be interested.

Again it's open but I think not in the public domain.   That's no big
deal as long as the switch is cheap at $20 with a dimmer, compared to
$10 for x10 without a dimmer.  I'm not going to pay $40 each yet.

The characteristic simulcast feature - either you think it's brilliant
or it's laughable.  Well the company white paper sounds like a 1st year
engineering student just completed a project.  On the RF side, you may
imagine a mesh of reliable RF repeaters. In reality, you should install
only two provided in the starter kid, to solve the phase coupling
problem, and no more.  Simulcast is self inflicted multi path
propagation interference.  FSK modulation is OK, less likely to reduce
the signal or cancel itself altogether, but any significant numbers of
RF devices simulcasting and repeating is mind boggling.  Insteon's idea
is for you to install the two, move them around if necessary to get
good reception, and don't move them afterward.  So if you buy the
coupler repeaters for X10, it's just as good.

Now back to the power line, simulcast using BPSK is a joke.  If you win
the luck of draw in the beginning, adding or removing a switch will
risk killing off the Insteon signal to some appliances or part of the
house.  Unaware users will experience some very subtle problems that is
hard to solve.  Getting a signal meter is a must.

So the few advantage of Insteon over x10 is the acknowledge signal.  So
if you turn off the lawn sprinkler because it just rained, the
sprinkler should really be turned off.  Ideally, the remote controller
should have an indicator light for the on/off status polled from the
device, and a toggle switch.  But the controllers I seen are x10 like,
one on and one off switch.  So the dumpiest way to do it is to repeat
the command 3 times if no acknowledgment is received, and then silently
give up.  It make little difference to x10.  If your dryer noise
disrupts the signal to the sprinkler, it's likely to kill the repeats
too.

Insteon suffers from the same power line problems.  All appliances
change the impedence to the signal, and bad DC adapters are very good
signal attenuators.  Two laptop chargers will have easily noticeable
unreliability effects, if not killing the signal altogether.  You can
plug in one Insteon RF repeater, but you can do it more elegantly with
a cheaper filter (isolator).

Noisy appliances are best dealt with by maintaining strong signals.
Outside effects, like power surge, spikes that drive the electronic
crazy, are supposed to be rare, and that should not happen with a good
power supply circuit within the wall switch.  Brownout effects are just
like the manual override feature of dust to dawn outdoor lanterns, that
should be an option.

So unless you have a lot of money, a $50 switch controlling $1000
lighting, or you need to control something that you cannot see, and it
have to be reliable, then x10 is not too bad.