Re: X-10 Mister House Motion sensor problems

["Robert Green" <ROBERT_GREEN1963@xxxxxxxxx>]

"Marc F Hult" <MFHult@xxxxxxxxxxxxxxxxxxxxx> wrote in message >
<ROBERT_GREEN1963@xxxxxxxxx> wrote in message

> >> In my experience, X-10 is almost never quick enough if the system needs
> to
> >> dim instead of toggle ON-OFF. And if you get up at 3AM , who wants the
> >> lights to go on suddenly at 100% bright?
> >
> >That's an easy one.  Just get a two-way wall switch or module.  They'll
> >remember the state last dimmed to and without any of the flash problems.
> >
> >But that's not and ideal solution.
>
> Two-way x-10 modules and switches cost ~$30 and up.

What, now you're price sensitive?  (-:  Yesterday you were trying to spend
$2000 on an upgrade for me for features I don't need and today you're trying
to save me $10.

The point I was making was that X-10 already had a solution to what you were
calling a problem.  I was also pointing out that the two way X-10 modules
are reasonably priced but not as dirt cheap as a vanilla lamp module.

> An INSTEON ICON wall  switch is $19.99  Why recommend
>  getting something new that costs more and does less or worse?
> Not Either/Or. Get it?

Get what?  That you think mongrelization to save $10 is a good thing? I
would NEVER recommend an Insteon Icon switch in place of a slightly more
expensive two-way X-10 switch for a newbie X-10 user for two reasons:

One is the lack of codewheels.  I like to be able to see what a module is
set at without a lot of hoopla.  I've seen how new users like my wife react
to devices with codewheels compared to Hawkeye "morse code" address entry.

I like modules and switches to retain their settings.  Especially if they
are buried behind wall plates, in drop ceilings and behind furniture.
Especially if I am going to be asked by that end user for help with their
system.  I am not convinced Insteon switches can retain their settings in
all environments.

The second reason is homogeneity.  I would not introduce a second protocol
into a system if I really didn't have to.  That's just introducing another
possible failure vector for no gain whatsoever.  I've been bitten by mixing
manufacturers in X-10.  Lots of little "gotchas" there is absolutely no
reason to inflict on a newbie.

The Insteon circuitry might be nice if I were planning to go to Insteon, but
I'm not headed that way so it's foolish to have it there.  The more
components and circuits, the more chances for failure.

If you recall, the discussion leading to this current message was about what
I would recommend to newbies and why.  I would NOT recommend "a bit from
here, a bit from there" approach unless it was absolutely necessary.

Why?  Well it's just too easy for tech support at X-10 to hear "Insteon" and
blame them and for Smarthome to hear "X-10" and do the same.  I've had it
happen to me more times than I can recall in the PC world so I would have to
ask myself:  Why mix?

I also think that Smarthome might be pricing their equipment very
agressively to edge out the competition.  I don't see their prices staying
so low.  Remember X-10's voucher giveaways.  From what I read they put over
$2M of free product into the pipeline as long a people actually bought
something else.  I believe they are not quite so generous anymore.  I'll
feel more comfortable when Insteon appears in gear not made by Smartlabs.

Now, if dimming from 0 level up at every lamp was incredibly important to an
end user, I might change my recommendation and go with the Insteon switches.
But I'd have to believe that the end user was techie enough to deal with
Insteon's way of setting up switch addresses.  I know a lot of people can't
program their EagleEyes correctly which I why I like codewheels for newbies
just like I prefer training wheels for children's bikes.  It helps them gain
their balance, learn what that basics are and enables them to take bigger
steps in the future.

I recall reading the manuals for some switches by Smartlabs and one of the
programming sequences was absolutely arcane.  It also, I believe, assumed
you owned a Maxi-controller.  I recall a number of people over the yearas
asking in CHA how they could get around that requirement.

No, I would not recommend Insteon to newbies for a number of reasons.  Get
it?  (-:

> >> Methinks you want this to work so badly that you don't see how badly it
> >> usually works ...
> >
> >One could turn that statement right around and say "methinks you want
this
> >NOT to work so badly that you don't see how well it CAN work."  (-:
>
> Touché !   ... or should we say "touchy"  ? ;-)

Or should we say "touched?"  :-)

> >It just takes a little effort.  It's not brain surgery.  I've solved the
> >problems that could be solved by reading here and doing research through
> >Google. I've worked around the rest.  Now the XTB will allow me to
> >eliminate a lot of compromises and really embrace a centralized
controller.
> >No matter what I tried before, the CM11A and the TW523 could not reach
> >every corner of the house.  Now they can.  That's a big, big change in
the
> >whole design of my system.
>
> Waiting for (eg) INSTEON's many bugs to be addressed is especially
> appropriate for folks that aren't making a clean install of all (eg)
> INSTEON. The firmware on most (eg) INSTEON switches and devices is not
field
> upgradeable, so there will be more than a bit of trauma involved for the
> early adopters who will find that the recommended solution to many of the
> current issues will be to replace all their hardware (Sound familiar ?).

OK - those are *definitely* not the words of an "rah rah" Insteon partisan.
You've just elucidated another reason why the XTB makes such great sense for
me at this moment in time.  I can now afford to wait until the bugs are
shaken out of the "next big thing."  The powerline clutter issues were
beginning to make X-10 unviable for me and I was feeling forced to make a
switch.  Now that pressure is gone.

> 'Course if we had waited until x-10 got all its bugs we would have missed
> out on all the fun over the years -- not to mention the wrath of SO's when
> the lights inexplicably went on at 3AM ...

Eventually, I will have enough haunted house stories to write a book.  But
that's part of the fun of home automation.  I suspect every protocol has its
horror stories.  Just the other night we heard a crackling noise coming from
the guest bedroom at 3AM.  Very strange, almost like a big bug chewing on a
stick.

Turned out it was an appliance module hooked to a window fan with dial
thermostat that had worn its way into X-10 chattering module heaven.  The
more complexity, the more modes for failure.  Yet I am not ready to go the
Thoreau route and own only a single chair that I hang up on a hook at day's
end.

> >It's a good time for it, too, because the mini-ITC PCs have proven
> >themselves to be very capable and reliable.  They'll make better
> controllers for far less $ than many of the hardwired panels out there.
>
> "Far less $ " ??? The mini-itx that Homeseer sells with its software
> installed -- only through distributors that will provide support -- costs
> about ~2700. This is all about the cost of support, and next to nothing
> about the cost of hardware which you seem to assume.

I said, precisely: "They'll (mini-ITX's) make better controllers for far
less $ than many of the hardwired panels out there."

That's straight up and simple.  In my case, and that of many other HA
enthusiasts, we already have the PC HW and SW smarts so we can assume the
cost of support is negligible.  I would NEVER give HS $2700 for a $200
mini-ITX ($100 used from Ebay) that I could (and did) assemble myself.  Nor
would I use HomeSeer until I got the sense that their plug-in problems have
been solved.  I've seen far too many complaints that plug-ins don't work as
advertised and getting plug-in support is way too difficult.

However, I do agree with their choice of a Mini-ITX as perhaps the perfect
home automation controller, which supports my contention that PC based HA,
specifically in the mini-ITX format, will eventually outperform hardwired
panels.  Even Homeseer agrees.   Get it?  :-)

> Also, if one steps away from the hypothetical to the actual, reality is
that
> hardwired panels provide secure, systematic hardware I/O connections and
> connectors in an appropriate enclosure. This is critical. If you look at
> (eg) the $2700 Mini-Itx Homeseer sells for their software, it has none of
> those needed features. It connects to other pc-centric hardware, not home
> wiring.

Let's get back on track with what I actually said first.  The mini-ITX *I*
am talking about costs about $200 for the most elemental version.  You can
mount it, just like an automation controller in a big box, if you like, or
in a case.  You can connect any number of I/O ports, ethernet devices, USB
devices, to that motherboard very nicely with breakout cables and serial and
USB I/O adapters.

In fact, you can build a far more easily maintainable system with a much
higher degree of "swappability" using COTS parts like the ITX than you ever
can with a proprietary panel.  And, in my case, a spare ITX machine can be
used as a regular PC until it's needed.  An Omni motherboard can't do
anything else except be an Omni motherboard.

You can keep a TWO spare ITX's MB's (at $100 each) on hand for a fraction of
what a spare Omni or Elk spare motherboard would cost.  Remember, the ITX
memory and CPU are socketed and easily repaired and upgraded in the field.
Not so with your hardwired panels.  The ITX machines have many more users
testing the system to the limits and have been revised far more often than
any proprietary boards to fix small annoyances.

There are innumerable adapters to connect PC's to relays and real world
devices.  It's just not the problem you're making it out to be.  If it were,
why would HomeSeer choose the mini-ITX as the platform of their choice?

If you're concerned about it not being in the appropriate enclosure, then
you've got 1,000's to choose from.  It's the size of laptop PC.  How hard is
it going to be to enclose?  Really.

> This is/was also part of the Elk MM443-Ocelot comparison that folks
> with no hands-on experience could never grok. The physical installation is
> critical -- not jist the smarts.

What, exactly, is so wrong about the Ocelot/ADI method of HA?  I've seen
top-notch Ocelot installations (at least photos of them) and they seem no
way inferior to the Elk way of doing things.  In fact, their internetworking
capabilities make them superior for the kind of HA work I want to do.  You
just have to love C-Max and ladder logic and to able to forget everything
you know about structured programming.  Well, that's hyperbole again, but by
now the whole world know's I am very hyperbolic.

If I were to give bad marks to the Ocelot/ADI entry, it wouldn't be
something I could easily correct at "enclosures'R'Us.com" it would be the
nastiness of programming the bugger.  I'm not sure what a pure, hardware
independent HA programming language should look like, but C-Max is not it.
I'll know it when I see it.  HomeSeer and CharmedQuark each come close on a
number of fronts, but from different directions.

> My experience with mini-ITX is limited to the three VIA mini-itx's I own
> including one diskless, fanless running XPe off a compact flash  that I
> purchased on eBay for $100.

A marvel of technology.  I often just sit and stare at the VIA because I
know how many man centuries of work have gone into refining every single
component of a modern mini PC.  They are engineering marvels.  And they are
reliable.  We've both noticed the tendency of HA and alarm hardwired panels
to grow more and more tendrils into the PC world.  Ethernet adapters, USB
adapters, all sorts of PC stuff.  That's a sign that eventually, the board
itself will get sucked into the PC.  My fixation with having spares around
for all critical system parts is very much a factor in switching away from
hardwired panels.  Too expensive to keep spares.  HAI went offline for a few
months this year due to Katrina.  Bad things can happen to large areas of
the country.  I want on-site spares.  Using a mini-ITX as a home automation
controller makes that very practical.

> Another runs MS Server 2003. The third is in the
> junk box  awaiting an assignment or recycling through eBay.
> Yes they are neat.

How's the mean time between involuntary reboots been?

> No they are not a cheap panacea IME.

Who's talking panacea?  I'm talking about the future, which I am convinced
will be PC-centric.  Elk and HAI just can't give you the bang for the buck
that a mass-produced motherboard like the VIA ITXs can.  More importantly,
with voluminous disk space, a PC-centric system is far better suited to
dealing with logging requirements than any hardwired panel I've seen.  Want
to add another ethernet connection?  $5 for a lowrise PCI ethernet card.
Tell me what it costs to add ethernet to Elk or HAI.  I'll bet you $5 that
it's more than $5.

> The original fanless 5000 and the new AN and CN-series
>
http://www.via.com.tw/en/products/mainboards/motherboards.jsp?motherboard_id
=400
> are at the ragged edge of what the current version of Power Over Ethernet
> (POE) can provide (~15watts). Mini-ITX will become even  more popular when
> the new POE standard supplying 30 watts+ is adopted. Adoption of
distributed
> and scavenged DC power will change the HA landscape considerably. Banish
> wall warts.

ITX is the way of the future, both for PC's and for HA.  At least that's the
direction I'm headed in.  If POE kills a few wall warts along the way, I'm
all for it!

--
Bobby G.