Re: Re: xPL/xAP an alternative to C-Bu$ ?

["K. C. Li" <li@xxxxxxxxx>]

On Thu, 20 May 2004, David Buckley wrote:

> I cant think of a single feature that makes them anything better
> than "average".  Things that would put them "up
there" would be a

The features you have listed are indeed nice to have but hardly essential
for most domestic or some commercial installations. The standard 4 or 8
channels C-Bus dimmers have zero crossing filter, thermal shutdown,
linearised brightness control, programmable soft start, power up load
status, etc. Admittedly, none of them are breath-taking but taken as a
whole, it is a pretty good device for the market that it is designed for.

> which is proprietry, to the point of being unavailable, short of
> some reverse engineering; you dont have the benefit of the RS232

No reverse-engineering is required. All one needs is a C-Bus PCI SIM
module for TTL interfacing. Add a MAX232 chip and a few capacitors, you
have an RS232 interface. However, some people are trying to move away from
RS232 to something that have at least multi-drop capability like RS485.

> interface.  I'm an NDA-er (part of the CBus enabled clan), so I know
> about the protocols once they are sanitised and made ready for human
> consumption.

And you say the protocol is unavailable, short of reverse-engineering
when you have good access to most aspects of the C-Bus protocol?

> That may be true, but it isnt the essence of X10, which is its a
> plug and play system.  It may work as you design it, but plugging
> something in may compromise the system, as many here have found out.

That is no difference in plugging some badly designed C-Bus, Dynalite, or
Lutron modules into the respective network and compromise the system. The
advantage of the private network system is that the system modules conform
to requirements by way of standards and licensing enforcement, something
that is not practical with X-10 and the mains system.

> However, underlying all that, is that although people may have an
> X10 system that works reliably for them, its not reliable per se,
> unlike for example CBus, which has reliability designed in as a base

It depends on what you mean by reliable. If you are talking about protocol
transmission reliability, X-10 indeed doesn't have much handshaking nor
any error correction capability. It does have, in theory, error detection
as X-10 transmissions are sent twice. Even without the modern protocol
facilities that one expects nowadays, X-10 do work remarkably well in
large number of installations throughout the world, and has done so for
over twenty years. It may be a slow and limited protocol but, in my
opinion, it is adequate for the majority of domestic and commercial needs
at affordable prices.

> requirement.  I havent seen the non-NDA CBus docs, but I guess they
> will have the same explanation in there thats in the NDA docs of
> what a network variable is in therel, and why its important.

X-10 have guidelines on its protocol handling in order to increase its
reliability. I accept that it is far from ideal.

> Finally, I'm not knocking either of these technologies - they are
> both fab, I've been a long time supporter of both.  But one has to
> have ones eyes open.

That is what this group is all about I guess but X-10 do get unfairly put
down at times.

Regards,

Kwong Li
li@xxxxxxx
Laser Business Systems Ltd.
http://www.laser.com
http://www.cbus-shop.com