Re: HA without fallback-CBUS

[Chris Hunter <cjhunter@xxxxxxxxxxxx>]

Ian -

>Not really ...

brilliant, many thanks ...

'just one question, please ... I'd assumed the sequence "orange /
white,
orange, green / white, blue, blue / white, green, brown / white,
brown"
was a misprint, but then you said "white colour white colour white
colour white colour" was not right ... 'just want to be clear !

Chris

PS: noticed a reference to WiMax today, as potential successor to WiFi ...


Ian Lowe wrote:

> > I've also heard that coax would be better than twisted-pairs, the
> separation & the
> > dielectric being the thing and the twisting just a way of keeping
the
> pairs
> > together - but using coax would be too expensive.   Am I right in
> all this
> ?
>
> Not really.
>
> We used to use co-ax for networks, the old BNC connectors and 10-Base2
> cabling that ran machine to machine in a long bus rather than the more
> modern hub to device cabling.
>
> There are situations where Coaxial cables are the best, but that tends
> to be
> for carrying very high frequency signals, which is not always the same
as
> high bandwidth. Think of it in terms of radios (because, basically,
that's
> what they are!) - if you have a single tuner then you can listen to a
> single
> station - a fixed amount of audio information. If you tune to a higher
> frequency, you can get another channel but still the same amount of
> information. Move to a completely different range of frequencies and
> you can
> pick up TV - again a single channel, but more information.
>
> Now, instead, picture that you have thirty or so tuners, all tuned to
> different stations - you can now receive a whole lot more audio
> information
> at once, at the same frequency as before by using more of the
available
> bandwidth. That's more what Gigabit Ethernet (and ADSL for that
> matter) do -
> they have lots of receivers on a chip, making use of much more of the
> available cable.
>
> Co-ax is really good for a single high frequency signal, like your
> cable-tv
> box, but twisted pair is much better for handling digital signals
(square
> waves essentially behave like an infinite number of sine waves going
down
> the same cable at once)
>
> Those twists are not just to hold the cable together, far from it.
Within
> each pair, one wire (normally the coloured one) is the signal wire,
> and the
> other (the colour + white one) is a ground wire. When a signal is sent
> down
> these wires, the signal wire gets the raw output, and the ground wire
gets
> the opposite of the signal (so, if the signal is a "1", the
ground
> wire gets
> a "0" and so on).
>
> This makes it dramatically better at carrying signals in a
"noisy"
> environment (like your house) - the signal wire and ground wire are
tied
> very closely together, so they get exposed to the same noise along the
> path,
> and the receiver can easily get the signal back at the other end.
>
> All of the pairs in the cable have a slightly different twist rate, to
> give
> even more protection - it helps stop the signals from bleeding between
> pairs
> (called "cross talk")
>
> Incidentally, this is the source of the biggest "gotcha" in
home cabling
> (and a distressing number of businesses too) If someone wires the CAT5
> into
> an RJ45 plug as "white colour white colour white colour white
colour", you
> get a strange bug - computers can transmit but not receive - they get
> a link
> light, but no data back. To make matters worse *some* PCs and network
> cards
> will work okay, and some won't.
>
> The reason - the first two cables are okay - that puts the Transmit +
> and -
> on a single pair, but the receive + and - (pins 3 and 6) end up on
> different
> pairs, which means that the signal is badly deteriorated. If it works
(and
> usually gigabit cards will work well at 100Mb speeds) it will be slow
and
> unreliable.
>
> > at the moment, it seems cables are running ahead of the equipment
being
> connected, at least > when it comes to Giga-Ethernet - but for how
long, I
> wonder - will the move to HD TV & DVD
> > change the situation, at least in multi-channel distribution
> situations ?
>
> I wouldn't worry about multi-channel distribution on HD. It's highly
> unlikely to be even possible, given the DRM paranoia that the studios
are
> building in right at the lowest level.
>
> Given that HDCP is all about a secure signal path, the chance of
> having two
> outputs active on a player at once is pretty much nil, and
conventional
> splitters won't work. That's the thing about HD and it's DRM B.S. -
it's
> going to prevent hobbyists like us from doing what we want to do (like
> pipe
> AV around the house, watch the movie that's playing "big
screen" on the
> projector on the wee LCD telly in the kitchen at the same time as well
> etc.)
>
> Watch for the "WARNING" page on HD DVDs - I'd bet the dug's
kidney
> that they
> will start to have "illegal if played on non-approved
hardware" messages
> soon. Remember what Vivendi's CFO said about DRM - "Pay per view
is
> coming,
> get used to it."
>
> I.