Signal wrote:

"Arny Krueger" emitted :

Every digital signal is received as an analog signal.

Hmmm... that doesn't sound right. Digital is, like, the opposite of
analogue, man. A signal cannot simply be "received" from an analogue
signal, it has to be converted. The analogue signal is a continuously
variable quantity, y'see, whereas digital is finite.

And if you understood the first thing about signal transmission you'd
understand why a digital signal is subject to 'analogue limitations'.
Like bandwidth for example.

Graham

"mc" wrote in message
. ..

"Richard Crowley" wrote in message
...
"Signal" wrote ...
"Arny Krueger" emitted :

Every digital signal is received as an analog signal.

Hmmm... that doesn't sound right. Digital is, like, the opposite of
analogue, man. A signal cannot simply be "received" from an analogue

It may not sound right to you, but Arny is correct.

He is correct, but he was deliberately misleading in order to get people
to think.

Yeah, its nice when people think and I try to stimulate it, but there is no
intentional missstatement or misapprehesnion.

Naturally, the signal going down a digital wire is not "analog" in the
same sense as an analog audio signal. It is not a voltage that is
supposed to be proportional to the displacement of a speaker cone.

Right, but a so-called digital signal on a line is a continuously-varying
signal with a lot more than just two different states.

But (and this is Arny's point) electricity does not consist of ones and
zeroes. (Well, there are individual electrons, but that's not what we're
dealing with here.) Voltage levels are on a continuum. If you decide
that 1 volt is 1 and 0 volts is 0, then what do you do when the actual
cable, with its resistance, capacitance, and inductance, delivers you 0.8
volt, or 0.5 volt? And there are no instantaneous transitions; if you
switch suddenly between 0 V and 1 V at the input, you will see a much less
sudden transition at the end of a long cable.

More specifically, a TTL signal in the real world is not just 0 volts or 5
volts and nothing in-between. The zeroes are usually some place between 0
and 1.x volts, and the ones are between 3.x and 5.x volts. Or it's the exact
opposite since a lot of logic signals are negative logic.

Furthermore, the transitions aren't instantaneous. Sometimes the transitions
can take seconds, but mostly they take nanoseconds or less. But even
nanoseconds can be "a long time" for modern digital logic.

"Richard Crowley" wrote in message
...
"Signal" wrote ...
"Arny Krueger" emitted :
Every digital signal is received as an analog signal.

Hmmm... that doesn't sound right. Digital is, like, the opposite of
analogue, man. A signal cannot simply be "received" from an analogue
signal, it has to be converted. The analogue signal is a continuously
variable quantity, y'see, whereas digital is finite.

It may not sound right to you, but Arny is correct.

This is a remarkable event. For the longest time Paul Dormer (who is posting
here as "Signal") has been able to say the weirdest things imaginable about
stuff I've posted on RAO, and even been reinforced by others who purported
to be superior authorities. Finally, he gets his comeuppance at the hands of
any number of people who have orthodox ideas about electronics and don't
have a stated agenda of humiliating and embarrassing me.

Thanks, guys - maybe Dormer will finally learn something useful about
electronics.

Mere wordplay. A digital signal exists in a physical world, hence is
analogue in one manner if speaking. So let's ban the term "digital
signal"? No, that would be silly.

Just found this sitting in my Drafts folder. Here goes anyway.


Ruud Broens wrote:

Elsewhere in this thread, you recommend coax for longer (30 ft.) stretches,
that is inconsistent with A and B, Arny :-)

Optical's a problem for longer runs because of the optical loss caused by the
cheap fibre not being optically 100% transparent. Not to do with signal
'degradation', just attenuation.

With respect to welldefined edges, optical is clearly superior - even the
plastic 850 nm home variety - to coax.

I don't think so. The edges are 'rounded off' by the opto coupler response
bandwidth. The early Toslink devices had only around 6 MHz to begin with IIRC
which made them only just suitable. They seem to have improved recently though.

Graham