Prev: How to run FreeRtos on MSP430
Next: 256-fold (or even more) increase in speed on copper wirepossible IMHO
From: Hans-Bernhard Bröker on 11 Mar 2010 15:54
RalfM wrote:
> Grant Edwards wrote:

>> I'm sure all those engineers at TI, Bell Labs, DEC, Intel, HP, etc.
>> were all wrong.

> I think they all thought of only bit-serial transmission, not going
> further in the dimension.

You have _no_ idea. None at all. Really.

You're basically calling 100+ years' worth of study in electrical signal
transmission theory worthless, purely on the basis of what _you_ "think
they all thought". Well, think again.

Better yet, stop writing for a while and use the saved time to read up
on the subject before you embarrass yourself any further. For starters
let me suggest you look up, and _understand_ all of the following
transmission standards:

56kbit/s analog modem over POTS
ADSL/SDSL technologies
QAM-256 modulation used for digital cable TV
Gigabit Ethernet

>> Do you think that everybody goes to such measures to avoid DC
>> signalling just for fun?

> My understanding is that one can do it better, faster, and cheaper.

That understanding lacks both completeness and correctness in rather
critical amounts.

> For example 1000BASE-T uses 4 pairs (!) of wires, IMHO a waste of wires
> and HW.

You've been investing an awful lot of bandwidth here into proving that
the 'H' in that 'IMHO' is a lie.
From: RalfM on 11 Mar 2010 15:56
Andy wrote:
>
> In a binary, single ended (not differential) signal, the transmitter
> is a conceptually just a switch that puts either the 1 level voltage
> or the 0 level voltage on the line, and can be very fast, very low
> power and/or very inexpensive. The receiver is conceptually just a
> comparator with a built in reference.
>
> What you are proposing, multilevel discrete signaling (as opposed to
> analog signaling or binary signaling), is already used in some
> versions of Ethernet, and internally in the latest generation of flash
> memory, where multiple bits are stored in each cell by having more
> than just two discrete signal levels stored per cell. Most of these
> systems that have to operate fast use primitive DAC and flash ADC
> circuits for transmitting and receiving the signal.
>
> Compared to binary signaling, multilevel discrete signaling has less
> noise margin for a given min-max signal swing (there is less
> difference between discrete signal levels).

Of course true, but 256 levels should IMO be within reach with todays chips...
From: WangoTango on 11 Mar 2010 16:11
In article <hnbe5j$o2u$1(a)speranza.aioe.org>, rm(a)invalid.invalid says...
> WangoTango wrote:
> > In article<hn9i18$m3h$1(a)speranza.aioe.org>, rm(a)invalid.invalid says...
> >>
> >> Just some crazy thoughts of mine... :-)
> >
> > http://en.wikipedia.org/wiki/Constellation_diagram
>
> Isn't this intended for AC signals? I was meaning DC signals like in Ethernet.
>
Ummm, Ethernet isn't DC.....

Be a bit tough to get those DC voltages through the isolation magnetics,
wouldn't it?

From: RalfM on 11 Mar 2010 16:12
Hans-Bernhard Br�ker wrote:
> RalfM wrote:
>> Grant Edwards wrote:
>
>>> I'm sure all those engineers at TI, Bell Labs, DEC, Intel, HP, etc.
>>> were all wrong.
>
>> I think they all thought of only bit-serial transmission, not going
>> further in the dimension.
>
> You have _no_ idea. None at all. Really.

Wrong. I have my own unchewed ideas:-)

> You're basically calling 100+ years' worth of study in electrical signal
> transmission theory worthless, purely on the basis of what _you_ "think
> they all thought". Well, think again.
>
> Better yet, stop writing for a while and use the saved time to read up
> on the subject before you embarrass yourself any further. For starters
> let me suggest you look up, and _understand_ all of the following
> transmission standards:
>
> 56kbit/s analog modem over POTS

That's analog. I'm interessted in digital technologies.

> ADSL/SDSL technologies

A waste of bandwidth by splitting it into 2 or 3 channels
instead of transmitting everything equally in digital packets, incl. voice,
by using a priority code in a header frame field. But this is already above
the wire level. Im talking of the low-level wire level.

> QAM-256 modulation used for digital cable TV
> Gigabit Ethernet

I have the feeling the computer network engineers can learn much from the digitial
cable TV engineers as the latter seem to have much faster transmission speeds
(just imagine trasmitting about 1000 (!) cable TV channels over just 1 coax cable!).
I wonder why the same underlying technology is not used with computer networks
to make much faster network transmissions possible over plain old copper wire.

>>> Do you think that everybody goes to such measures to avoid DC
>>> signalling just for fun?
>
>> My understanding is that one can do it better, faster, and cheaper.
>
> That understanding lacks both completeness and correctness in rather
> critical amounts.
>
>> For example 1000BASE-T uses 4 pairs (!) of wires, IMHO a waste of
>> wires and HW.
>
> You've been investing an awful lot of bandwidth here into proving that
> the 'H' in that 'IMHO' is a lie.
From: RalfM on 11 Mar 2010 16:36
To all the people who have dicussed this topic:

Imagine Satellite TV: it is possible to get more than 1000 TV channels
via the dish. Now this is a huge data rate and data quantity that gets received
by the dish and made available in the receiver.
Now, what do you think: is a satellite link faster than a cable link?
I think a cable should allow more capacity and reliability than air transmission.
But then why don't we see such transmission rates in computer networks?...
First  |  Prev  |  Next  |  Last
Pages: 1 2 3
Prev: How to run FreeRtos on MSP430
Next: 256-fold (or even more) increase in speed on copper wirepossible IMHO