[A83] Re: calculator networking - good idea?
[Prev][Next][Index][Thread]
[A83] Re: calculator networking - good idea?
ever heard of "digital phase lock"(ing) ....
lemme help:
manchester 2 encoding works this way
when the line changes from low to high ( _|~ ) it's a 1 .. when it goes from
high to low ( ~|_ ) it's a zero.... now...
the Manchester II packet is like this....
56 bits preamble "10101010101010....."
1 byte: "10101011" marks end of preamble..
48 bits: dest. address
48 bits: source address
16 bits: data lenght
xx bytes: data...
32 bits: CRC checksum...
hte purpose of the preamble is exactly for "syncing" the lcocks.. knowing
that they cannot be exactly the same...the clocks used to read data are
slightly faster .. and using the preamble... the delay needed to make the
clocks synched is calculated (digital phase lock)... and then you're done...
-Nicolas Gilles
----- Original Message -----
From: "Gavin Olson" <gtolson@comcast.net>
To: <assembly-83@lists.ticalc.org>
Sent: Saturday, February 08, 2003 13:14
Subject: [A83] Re: calculator networking - good idea?
> At 11:19 AM 2/8/2003 +0100, you wrote:
> > > Van: Nicolas Gilles <eee20154@port.ac.uk>
> > >
> > > I suggest using 1 hub where the red of each is connected to the white
of
> > > every other (and vice versa).. and use manchester II encoding (or
> > > similar) for transmission.. just like ethernet works :) obviously it
> > > won't be 10 mbps ... rather someting like 8000 bps but what do you
> > > expect... the only problem you face will be collision (but you can
handle
> > > them the same way ethernet cards deal with it ...) .. syncronisation
> > > won't be a problem...
> >
> >A nice idea, though I think it won't work. The Ti's can't synchronise
their
> >clocks, that's the whole purpose of manchester encoding, detecting
desync.
> >
> > Henk Poley <><
>
>
> ...Unless you installed a heating unit near the clock circuitry, and
> cranked it up until the speeds matched :) J/K
>
>
>
>
>
References: