Re: SD: Networking - tell me about it
[Prev][Next][Index][Thread]
Re: SD: Networking - tell me about it
Assigning the addresses is a fairly simple task so it would be a waste to
have a calc spending all its time doing that. Why not just let the first
calc on the network be the one to assign addresses, and let the whole thing
be a software solution (except for the link cable), that would make the
whole thing a lot easier to make.
Dines
----- Original Message -----
From: "Jimmy Baker" <ashbar83@yahoo.com>
To: <shell-developers@lists.ticalc.org>
Sent: 10. januar 2000 14:58
Subject: Re: SD: Networking - tell me about it
>
> Yes, I remember some sort of protocol designed. I
> think the best thing for this is more like a LAN
> system. Have everyone hook up to a hub, and have one
> calc dedicated to dishing IP's. Wing (of TCPA) and I
> have been discussing this concept for quite a while,
> we just don't have the engineering experience to make
> it work. The server calc would ONLY be used for IP
> assignment, so an 82 could work for that.
>
> --- Grant Stockly <gussie@stockly.com> wrote:
> >
> > I made one of those afew years ago, but NO one
> > wanted to design programs
> > for it.
> >
> >
> > >Shell developers and hardware designers,
> > >
> > >Has a networking system and protocol been created
> > for the TI calcs? If not,
> > >I have a bit of an idea, but it will need all of
> > your help for its
> > >realization.
> > >
> > >For the hardware, there are two options. Either a
> > server system, or a hub.
> > >
> > >The server option would be like a memory expander
> > with a few link ports, but
> > >having a chip controlling traffic with a command
> > buffer, acting much like a
> > >multi-port dial up server. I believe this option
> > would have less work in
> > >terms of the software side of things, as it would
> > interact with the calc
> > >much in the same way as the memory expander.
> > >
> > >The second option would be to actually have a
> > 'network' using addresses.
> > >This would have a hub, much like an ethernet hub,
> > or like a Gameboy
> > >multi-tap (remember those things?); or it could use
> > a daisy-chain method,
> > >using special transceivers. This option would
> > require more in the ways of
> > >software, as it would require a completely
> > different os or shell, or at
> > >least a major patch to the ti-os's linking section.
> > Game's would also have
> > >to be changed to send different commands to the
> > link port, but we would be
> > >able to play real multi-player games.
> > >
> > >Just a few ideas. Tell me about the reality of
> > these ideas. Note that I am
> > >sending this to both shell-developers, and
> > ti-hardware, so you may want to
> > >send responses to both. thanx
> > >
> > >--
> > >M M IIIII TTTTT CCC H H OOO
> > >MM MM I T C C H H O O
> > >M M M I T C HHHHH O O
> > >M M I T C C H H O O
> > >M M IIIII T CCC H H OOO
> > >
> > >"Guitar picks, yeah, those make great fingernail
> > cleaners, screwdrivers, and
> > >percussion sounds in your vacuum cleaner, just in
> > case you didn't know." -
> > >anonymous
> >
> >
> >
>
> __________________________________________________
> Do You Yahoo!?
> Talk to your friends online with Yahoo! Messenger.
> http://im.yahoo.com
>
References: