Re: Networked 8x calcs :)


[Prev][Next][Index][Thread]

Re: Networked 8x calcs :)



On Thu, 30 Oct 1997 23:42:37 -0500 "Thomas J. Hruska" <thruska@TIR.COM>
writes:
>At 02:51 PM 10/30/97 PST, you wrote:
>>>Okay, I have this strange idea for creating a network for the TI-8x
>>>graphing calculator series.  I am picturing a small box where 4-5
>>people
>>>connect their calcs via a link cord to it.  From there, they all
>enter
>>a
>>>game and begin playing against each other.  Is this possible or even
>>>feasable to be done?  NETBIOS networks basically do the same thing
>as
>>what
>>>I just suggested.  The only thing that we would find difficult to
>>program
>>>is the synchronization of the packet protocol.  What do you think?
>>>Possible?  Worth the effort?
>>>
>>I have been toying around with this idea for a while. Basically it
>would
>>require a box with each calc having an "IN" line (to calc) and an
>"OUT"
>>line (to hub). Since linkports have 2 lines, this is not a problem.
>The
>>hub would connect the OUT line from 1 calc to the IN line of all
>except
>>the sending calc, so 1 woud send to 2,3,4; 2 would send to 1,3,4;
>etc.
>>The sending calc would wait for silence, and then transmit a
>"reserve"
>>signal for a few milliseconds. after this, it would check the IN line
>to
>>see if anyone else is putting out a reserve signal. If this happens,
>all
>>calcs stop trying to send for a random amount of time. If there is no
>>collision, the sending calc would then send a synch signal and then
>the
>>packet, which I would imagine to be start:to:from:size:data:CRC:end.
>The
>>protocol used in the data segment would be different for each game.
>How
>>does this sound?
>
>Sounds pretty good.  You must have been thinking on this for a while.
>I
>know the protocol part is pretty simple to figure out, but what I want
>to
>know is, how many people will make the device and is it really worth
>the
>effort?

You could expand the idea further and use ir or radio links, (if they are
ever fully developed) with a few modifications.  It would be simpler to
have two, one-way channels then the two two-way channels used by standard
ti protocol, thereby simplifying the design of the wireless links.  These
ideas have been discussed before using the two way channels an i2c
routines, but this method would probably be much easier to implement.


Follow-Ups: References: