[TI-H] Re: I remember hearing of a ti hub project a few years ago...


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

[TI-H] Re: I remember hearing of a ti hub project a few years ago...




Strictly speaking, a hub would merely amplify the signals and retransmit
to ALL connected devices.  Even for a broadcast (one calc sends to
multiple), this would not be compatible with the built-in link protocol. 
(which needs bidirectional communications no matter what)

A simple way to make a hub/router work with basic programs is to have a
connect packet set to it with an asm program to set up where everything
goes until another connect packet is recieved.  This does require the
device to track packet start&ends, but that shouldn't be too much of a
problem.  ((If done properly, this might even let you connect multiple
hub thingies together for a larger netwoek.)

The data buffer with a pic16f84 is essentially zilch.  You can't hope to
buffer entire packets except perhaps VERY small ones.  There probably is
enough ram to track packet start&ends, so it's likely not a bad
microcontroller to use (if you can get one ;)

An interesting thing with this chip: I have had a circuit work with no
power, just connected to one ti-86 on two io lines.  It even had enough
power to drive an led display.

-josh


On Sun, 29 Apr 2001 09:55:15 -0700 (PDT) Adam Thayer
<krevinek@matrix-z.net> writes:
> 
> Here is another idea:
> 
> Get someone to learn how to program a PIC Microcontroller and use it 
> to setup a basic 4-Port Hub which will send data to each calc if 
> data is sent to it. Of course this particular setup will not be 
> TI-OS compliant to an extent. You could send vars to multiple calcs, 
> but using TI-Basic for data transfer is out, unless drivers add a 
> couple BASIC functions for hub use. 
> 
> All one would have to do is wire a set of submini jacks into the B 
> array on the Microcontroller and have some power supply to make sure 
> the PIC gets 5V and a clock. One with minor skills in the PIC would 
> be able to hack up a basic schematic in about an hour, and if we 
> could get the IR-link code, could use that to code up the 
> Variable-transfer code. The only drawback is that the data buffer is 
> VERY small, a measly 16k or less, so transfer conditions have to be 
> just right, unless some sort of Static RAM chip of something is also 
> used off-chip. The device wouldn't be difficult, but since I cannot 
> seem to get any PIC to program (especially the F84), I am kinda 
> stuck on development myself.
> 
> 
________________________________________________________________
GET INTERNET ACCESS FROM JUNO!
Juno offers FREE or PREMIUM Internet access for less!
Join Juno today!  For your FREE software, visit:
http://dl.www.juno.com/get/tagj.



Follow-Ups: