console multiplexor

From: ajp166 <ajp166_at_bellatlantic.net>
Date: Wed Oct 4 17:38:54 2000

From: Chuck McManis <cmcmanis_at_mcmanis.com>


>Clearly the simplest solution is a 6PnT rotary switch, some cables wired
up
>with MMJ plugs crimped on their ends. I'm planning to build one of these
as
>I have all the parts and that will get the console going on the
micro-cluster.


That would work if you assert RxD to spacing on disconnect.

Me, I cheat. I don't bother connecting a tube (unless required like
during initial install)
and when needed I use the VT1200 to connect via the net once that is up.

>However, if you're willing to be really clever then you can do much
better
>than this. I've got a nice color terminal (the Link MC70 although its
color
>is a bit unstable these days :-() or I could use a DEC VT340 (probably
the
>actual version for display) and you could, with a bit of smarts and some
>buffering take output from the RS-423 lines, and prior to forwarding
them
>on to the terminal you could inject ANSI color codes on/off. This would
>make the output from each console a different color. (keeping them
straight
>is of course the challenge) If you are really clever you don't allow one
>console to interrupt another mid-line either. I've got an old single
board
>Z80 system that could probably do this, or I could wimp out and use a
CPLD
>feeding a Scenix chip.


Could be done with Vt125 or better yet....

Like I used to do with a Vt320, the system write a status line (25th) in
response to a simple request "_at_who" which runs a DCL script. the
request was stored as the "respones line" on the terminal or as a
simple "w" [set W*ho="_at_who.com". Also other defined commands
terminate with a call to who. The status line also has current
directory
and account. the switch box was a LQPX2-SW and MMJ to DB/P25
cabling.

Allison
Received on Wed Oct 04 2000 - 17:38:54 BST