Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site sftig.UUCP
Path: utzoo!linus!decvax!decwrl!amd!dual!zehntel!ihnp4!mhuxj!mhuxr!mhuxv!mhuxt!mhuxm!sftig!rbt
From: rbt@sftig.UUCP (R.Thomas)
Newsgroups: net.dcom,net.periphs,net.news
Subject: Re: 9600 full-duplex, by trickery?
Message-ID: <467@sftig.UUCP>
Date: Wed, 17-Oct-84 02:11:04 EDT
Article-I.D.: sftig.467
Posted: Wed Oct 17 02:11:04 1984
Date-Received: Fri, 19-Oct-84 05:35:07 EDT
References: <4468@utzoo.UUCP>
Organization: AT&T Bell Laboratories, Summit, NJ
Lines: 17

>                                        a fairly ordinary 9600-baud half-
> duplex modem -- plus an "EC100" box which does error correction *and*
> makes the half-duplex modem look full-duplex!
> 
>                                            If one assumes that total
> average traffic, both directions together, is somewhat under 960 cps,
> then making a half-duplex channel look full-duplex is indeed feasible.
> They're pushing it for terminals, where the assumption is plausible,
> but it might work well enough for things like uucp.
> 
> 				Henry Spencer @ U of Toronto Zoology
> 				{allegra,ihnp4,linus,decvax}!utzoo!henry

It might work *very* well for uucp.  Uucp is essentially half-duplex.
When one side is sending, the other side is (nearly) idle (except for
things like x-on/x-off).  The only problem is to make sure that both
sides have one of them.  (The old chicken and egg problem!)