Path: utzoo!attcan!uunet!lll-winken!lll-tis!ames!umd5!brl-adm!brl-smoke!gwyn
From: gwyn@brl-smoke.ARPA (Doug Gwyn )
Newsgroups: comp.sys.apple
Subject: Re: 'Keeping up' at high speeds (was Re: Kermit at 19200bps)
Message-ID: <8179@brl-smoke.ARPA>
Date: 28 Jun 88 17:14:54 GMT
References:  <8806242338.aa06649@SMOKE.BRL.ARPA> <2076@pt.cs.cmu.edu>
Reply-To: gwyn@brl.arpa (Doug Gwyn (VLD/VMB) )
Organization: Ballistic Research Lab (BRL), APG, MD.
Lines: 16

In article <2076@pt.cs.cmu.edu> ralphw@ius3.ius.cs.cmu.edu (Ralph Hyre) writes:
>But I don't consider that to be 'keeping up', it should be able to
>deal with 19.2 kbps all the time, without any flow control at all.

That's simply not realistic.  Depending on the incoming data stream,
an arbitrarily large amount of work may get backlogged.  Some form of
flow control is essential for any "intelligent" terminal.

>XON/XOFF is a hack at best, and it can cause serious problems with
>certain applications (emacs, etc).

DC3/DC1 control functions are part of the ASCII standard, and they are
widely used, probably more than any other form of flow control.  EMACS
will work fine if it pays attention to the settings of the terminal
handler on the host (in particular, whether DC3/DC1 flow control is
enabled when EMACS is started).  Some implementations of EMACS do this
right.