Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!dali.cs.montana.edu!uakari.primate.wisc.edu!caen!umich!sharkey!rjf001!mudos!mju
From: mju@mudos.ann-arbor.mi.us (Marc Unangst)
Newsgroups: comp.dcom.modems
Subject: Re: Telebit Announcement of V.32 bis for T-1600
Message-ID: <1991Jun29.015528.3458@mudos.ann-arbor.mi.us>
Date: 29 Jun 91 01:55:28 GMT
References: <1991Jun26.202632.21097@telebit.com> <1991Jun27.030207.1969@shaman.com>
Organization: The Programmer's Pit Stop
Lines: 25

In article <1991Jun27.030207.1969@shaman.com> jiro@shaman.com (Jiro Nakamura) writes:
>but now with the advent of v32bis with 19200 bps FULL DUPLEX, I think

Minor nit-picky point: V.32bis is 14400bps FDX, not 19200bps.  (However,
with V.42 and even a little bit of compression from V.42bis, you'll probably
get at least 19200bps throughput.)

>happened, you can easily double PEP's throughput (or at least make it
>full duplex) can't you?

One thing I'd like to see in PEP is dynamic, partial line turnaround.
The connection starts out as 50/50 send/receive, and as more data is
sent in one direction than the other, the modems gradually turn around more
and more channels.  If protocol spoofing is active, perhaps the connection
could start out as 75/25 send/receive or something like that, to give you
increased initial throughput.  Such a modification would allow much better
performance on full-duplex links without much effort.  The modems would
just have to keep track of how many characters have been sent and
received (something I believe they already do, anyway), and negotiate
some way to turn around individual channels.

-- 
Marc Unangst               |
mju@mudos.ann-arbor.mi.us  | "Bus error: passengers dumped"
...!hela!mudos!mju         |