Path: utzoo!attcan!uunet!wuarchive!zaphod.mps.ohio-state.edu!uakari.primate.wisc.edu!aplcen!haven!adm!smoke!gwyn
From: gwyn@smoke.BRL.MIL (Doug Gwyn)
Newsgroups: comp.sys.apple2
Subject: Re: How do I get kermit to go faster than 2400 baud
Keywords: IIGS, TWGS, Telebit Trailblazer, Kermit
Message-ID: <13549@smoke.BRL.MIL>
Date: 14 Aug 90 13:31:55 GMT
References: <1990Aug13.135356.11691@lgc.com> <1990Aug13.212155.29044@laguna.ccsf.caltech.edu>
Organization: U.S. Army Ballistic Research Laboratory, APG, MD.
Lines: 19

In article <1990Aug13.212155.29044@laguna.ccsf.caltech.edu> toddpw@tybalt.caltech.edu (Todd P. Whitesel) writes:
>max@maxsun.lgc.com (Max Heffler) writes:
>>implying that I had a 19200 baud connection; however, the throughput I
>>measured under kermit was only around 2400 Baud.
>Kermit 3.86 can handle 9600 (through a Super Serial Card and through the GS's
>modem port) without too much trouble. I'd check your control panel settings --
>make sure THEY are set to 19200.

I think this misses an important point -- measured throughput cannot exceed
the communication channel bit rate, but it certainly can fall far short of it.
There are delays on both ends and, additionally, such high bit rates over
voice-grade phone lines rely upon statistical compression techniques and half-
duplex operation to attain apparent rates like 19.2Kbps.  If the data does not
fit the assumed statistical model, as compressed files would not, or if the
communication protocol requires frequent turnaround for acknowledgment, as I
think KERMIT does, then the so-called 19.2Kbps modem simply cannot provide
19.2Kbps apparent throughput.  Such modems work best when sending a large
amount of fairly patterned data in one direction, such as display characters
for a display screen, or large blocks of data in a batched transfer protocol.