Path: utzoo!utgpu!news-server.csri.toronto.edu!rutgers!usc!sdd.hp.com!decwrl!ucbvax!ucsfcgl!cca.ucsf.edu!wet!tempest
From: tempest@wet.UUCP (Ken Lui)
Newsgroups: comp.dcom.modems
Subject: Re: PEP vs MNP vs v.32 (& v.42bis)
Message-ID: <1458@wet.UUCP>
Date: 17 Aug 90 09:01:23 GMT
References: <32867@cup.portal.com>
Reply-To: tempest@wet.UUCP (Ken Lui)
Distribution: usa
Organization: Wetware Diversions, San Francisco
Lines: 92

In article <32867@cup.portal.com> Carra_its-me_Bussa@cup.portal.com writes:
>(Sorry if this is a rehash, but I can't exactly find any comparisons....) 
>
>I'm planning on getting a high-speed modem.  There SEEM to be 3 types:
>
>USR HST modems using MNP 5 (giving 19.2)
>v.32 modems using v.42bis (Hayes V-series & others)
>Telebit PEP modems
>
>HST (5+) uses compression to increase their speed over the single carrier.

The HST sports the fastest transfer speed over normal telephone
lines.  Its 14.4kbps carrier used in conjunction with MNP 3 and
above results in normal transfer rates of 16.0kbps-17.0kbps.
These figures come from transferring pre-compressed files.

>V.32 runs at 9600 FDX, w/v.42bis to double it's speed over the single carrier.

V.32 with V.42bis will not double its transfer speed when
sending/receiving pre-compressed files.  It works great with TEXT
files, but will be roughly give you the same performance when
transferring pre-compressed files.

>PEP uses multiple carriers (unlike Hayes/USR HST) and can slowly(?)
>     lower the baud and/or change working carriers as the line changes.
>
Yes, and that's the great thing about PEP.  Connections are solid
unlike V.32.  My experiences in calling systems though _satellite
links_ show that V.32 will almost always drop carrier during a
session while a PEP connection rarely drops carrier.  This was
during the Alaskan volcano eruption when tons of ash spewed into
the atmosphere blocking the satellite link for brief periods.
Transfer throughput was down but I rarely lost a connection.
V.32 connections are reliable within the 48 states, however.

>USR has the dual v32/HST modem; Hayes has the Ultra modem.  Both of these
>do MNP5 and v.32, and can talk to each other (right?)
>
Yes.  The two will also talk to a Telebit T2500 in V.32 mode and
a T1500 (the T1500 is only V.32, no PEP).  If you only want PEP,
there's the T2000 that also supports SNA/HDLC connections.  The
Trailblazer Plus supports only PEP.  The T2500 supports
everything the T2000 supports and V.32.
>
>The telebit modems use PEP, but the 1000 is PEP 9600, the + is 9600/v.32,
>and the 2500 is PEP/v.32/HST (?)  (MNP 5??)
>
The T2500 _DOES NOT_ support HST.  You have to be aware of this.
The current Telebit modems are:
	Trailblazer Plus	PEP
	T1000			PEP at 9600bps
	T1500			V.32
	T2000			PEP and SNA/HDLC
	T2500			PEP, SNA/HDLC, and V.32
As far as I know, all PEP modems support protocol
spoofing--XMODEM, Kermit, UUCP.  Currently, all V.32 supported
Telebits sports V.42/V.42bis.

There are Trailblazers for cellular phones, but I don't know too
much about them.

>So I can use the (external?) T2500 and talk to any modem known to man?
>
No.

[I almost wanted to comment that PEP and HST are primarly
half-duplex while V.32 is full duplex; but apparently there's
been some really hot debate whether this is true for the HST on
another network.  Can anyone who is qualified post the answer?
It's true that the HST is full duplex, but transfers data
asymmetrically.  I suppose my question is: "When one says 'full
duplex,' does it automatically imply symmetrical full duplex IN
THE TELECOMMUNICATIONS WORLD?"  i.e. When one says "I'm going to
build a new type of full duplex modem," do all engineers think
it's symmetrically full duplex without further inquery?  Or does
one have to specify symmetrical full duplex to imply symmetrical
full duplex?]
>
>Thanks!
>
>Carra Bussa @ cup.portal.com
>

Ken

I have no connection with Telebit other than a satisfied
customer.
-- 
_____________________________________________________________________________
     Kenneth K.F. Lui	   |  UUCP:	...{ucsfcca|claris}!wet!tempest
     tempest@wet.UUCP	   |  Internet:	cca.ucsf.edu!wet!tempest@cgl.ucsf.edu
			   |	-or- 	claris!wet!tempest@ames.arc.nasa.gov