Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!usc!wuarchive!sdd.hp.com!spool.mu.edu!agate!ucbvax!ralvm31.vnet.ibm.com!petty
From: petty@ralvm31.vnet.ibm.com ("Jack Petty")
Newsgroups: comp.dcom.modems
Subject: V.32 vs Telebit: Which should I buy?
Message-ID: <9105210700.AA14518@ucbvax.Berkeley.EDU>
Date: 20 May 91 11:54:22 GMT
Sender: daemon@ucbvax.BERKELEY.EDU
Organization: The Internet
Lines: 33

Ref: article of 16 May by tech@mich-ns.Michigan.COM

I think it is important to differentiate between the requirements in the
V.32 recommendation and the implementations of V.32 done by different
vendors.

The author writes:

> V.32 will fall back to half-speed with a noisy
> line and then will quit altogether if things get worse.
>
> Once V.32 downshifts, it cannot recover.

Both points are implementation decisions.  The author may be very
familiar with the T2500 and assumes incorrectly that all V.32 modems
must behave the same way.  I have seen one V.32 modem
that does not fall back at all - it just goes on-hook if the connection
degrades.  I have seen another that will fall back from 9600 bps to
4800 bps and then from 4800 bps down to V.22 bis 2400 bps.  I have
also seen a V.32 modem that will fall forward from 4800 bps to 9600
bps.  For me, all of these can be considered to meet the requirements
of the V.32 recommendation.

The author also indicates Telebits work "fine" for interactive use.
"Fine" is often a matter of personal taste.  What is the single character
round trip time?  I think the IBM 7855 is a typical V.32 modem and its
single character round trip time is in the range of 60 to 130
milliseconds (depending on whether MNP or compression is turned on).
I would expect a USR modem with its slow speed back channel to be about
the same for single characters.
Have I read that for the Telebit proprietary modulation scheme it takes
about 1 second to turn the line around?
Jack Petty