Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!wuarchive!uunet!hayes!tnixon
From: tnixon@hayes.uucp
Newsgroups: comp.dcom.modems
Subject: Re: Hayes SM 9600 V-series calls Intel 9600ex; sad results
Message-ID: <3976.284e2f93@hayes.uucp>
Date: 6 Jun 91 12:50:27 GMT
References: <1991Jun5.211611.7156@aero.org>
Distribution: na
Organization: Hayes Microcomputer Products, Norcross, GA
Lines: 45

In article <1991Jun5.211611.7156@aero.org>, pthomas@arecibo.aero.org
(Peter L. Thomas) writes: 

> I'm trying to connect to the Intel modem at 9600 v.42bis, a feat which I
> think both are theoretically capable of.  The Intel modem responds with a
> tone I'm not familiar with (it sounds like the old v.22bis (?) tone with a
> series of "clicks" interspersed in i) bee-click-bee-click-bee-click, and then

The "clicks" in the answer tone are phase reversals.  These phase 
reversals cause any echo cancellers in the long distance network to 
be disabled.  V.32 modems do this, because they provide their own 
echo cancellation capability (which is more accurate than the 
network's, since the network is only concerned about it being "good 
enough for voice").  The phase reversals do not affect operation at 
other speeds.

> it falls back and we get a regular 2400 connection.  

This happens because your Intel modem supports only CCITT 
V.32-compliant modulation at 9600bps, which your VSM9600 supports 
only Hayes' Express 96 proprietary fast-turnaround ping-pong 
modulation at 9600.  These are not compatible.  The highest speed 
the modems have in common is 2400bps.

>                                                       Of course it gets ugly
> then, because I think I get a 2400/ARQ connection--and the other end thinks
> differently--results in a hang.

That shouldn't happen.  Both modems support V.42 and V.42bis, so you 
should get a good connection.  Your VSM9600 registers are set 
properly for a V.42bis connection, so I don't have any good ideas.

Have you tried setting &Q0, and see if you can connect to that modem 
in plain old 2400 without error control?  It would be an interesting 
experiment.  If that works, try MNP alone, by setting &Q5 S36=4 
S46=0 S48=128 (for MNP4; try S46=2 for MNP5).  If that works, try
forcing V.42 alone, using &Q5 S36=0 S46=136 S48=7 (or S46=138 for 
V.42bis).  Let me know what you find.

	-- Toby
-- 
Toby Nixon, Principal Engineer    | Voice   +1-404-840-9200  Telex 151243420
Hayes Microcomputer Products Inc. | Fax     +1-404-447-0178  CIS   70271,404
P.O. Box 105203                   | UUCP uunet!hayes!tnixon  AT&T    !tnixon
Atlanta, Georgia  30348  USA      | Internet       hayes!tnixon@uunet.uu.net