Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!sun-barr!olivea!uunet!hayes!tnixon
From: tnixon@hayes.uucp
Newsgroups: comp.dcom.modems
Subject: Re: Hayes announces Ultra 144 and V.32bis upgrade
Message-ID: <4044.286b25d7@hayes.uucp>
Date: 28 Jun 91 12:04:39 GMT
References: <5234@orbit.cts.com>
Organization: Hayes Microcomputer Products, Norcross, GA
Lines: 47

In article <5234@orbit.cts.com>, kurt@pnet51.orb.mn.org (Kurt
Sletterdahl) writes: 

> Pardon my dumbness, but why don't V.32bis modems have a DTE interface speed of
> 57,600bps? I thought, with V.42bis data compression, that modems were supposed
> to have a DTE that was four times the modulation rate. Thanks in advance for
> your patient response.

I suspect that most manufacturers WILL support 57,600 DTE speed in 
their second generation V.32bis products.  But the first generation 
is, for the most part, an upgrade to a V.32/V.42bis product, many 
of which are already pushing the limits of the processing power of 
their microprocessors (the ones used for error control and data 
compression).  If you increase only the phone line speed by 50%, you
only increase the total number of interrupts per second that the
modem may have to process by about 12%, which is probably reachable
by most modems.  But if you increase the DTE speed to 57,600, the
number of interrupts per second then is 50% over the original
V.32/V.42bis modem, and few modems are built with this much excess
processor capacity (if they were, you'd be paying for power you
can't use). 

In my discussions with representatives of other modem manufacturers, 
the consensus seems to be that the primary goals of the first round 
of V.32bis product introductions is to (a) provide 14400bps 
synchronous capability, for full duplex dial-backup of V.33 leased
lines; (b) provide 1700 cps throughput for transfer of 
pre-compressed or uncompressible data; and (c) support a full 
3840cps throughput for a larger portion of compressible files, since 
you only need to achieve a 2.25-to-1 compression ratio with V.32bis 
instead of 3.33-to-1 with V.32 (it's not really 4-to-1, because some 
of the improvement comes from the stripping of start and stop bits).  
None of these require a 57,600bps DTE interface speed. Meeting these
goals quickly is considered to be more important than expending,
right now, the development funds to increase the speed of the entire
modem by 50%, especially since there are few applications and
systems that can really handle 5760 characters per second anyway,
and only a relatively small amount of the data typically transferred 
is really compressible at 3.33-to-1 by V.42bis.

	-- 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       tnixon%hayes@uunet.uu.net