Path: utzoo!attcan!uunet!ogicse!hakanson
From: hakanson@ogicse.ogi.edu (Marion Hakanson)
Newsgroups: comp.dcom.modems
Subject: Re: Searching for inexpensive V.32 9600 bps modems (was cheap telebits)...
Message-ID: <7734@ogicse.ogi.edu>
Date: 5 Mar 90 19:15:28 GMT
References: <8398@potomac.ads.com> <1990Mar2.220228.2836@ux1.cso.uiuc.edu> <1990Mar3.164553.29588@virtech.uucp> <9977@cbmvax.commodore.com> <8612@xenna.Xylogics.COM>
Organization: Oregon Graduate Institute (formerly OGC), Beaverton, OR
Lines: 21

In article <8612@xenna.Xylogics.COM> loverso@Xylogics.COM (John Robert LoVerso) writes:
>. . .
>The really expensive part is the bit-pump.  I.e., the T2500 (and newer-style
>TB+) have a considerable amount of computing power.  And this is why the
>T1000 can't do better than 9600bps; it just doesn't have the power!

You are correct that the T1000 cannot go as fast as the TB+ or T2500,
but you are mistaken about 9600bps being the T1000's top speed.  If you
check the appropriate S-register (there's one for xmit and one for rcv
bit rates), a clean line will report the maximum of 11600 (or in that
ballpark, I forget the exact numbers).

To verify their claim, I measured throughput (using zmodem) with the
serial line set at 9600bps and compared it to a transfer with the line
set at 19200bps (both trials were with the T1000 connected to a TB+
running at 19200bps on the other end).  The higher line speed yielded
higher throughput.  So don't sell the T1000 short on that count.

-- 
Marion Hakanson         Domain: hakanson@cse.ogi.edu
                        UUCP  : {hp-pcd,tektronix}!ogicse!hakanson