[comp.dcom.modems] T2500 at Bell 103 modulation to Compuserve

gandrews@netcom.COM (Greg Andrews) (03/20/91)

In article <10@peanut.UUCP> dts@peanut.UUCP (Daniel Senie) writes:
>I've been trying to dial into Compuserve at 300 BPS, due to better rates.
>For certain uses on CSERVE, I find 300 is actually fast enough, since I 
>can read as fast as the characters come out that way.
>
>The problem I'm having is that with recent revs of the T2500 ROMs (6.01 and
>7.0) I have been unable to complete Bell 103 calls. The modems answer and
>think they're happy, but no amount of playing with baud rates/parity/etc.
>will make anything other than garbage come out.
>

Those are the versions of firmware that included V.42 as well as MNP.
Try turning off both error correction types (S95=0 S97=0 S106=0) and
see what happens.

I assume you're setting S50=1 to get 300 bps?  As long as your modem has
S90=0 (the default), you're getting a Bell 103 connection.

Without knowing other details (comm software, other register settings,
etc.), that's the best I can do.  Could your comm software be unable
to recognize the 300 bps connect message?


-- 
.------------------------------------------------------------------------.
|  Greg Andrews   |       UUCP: {apple,amdahl,claris}!netcom!gandrews    |
|                 |   Internet: gandrews@netcom.COM                      |
`------------------------------------------------------------------------'

dts@peanut.UUCP (Daniel Senie) (03/21/91)

I did some more experimenting and found some interesting results. The problem
with 300BPS with T2500 seems to be a problem with auto-reliable mode in MNP.
If I turn off MNP and V.42, everything worked fine. Also, using the modem
with V.42 turned on and falling back to no-protocol (S106=1, S95=0) worked
fine.

Setting up for MNP only with fallback to no-protocol (S106=0, S97=0, S95=2)
fails. In this mode, the tones sound like V.21 instead of Bell 103. I suspect
this is a bug, plain and simple. Incidentally, everything else in the modem
is set for Bell modes as opposed to CCITT modes (for speeds where that matters).

Thanks for the response. I'll see about filing a bug report with Telebit.

-- 
Daniel Senie               UUCP: uunet!lectroid!peanut!dts
Daniel Senie Consulting    ARPA: peanut!dts@lectroid.sw.stratus.com
48 Elm Street              CSRV: 74176,1347
Clinton, MA 01510          TEL.: 508 - 365 - 5352