1st 3Com v.90 Modem Problem

"First 3Com/USR V.90 modem problem"

2/25/98 updated & "solved" 2/26/98

Dean purchased a new USR V.90 modem and can no longer get a 56k rate with his local ISP (Prodigy Internet). He posted a message in newsgroup, and I've had some correspondence with him.

When he dials the USR BBS (888-877-9248) and IBM's x2 toll-free access number (800-590-4857), he gets a 54,666 connect - an excellent 56k rate (although the actual rate is 52,000 just a few seconds after connect). When he dials his local number, he gets v.34+ - no x2!

Solution: disable V.90 on the new modem (S32=66). Apparently there's x2 server equipment in the field that won't negotiate an x2 connection with V.90 enabled on USR's new modems.
> From: Robert Sanders <rsanders@mindspring.net>
> To: usr-tc@xmission.com
> Subject: Re: (usr-tc) new V.90 modem problem
> 
> I'd bet that the Splitrock modem refusing to negotiate X2 is a Bay
> 5399/8000 RAC. We have one running X2 code that will fallback to V.34
> if I call with a V.90 Sportster. If I disable V.90 on the Sportster
> (S32=66), it connects at X2. Go figure.
> 
> So, if you want to try Hypeterminal - give command AT S32=66
> then ATDT (local prodigy #)
> 
> If you get a 56k connection, you'll need to put the S32=66 in the 'extra
> settings' for your modem. My page at
> https://modemsite.com/56k/trouble.asp has instructions on how to do
> that.

It works I connect 54666 bps every time now ,thanks for the info.Here
are the test results from my local prodigy # again . the string I use
now in extra settings is AT&F1S32=66M0

at s32=66
OK
atdt 252-2170
CONNECT 54666/ARQ/x2/LAPM/V42BIS


Annex Command Line Interpreter * Copyright (C) 1988, 1997 Bay
Networks
Checking authorization, Please wait...
Port password:         
Here are the original messages:
From: Dean <doz911@hotmail.com>
Newsgroups: comp.dcom.modems
Subject: first problem with v.90-x2 modem
Date: Tue, 24 Feb 1998 16:04:11 -0600
Organization:   
Lines: 6
Message-ID: <34F343DB.1638@hotmail.com>
Reply-To: doz911@hotmail.com
NNTP-Posting-Host: 209.156.60.45


I just baught a new v.90-x2 USR internal faxmodem model 5687 and now I
only connect at 31200 to my isp Prodigy Internet .any one else have this
problem.I use the string At&F1 do I need some special string now?.Also I
had to set the jumpers because the default plug&play jumper settings
didn't work .when I turned on my computer I just got a blank screen .I
think that has to do with my old phoenix bios though.         
Reply I posted:
Could you post a paste of a Hyperterminal screen to the modem giving it
ATI6 I7 I11
after attempting a connection, please?
Also, try using Hyperterminal:
ATDT 1 888 877 9248
and paste your connect message for us to see.
Also, try
ATDT 1 800 590 4857
and paste your connect message for us to see.

Would much appreciate the information.
Thanks,
Richard         
Dean's reply:
I still always conect at 31200 and I used the inf from the cd-rom


ati6 i7 i11
U.S. Robotics 56K FAX INT Link Diagnostics...

Chars sent 0 Chars Received 0
Chars lost 0
Octets sent 0 Octets Received 0
Blocks sent 0 Blocks Received 0
Blocks resent 0

Retrains Requested 0 Retrains Granted 0
Line Reversals 0 Blers 0
Link Timeouts 0 Link Naks 0

Data Compression NONE
Equalization Long
Fallback Enabled
Last Call 00:00:00

Disconnect Reason is Keypress Abort

Configuration Profile...

Product type US/Canada Internal
Product ID: 00568700
Options V32bis,V.34+,x2,V.90
Fax Options Class 1/Class 2.0
Line Options Caller ID, Distinctive Ring
Clock Freq 92.0Mhz
EPROM 256k
RAM 32k

FLASH date 2/12/98
FLASH rev 4.7.31

DSP date 2/12/98
DSP rev 4.7.31

U.S. Robotics 56K FAX INT Link Diagnostics...


Modulation Unknown Speed
Carrier Freq (Hz)
Symbol Rate
Trellis Code
Nonlinear Encoding
Precoding
Shaping
Preemphasis (-dB)
Recv/Xmit Level (-dBm)
Near Echo Loss (dB)
Far Echo Loss (dB)
Carrier Offset (Hz)
Round Trip Delay (msec)
Timing Offset (ppm)
SNR (dB)
Speed Shifts Up/Down 0/0
Status :

OK

atdt 1 888 877 9248
CONNECT 54666/ARQ/x2/LAPM/V42BIS
CONNECT 33333/ARQ/X2/LAPM/V42BIS
CONNECT 115200 / 02-24-98 (21:22:20)
(Error Correcting Modem Detected)

USR Support BBS - Node 21 - Total Control Rack
PCBoard (R) v15.3/250 - Node 21

Testing your system capability...

Do you want graphics (Enter)=yes? ( )

ATDT 1 800 590 4857
CONNECT 54666/ARQ/x2/LAPM/V42BIS         
My follow-up:

The I6 & I11 would be more meaningful if you issue them after you connect to the toll-free #s - after the CONNECT, type +++ when you get OK, type ATH, then give it the AT I6 I11.... but, from what I see so far, it looks like you are [initially] connecting at VERY high speeds on toll circuit. Although, it may be dropping immediately. The USR msg shows 2 connects - 54666 & 33333... You should also try ATDT[your local PI #] and see what connect speed you get, wait about 15 seconds, give it the +++; then ATH, then AT I6 I11 to see what your speed is at time of disconnect.

Dean's follow-up:

ok i did the test to all the #s atdt 252-2170 CONNECT 31200/ARQ/V34/LAPM/V42BIS Annex Command Line Interpreter * Copyright (C) 1988, 1997 Bay Networks Checking authorization, Please wait... Port password: Password Entry Time Out CLI authorization impossible due to error: Connection timed out. NO CARRIER ath OK AT I6 I11 U.S. Robotics 56K FAX INT Link Diagnostics... Chars sent 19 Chars Received 229 Chars lost 0 Octets sent 19 Octets Received 229 Blocks sent 19 Blocks Received 5 Blocks resent 0 Retrains Requested 0 Retrains Granted 0 Line Reversals 0 Blers 0 Link Timeouts 0 Link Naks 0 Data Compression V42BIS 2048/16 Equalization Long Fallback Enabled Protocol LAPM Speed 31200/33600 Last Call 00:01:01 U.S. Robotics 56K FAX INT Link Diagnostics... Modulation V.34+ Carrier Freq (Hz) 1959/1959 Symbol Rate 3429/3429 Trellis Code 64S-4D/64S-4D Nonlinear Encoding ON/ON Precoding OFF/ON Shaping OFF/ON Preemphasis (-dB) 6/4 Recv/Xmit Level (-dBm) 18/9 Near Echo Loss (dB) 33 Far Echo Loss (dB) 55 Carrier Offset (Hz) 672 Round Trip Delay (msec) 5 Timing Offset (ppm) -320 SNR (dB) 40 Speed Shifts Up/Down 0/0 Status : OK ATDT 1 888 877 9248 CONNECT 54666/ARQ/x2/LAPM/V42BIS CONNECT 33333/ARQ/X2/LAPM/V42BIS CONNECT 115200 / 02-24-98 (23:47:43) (Error Correcting Modem Detected) USR Support BBS - Node 17 - Total Control Rack PCBoard (R) v15.3/250 - Node 17 Testing your system capability... aahyou want graphics (Enter)=yes? (++++++++ ati6 i7 i11 U.S. Robotics 56K FAX INT Link Diagnostics... Chars sent 28 Chars Received 354 Chars lost 0 Octets sent 21 Octets Received 322 Blocks sent 10 Blocks Received 21 Blocks resent 0 Retrains Requested 0 Retrains Granted 0 Line Reversals 0 Blers 0 Link Timeouts 0 Link Naks 0 Data Compression V42BIS 2048/32 Equalization Long Fallback Enabled Protocol LAPM/SREJ Speed 52000/31200 x2 Peak Speed 52000 Last Call 00:00:23 Disconnect Reason is DTR dropped Configuration Profile... Product type US/Canada Internal Product ID: 00568700 Options V32bis,V.34+,x2,V.90 Fax Options Class 1/Class 2.0 Line Options Caller ID, Distinctive Ring Clock Freq 92.0Mhz EPROM 256k RAM 32k FLASH date 2/12/98 FLASH rev 4.7.31 DSP date 2/12/98 DSP rev 4.7.31 U.S. Robotics 56K FAX INT Link Diagnostics... Modulation x2/V.34 Carrier Freq (Hz) None/1920 Symbol Rate 8000/3200 Trellis Code None/64S-4D Nonlinear Encoding None/ON Precoding None/ON Shaping ON/ON Preemphasis (-dB) 7/4 Recv/Xmit Level (-dBm) 17/10 Near Echo Loss (dB) 13 Far Echo Loss (dB) 0 Carrier Offset (Hz) NONE Round Trip Delay (msec) 18 Timing Offset (ppm) -251 SNR (dB) 0.0 Speed Shifts Up/Down 3/0 Status : 0000,1001,1010,1000,1010,1000,1000,1000 OK atdt 1 800 590 4857 CONNECT 54666/ARQ/x2/LAPM/V42BIS OK ath OK ati6 i7 i11 U.S. Robotics 56K FAX INT Link Diagnostics... Chars sent 3 Chars Received 0 Chars lost 0 Octets sent 3 Octets Received 0 Blocks sent 3 Blocks Received 0 Blocks resent 0 Retrains Requested 0 Retrains Granted 0 Line Reversals 0 Blers 0 Link Timeouts 0 Link Naks 0 Data Compression V42BIS 2048/32 Equalization Long Fallback Enabled Protocol LAPM/SREJ Speed 52000/31200 x2 Peak Speed 52000 Last Call 00:00:29 Disconnect Reason is Escape code Configuration Profile... Product type US/Canada Internal Product ID: 00568700 Options V32bis,V.34+,x2,V.90 Fax Options Class 1/Class 2.0 Line Options Caller ID, Distinctive Ring Clock Freq 92.0Mhz EPROM 256k RAM 32k FLASH date 2/12/98 FLASH rev 4.7.31 DSP date 2/12/98 DSP rev 4.7.31 U.S. Robotics 56K FAX INT Link Diagnostics... Modulation x2/V.34 Carrier Freq (Hz) None/1920 Symbol Rate 8000/3200 Trellis Code None/64S-4D Nonlinear Encoding None/ON Precoding None/ON Shaping ON/ON Preemphasis (-dB) 7/4 Recv/Xmit Level (-dBm) 16/10 Near Echo Loss (dB) 13 Far Echo Loss (dB) 0 Carrier Offset (Hz) NONE Round Trip Delay (msec) 41 Timing Offset (ppm) -238 SNR (dB) 0.0 Speed Shifts Up/Down 3/0 Status : 0000,1001,1010,1000,1000,1000,1010,1000 OK

My response:

Dean, It appears to me that you have an excellent telephone line and are getting VERY high speeds on the toll circuits. Your modem is failing to negotiate x2 on the local call. My guess is the local problem is either (a) the firmware in the server modem you are calling is not up-to-date (b) 3Com has failed to address the [widespread] problems with 56k LOCAL connections. [It appears to me your Prodigy node is on the Split Rock system - I've heard a lot of negative stuff about them...] I'd suggest you try other local 56k access #s to see if you get a 56k connection with them. If these #s are local to you - try them - I've found IBM (612)-943-5801 AOL (612)-502-9000 AOL (612)-630-0771 If any of those #s are local and you get 56k to them, it most likely is a problem with firmware/setup of the Prodigy node for you. If you don't get it with any of them, .... well.... Thanks for keeping me informed.

The documentation for x2 server firmware release 5.6 (the most recent released version for the ISPs is now 5.8), indicates a problem with older firmware if a 'new version' of client firmware (which you have) is released: "If any new version of x2 client code is released, the existing version of the Quad Modem cannot identify the new version of x2 client and will train to V.34 speeds. Quad Modem 5.5.7/5.6.7 checks the client's x2 version number and will train to the x2 version currentlyimplemented in the Quad Modem instead of falling back to V.34." The problem here becomes trying to find out what version your POP is using - if you ask Prodigy, the people you can get to probably don't know.