This is
Do you have a modem-related question?
Modemsite's free public modem board: Q&A - Helping you get the best performance from your modem
 Troubleshooting News Technical Search
 Home Forum 56 Premium Site Map
Home News & Updates    V.92 Special Report USR V.92 Bug

USR V.92 Bugs Updated 10-May-01

When a USR V.92 modem calls a V.92 server that does not support PCM upstream, the USR modem will ignore the server's indication PCM upstream is not supported; without disabling V.92 on the USR (+PIG=1), the connection will fail.

In addition to the bug reported below, it appears that some people who attempt to upgrade their modem are having a more serious problem: the flash does not complete properly and the modem is rendered unusable. If this occurs, the modem must be returned to U.S. Robotics for replacement and will require you to get a return authorization. This 'bug' apparently involves the newest Windows-based flash program (as opposed to the flash itself), and will not occur on all systems; however, there is no way to know if any particular system will fail or succeed with the flash - so unless you're willing to risk the destruction of your modem, I do not recommend you attempt to upgrade 5686-03 modems at this time.

A thread in comp.dcom.modems indicates some trouble and a bug in at least the firmware for the USR 5686-03 external modem. Remember: USR with hardware-controller modems, the upgrade involves flashing memory that holds the software that makes the modem work. USR still does not provide any easy means to flash your modem back to a previous version. So if you flash and you have trouble, you're really in trouble! (Also see: Flashing back a Sportster)

BUG: Apparently, the diagnostic report on the 5686-03 will indicate x2 modulation when in fact, a V.90 connection is made:

U.S. Robotics 56K FAX EXT 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)      8/2
Recv/Xmit Level  (-dBm)     24.5/17.5
Near Echo Loss   (dB)       28.6
Far Echo Loss    (dB)       0
Carrier Offset   (Hz)       0
Round Trip Delay (msec)     7
Timing Offset    (ppm)      1368
SNR              (dB)       43.2
Speed Shifts Up/Down/Null   2/1
Status :                    uu,5,12N,18.7,-7,1N,000,46.4,-51.3,52

Note however that the Status line returns diagnostic in the V.90 format. (When an x2 connection is made with a USR modem, the status line is in a binary-coded 0/1 format. See: 3Com/USR Diagnostics.) Also note that the status line includes an additional value compared to the V.90 modem.

This bug may not affect the operation of the modem, and when the connection is actually made, V.90 is reported (this can be verified by looking at your modemlog). Recv: <cr><lf>CONNECT 41333/ARQ/V90/LAPM/V42BIS<cr><lf>

Trouble: The real problem, at least for the user who started the thread this page is based upon, is the modem connected at 44,000-48,000bps before the upgrade, and now connects consistently at 41,333bps.

Home  |  Links  |  Send Feedback  |  Privacy Policy  | Report Broken Link
Legal Page  |  Author's Web Sites   |  Log In 1998-2022 v.Richard Gamberg. All rights reserved.