I have devised a workaround whereby I can force my modem (an ACF2
chipped Rockwelloid) to retrain using TIES (Time Independent Escape
Sequence). I create two blank email messages, both addressed to
myself, the first with a subject line of "+++ AT Q1 S2=93 M2 L3 O1",
the second with "]]]AT Q0 S2=43 M1 O". Upon sending the first message,
my modem switches to command mode, turns on the speaker, returns to
online mode, and then retrains. After 10 to 20 seconds, when the
retrain is complete, I send the second message to turn off the
speaker. I use Q1 in the first message to suppress the CONNECT message
so as to reduce the number of CRC errors in my ppplog. I use the S2
command to temporarily change escape codes, from "+" to "]" and then
back again, otherwise the email messages will not go through.
Note that the "O" and "O1" in the AT strings are ohs, not zeroes. Note
also that there is no space between the +++ and AT in the first
string. I have added it to enable me to post this message.
-- Franc Zabkar
Please remove one 's' from my address when replying by email.
--
Hooda Gest
"In a New York minute, everything can change..."
"Franc Zabkar" <fza...@optussnet.com.au> wrote in message
news:f6619vgvbtg01nsjc...@4ax.com...
>Why not just find out if the problem is inside or outside and get the telco
>to repair it?
Until now the problem has been very intermittent. As the results of
past service calls have been a crossed line (with another subscriber),
and pair gaining, I have been loathe to risk further degradation of my
service. In fact, I just recently downloaded Open Office (50MB)
without incident. Having said that, today I have experienced a
substantial worsening of my problems (due to audible line noise, ie
crackling), with several disconnections due to retrain failures, even
in the absence of incoming calls.
Anyway, I have placed a service call to my telco whose testing has
fortunately found an outside fault. Hopefully the problem will be
rectified properly this time.
Problems aside, I believe my technique can still be useful in certain
cases. For example, during a long download I notice that my modem's
throughput sometimes slowly decreases, with the post-call stats
showing a corresponding downward speedshift. I don't believe this is a
thermal issue because if I disconnect and reconnect, the throughput
returns to normal. In fact, after the long 50MB download I alluded to
above, I was able to force an upshift using my technique.
>--
>Hooda Gest
>"In a New York minute, everything can change..."
>
>
>"Franc Zabkar" <fza...@optussnet.com.au> wrote in message
>news:f6619vgvbtg01nsjc...@4ax.com...
>> I have an ongoing problem with crosstalk between my voice and data
>> lines. There is occasional hum on my data line which suggests a water
>> problem, as was previously diagnosed by my telco. If my voice line
>> rings whilst I'm connected on my data line, the crosstalk is often
>> severe enough to cause a disconnect. At other times, especially if I'm
>> quick to answer the call, the modem shifts to a lower speed, often to
>> V.34, where it stays for the remainder of the dialup session.
>> Curiously, the crosstalk is not loud enough for me to hear.