Of interest - I have turned up another interesting lead in the
"produces horrible results in one direction only" problem.

Sold one of my Hayes Optima 288 v34's to a mate, who had been running a VFC
on the same line.  He's getting the problem.

Punch line - it doesn't happen with a VFC.  It happens with the v34, if and
ONLY if, the calling tones are enabled.

Thinks.  If the blighters are fitting fax processors at trunk exchanges
generally, v34 may be triggering malfunctions in a way that VFC doesn't,
via the calling tones.  (Or perhaps a bit of video processor gear?  It DOES
correlate to the date of starting serious pay TV installation.)

This correlates to v32 not causing the problem either.  Confirmed here -
the problem reappeared two days ago after a considerable absence - so I set
up a VFC Microcom on the line.  Sure enough - it plows through the horror
time with no problem.  The USR v34+ had a useful connection life of under
five minutes whan I was running the A/B tests - the VFC then held for
nearly two hours after I'd swapped back and forth three times and confirmed
VFC no drop, v34 dropout, switch back...   I didn't have enough time to run
tests vis do useful work to try reconfiguring the USR to no calling tones -
not tested.

My immediate fix is to go back to VFC on my dial-out line.

What the $#{at}# is going on?   And why?

Sigh.  Dave Hatch

--- Msgedsq 3.20