BG> Nope, the anomaly occurs because the USR doesn't
BG> auto-baud its serial port on an ATZ command.

IS> Indeed, part of what ATZ does (and arguably _should_ do) is restore the
IS> AT&W's port rate, as it restores all the other saved &W settings, no?

The problem with that claim is that only the USR does it that way.

BG> Agreed, to do it any other way would be stupid.

Now try explaining why all the other modem manufacturers except
USR are doing it the 'stupid' way, and just what is actually
ACHIEVED by doing it the way USR does it. The short story is
that there is NO good reason why the ATZ command ALONE discards
the speed determined by the AT letters on that AT command.

BG> However, I'm still not convinced that what Paul's complaining about
BG> is buggy behaviour insofar as all USRs function in this manner and AFA I
BG> can determine, always have done (my 1988 Courier HST/2400 certainly does).

All that shows is that *USR* has been doing that for quite a while. Says
nothing useful about what all the other modem manufacturers have been doing
with the ATZ command OR why it makes sense to do it the way the USR does it.

IS> Agreed - that was my point, too.

BG> I've left in the quoting so that there can be no further misunderstandings
BG> re USR's requirement of saving the baud rate to NVRAM, BTW.

You still havent actually demonstrated
why it makes sense to do it that way tho.

BG> As it happens, I can sort-of see his point, but this relates
BG> more to USR's apparent failure to adequately document the
BG> necessity of using &W under the aforementioned circumstances.

Nope, you still have to actually justify why it makes the slightest
sense for USR to be doing it differently to everyone else. I'm not
convinced its actually possible to do that. And when its got a real
tendency to bite you on the bum when you least expect it, for no good
reason, I dont think you can actually show why it makes sense for USR
to be doing it differently to the way all the rest do it, including Hayes.

BG> All they say is that it must be done,

They DONT, thats ONLY for the &B2 mode, which isnt actually being used.

BG> no reason given (or none that I can find, at least).

The reason is pretty clear with the &B2 mode, after a non ARQ connect,
the modem has to use some speed for an ARQ connect, and clearly if the
modem hasnt received an AT command in the mean time, THAT speed has to
come from the NVRAM, coz there aint no other place to get it. Tho you
can still quibble with the use of the NVRAM speed for an ARQ connect
*IF* and AT command has been received by the modem in the mean time.

IS> Sounds like it's just not the right modem for his
IS> (or Rod's) personal tastes.  Can't please everyone, so ..

Nope, you have to justify why it makes any sense for USR to be
doing the ATZ different to ALL the other modem manufacturers,
let alone not even bothering do document that in the manual.

BG> Agreed.  :)

Wont wash Bill.
@EOT:

---