PE>> And I presume you are using a Rockwell modem.  Yes, you are
PE>> experiencing what any decent implementation of auto-baud 
PE>> does. 

DD>> The USR Courier DOES "auto-baud" as well as any modem.  

PE> No it DOESN'T.  It fails to auto-baud on ATZ, making it 
PE> WORSE than a Rockwell.

DD> What's the point of continuong Paul?

Well I was actually hoping you would understand eventually that
it makes no sense to change baud rate on the terminal, no 
terminal expects that to happen.  You will notice that a couple
of people changed their mind on this issue after it was
explained a couple of times to them.

DD> You say it is a bug - I say the characteristic is documented.

Bug/design flaw, yes, EVEN if it was documented, which it isn't
anyway.

DD> You say it is f*cked by design, I say it is a perfectly usable and useful 
DD> modem.

I only said "f*cked by design" about the lack of sensible autobaud
on ATZ, I didn't write the whole modem off because of that.  It
most certainly is a usable and useful modem.  Even my Spirit fell
into that category, plenty of connects every day with it.  However,
both modems have warts, which I documented.  And now that I have
a Netcomm, I have documented 1 problem, and expect to be documenting
another in a day or so.  Both with USRs I might add.  Which does
not mean that it is a USR bug either.

DD> We don't agree, so that may as well be the end of the thread.

Yeah, if there's no chance of you ever understanding that the
Rockwell/rest-of-the-world method makes more sense than USR's
method, when you look at the fact that not one terminal or 
comms program in the world expects the baud rate to be changed
on it after issuing an ATZ, and can adapt after issuing the
command, then there is little point continuing.  I really
thought most people would be capable of reasoning this out for
themselves.  BFN.  Paul.
@EOT:

---