PE>>> The modem was never locked at 57600, it merely happened to 
PE>>> be the speed the com port was set at at the last &W.  

DD>>> That is how you lock the speed into the Courier . . . .

PE>> No it isn't.  You lock the speed by setting your comms 
PE>> software to a particular speed, whilst making sure that 
PE>> Courier has &B1 in it (not necessarily in NVRAM either).  
PE>> Try it on an outgoing call and you'll find it works a 
PE>> treat.  You can do it on an ingoing call too, just make 
PE>> sure you have done at ATS0=0 or some other harmless 
PE>> command.

DD>> For Christ's sake! check page C-6 in the manual.  Read the 
DD>> description of ATI5 - it displays the NVRAM settings.  One 
DD>> of these is the speed that the modem will use to 
DD>> communicate incomming calls to the computer.  It is set 
DD>> with AT&W.

PE> In the absence of autobauding, that is the default speed 
PE> used. Like I said, do the test above, and you will find that 
PE> that speed is NOT used.  Try it yourself, on an outgoing 
PE> call.  Unless you're going to say that THAT is the bug in 
PE> the USR, that it doesn't always communicate at the speed 
PE> stored in NVRAM?  Come on, answer this straight question.

DD> The ATZ in the INIT string will reset ALL memory settings to those stored 
DD> in NVRAM.

Come on, answer the straight question.  Do the last &W at 57600,
set your comms program to 38400, and then do an ATDT and call
me or someone else, and then see if the modem communicates at
57600 like you say it is meant to, according to this mysterious
quote which you didn't bother quoting.

PE>> Oh, and then explain why the way the USR is doing it makes 
PE>> any sense whatsoever, and how it could possibly be better 
PE>> than the way Rockwell do it.  ie, the USR is fucked by 
PE>> design.

DD>> Different = fucked?

PE> No "implemented in a manner of no use to anyone" = fucked.

DD> Works EXACTLY as documented.

It's documented that ATZ does not autobaud?  Where is that pray
tell?  BFN.  Paul.
@EOT:

---