Paul, at 08:33 on Feb 14 1996, you wrote to Bill Grimsley...

BG> I believe USR's ATZ (which works exactly the same as ALL Hayes-type modems) 
BG> to be perfectly correct.  It simply restores ALL &W defaults as expected.

PE> No, that is NOT what is expected. 

Well, it happens to be exactly what I'd expect.

PE> What is expected is that it will auto-baud on the AT command.  That's what 
PE> auto-baud is all about.

On modems which don't save the baud rate to NVRAM, sure.  But the USR does.

PE> Unless USR have documented "Our modems do auto-baud, except on ATZ
PE> where we have totally fucked it up"?

I've already agreed that they could have documented that more clearly.

PE> You saw it too, Bill.  And if David D changes his binkley.cfg and
PE> mode to a different baud rate, he will see it too.  Unless he does
PE> the EXTRA step in changing the baud rate in NVRAM to work around
PE> the USR bug.

Just like USR tell you to, you mean?

PE> I ran for years with my Spirit never having to worry about the
PE> speed of last &W.  Remember I changed the speed to 19200 for
PE> some reason when trying to help you with your connects?  I
PE> didn't need to go &W.

That's because S2s and Rockwells don't save the baud rate to NVRAM !

BG> I don't accept that the failure of ATZ to auto-baud the modem is a fault or 
BG> a bug at all.

PE> Uh oh, Ostrich Impersonations!

As you wish, Paul.  Now pardon me, I have to go bury my head...

Regards, Bill

--- Msgedsq/2 3.20