David, at 15:59 on Feb 04 1996, you wrote to Paul Edwards...

PE> No, I expect that David Begley would have managed to get at 
PE> least one 28800 connect though, instead of locked it at 
PE> 26400 first time every time. 

DD> Perhaps the Netcomm rendition of V.34 isn't that compatible with the USR 
DD> version (at the higher speed).  Netcomm to Netcomm gets 28,800 (so you 
DD> say), and USR to USR gets 28,800 (or faster). 

It's already common knowledge that Rockwell haven't implemented V.34 to the
letter, as shown by their use of different symbol rates and trellis coding,
which makes it pretty easy to point the finger at the guilty party when the
modems fail to connect properly.

DD> I remember when the Rockwell V.FC didn't play well with the USR's version 
DD> too.

And whose fault is that?  Rockwell changed their V.FC code in each of the
first 19 revisions of their RC288dpi chipset, so can you blame USR for not
being able to keep up?  Anyway, V.FC is essentially dead now.

Regards, Bill

--- Msgedsq/2 3.20