-> USR chooses the 3200 symbol rate when answering a call from a modem
-> that doesn't support the 2743 symbol rate and that indicates the same
-> maximum projected data rate for both the 3200 and 3429 symbol rates.
-> This does not match what other V.34 implementations do.  By choosing
-> the 3200 symbol rate instead of the 3429 symbol rate USRs are giving
-> up about 2dB of SNR.  On a lot of lines that is the same as giving up
-> one data rate.

This entire paragraph was "cut-n-pasted" from a message originally
written by Dan Moore of Supra, shouldn't you be giving him some credit
for it?  It contains some minor technical errors, but it explains things
pretty well for anyone who is armed with a copy of the V.34 spec.

-> However were you to examine your INFO1c and INFO1a sequences you will
-> see what has and what has not been deliberately done to deceptively
-> try and enhance USR's position in the market place.  Shame on Skokie!

What exactly has been done, and in what way does it "enhance USR's
position"?

Be careful, I *DO* know how to examine the INFO sequence data...

-> It's called Modem Apartheid.  ie USRs are deliberately configured to
-> recognise the colour of callers' money spent on modem in deciding
-> what performance to achieve.

This is completely false.  V.34 provides *NO* way of identifying the
manufacturer of the modem at other end.

USR's symbol rate/speed selection logic is designed to operate at the
fastest reliable rates under all circumstances, even if the remote modem
does something foolish or out-of-spec.  It does not "penalize" other
manufacturers, and it is capable of operating reliably even when
connecting to some first-generation modems which are notorious for
handshaking bugs.

I would suggest you re-read Dan Moore's message, you seem to have
missed the important parts of it.   He does explain why this happens,
and it *isn't* a bug or secret plot.

Joe
--- FidoPCB v1.5 beta-'j'