In looking at the connections on my chassis, I noticed something that seems very unlikely. It's that all of the non-x2/v.90 modem connections always have a rcv trellis rate of 64S-4D while they have a transmit trellis rate that is almost (9 in 10) anything _but_ 64S-4D. I don't have any quad modems any more, but if I did it would be interesting to see if they had the same anomoly. I'm not a modem designer, but I'm going to guess that the 64 is the number on points in the constellation. If that's the case, I'd imagine that the forward channel, the one that is from the ISP to the end user, the one that is digital, would be the one with the best performance, least phase noise, and have the most constellation points. It seems odd that in fact the reverse is true, that the receive trellis rate is always pegged at 64S-4D while the transmit trellis rate is variable, but almost never 64S-4D! Another observation I made is that the S55 register has absolutely no effect on the selected trellis rates in either direction. You can set S55 to any value and the rates will not be negotiated any differently. Is it possible some debugging switch got left thrown in the DSP code to account for this? Thanks, -- Aaron Nabil - To unsubscribe to usr-tc, send an email to "majordomo@xmission.com" with "unsubscribe usr-tc" in the body of the message. For information on digests or retrieving files and old messages send "help" to the same address. Do not use quotes in your message.