Olivia, Domino, and even MFSK16 can both fail to get good copy for the exact same reason. So CMSK is not alone for sure.
However, it's surprises me that MultiPSK seldom fails to decode an olivia signal while FLDigi (my fav program) does it much more often. I suspect there is something that can be done to mitigate that problem in the DSP code but NOT being a DSP programmer I have no idea what it is. Since I had already "calibrated" my sound card with checkSR program I wonder if Phil had done so too? One would assume that IF checkSR had calculated the correct offsets for my SignalLink USB (-14 for both input and output) that I should have been able to receive another calibrated station with a Sample rate of 7986 and NOT 8062. BUT then again, I don't know what is going on under the hood with CMSK program. When I set the sample rate does that apply to input and output? Rx and TX both?? Maybe not. I assume, maybe wrongly, that the CMSK is sort of a test program and NOT the final word on this mode. I wonder if the source code will be open source? I would like to think that it will be available to other programs but so far don't really know that much about it other than that when it works - it is kinda neat mode - 32wpm in 63hz signal with good FEC. ---Gary WB8ROL --- In digitalradio@yahoogroups.com, "g4ilo" <jul...@...> wrote: > > I don't think it's the only mode that is sensitive. I've noticed similar > problems with a few stations on Olivia. I'd get a few seconds of perfect copy > and then several more of complete garbage, even if their signal is strong. > I'm sure the problem isn't my end as I've had lengthy Olivia QSOs when the > signals were really weak and had perfect copy throughout. Perhaps both our > sound cards were identically malajusted? > > Julian, G4ILO > > --- In digitalradio@yahoogroups.com, jhhaynes@ wrote: > > > > Hmmm - if it's that sensitive to sample rate settings then we need > > some method of assuring that the sending station has a "correct" > > sample rate and then a way of setting the receiving station to > > agree with it. Unless someone is sending a long test message I can't > > see just randomly playing with the sample rate setting until I get > > good copy. > > >