> As far as I can see or know, Andy's repeater doesn't buffer or 'hold' any > data, it is purely passed directly and almost instantaneously into the TX > from the RX, after the comparator circuit that helps smooth off the 0's and > 1's.
Andy then is almost certainly transmitting "plain" FM, not GMSK-FM. This will work RF-wise moderatly well and is what NU5D did before he had an ICOM stack. The bandwidth will be somewhat greater, and the S/N performance worse, but the FEC will overcome some of this. Practically, I think the communciation range is nearly the same and the main difference will be bandwidth. NU5D should have done a side-by-side spectral comparison of his "FM DStar repeater" and the ICOM stack, but I don't think he ever had both going at the same time since he used the same duplexer and antennas. The ICOM stack may also not have the best designed GMSK modulation possible, so it may not be representative of GMSK performance per se. I think NU5D's main problem was transmitter key-up delay - enough of the sync data had already gone by that the receiving DStar radio had trouble locking up. 73 -- John
