I like to use the narrowest possible bandwidth mode possible, but sometimes conditions are so bad, and there are unlikely to be many other stations that going to a wide bandwidth 1000 or even 2000 Hz mode I think it can be justified.
So far, the consensus seems to be that Olivia will beat all other digital modes assuming you don't consider the bandwidth. While I consider 1000/32 Olivia, with its ~24 wpm to be too slow for practical keyboard use, I can see where it is better than not getting anything through at all. Olivia 500/16, at under 20 wpm, is even slower than 1000/32 and 500/8 is just under 30 wpm so while relatively slow, it is a little faster than the other two mentioned modes. The first two use 31.25 baud vs. the 62.5 baud for the 500/8 which may be pushing the limit for multipath. It would have been interesting if we could have gotten one of the Olivia modes to work better than CW which was extremely difficult and I was not able to copy much better than 70% or so. When conditions deteriorate like they did, it was not possible to communicate what alternative mode to switch over to. But, again, the problems that we are up against on the lower bands during warm weather are primarily the unrelenting QRN static crashes that blend into a continuous roar with a few spikes even higher than that. Except for a very few SSB stations, running full power, you don't hear that much on the lower bands with these conditions, without specialized antennas to improve S/N ratios, (beverage, flag, pennants, etc.) It is a lot to expect any non-ARQ mode to handle this kind of environment since even if you have signals usually above the lowest S/N ratio of AWGN for that mode, since the QRN is not AWGN and the spikes will often cause hits that I doubt can be repaired by any of the modes, and that includes modes that spread out and make the data redundant. There is a limit to how much QRN is too much, and I think that I have already found that limit. We will have to try some of the Olivia modes again, even though they are slower when compared to DEX and MFSK. From previous tests I recall that MFSK typically outperformed Olivia but maybe my memory is faulty? I do like the more forgiving tuning accuracy with Olivia. The other station likes Hell modes, although I do not really care for this mode either, and it is also very slow at 25 wpm, but maybe we would be able to decode better than the machine? I would like to hear of other experiences for those who have tested the various modes and found what worked best for them under extreme QRN conditions. 73, Rick, KV9U expeditionradio wrote: >> 1000/32 would blow all the QRN away, methinks >> John >> VE5MU >> > > Hi John, > > Yes, I agree, Olivia 1000/32 is excellent, it decodes at -13dB SNR. > > Olivia 500/16 is about 1dB better than 1000/32. > Olivia 250/8 is about 2dB better than 1000/32. > > Here is a chart (courtesy of Patrick F6CTE) > http://www.hflink.com/olivia/#formats > > I've noticed that going from PSK31 to Olivia 500/16 is like turning on > a kilowatt amp. > > 73---Bonnie KQ6XA > > >
