You are comparing apples to oranges. A -17 FH signal is not the same as a -17 SF signal. Per the FH docs "for N Slots = 1, 2, 3, 4, and 5 the average power per signal will be down from P0 by about 0, 6, 9.5, 12, and 14 dB, respectively". So you have a 9.5 dB hit per stream for three streams. Or to put it another way, a single stream would be 9.5 dB stronger. SF is a single stream transmission so it is at 0 dB all the time. This is covered in greater detail in the developer docs but there is a net "gain" for SF over a certain number of FH streams.
I have what you describe as a small station. Again, I have N5J contacts in the log that I would not have if they were running multi-stream FH. It was six days in before I heard and worked N5J on 10 and 12 SF but patience and good software tools got me to the right spots (no pun intended) at the right time. 73 -Jim NU0C On Sat, 24 Aug 2024 17:12:25 -0400 Gary via wsjt-devel <wsjt-devel@lists.sourceforge.net> wrote: > Exactly. The reduced overall sensitivity compared to 3 stream regular > F/H was at play in my case. On 3 streams, they decoded on regular F/H > at -17. Weak, but well within the -24db decoder limit. A -17db signal > on Super F/H is outside of that decoder's limit of -16. It makes a > difference for those of us forced to use marginal stations. Also, the > 1500hz bandwidth makes it more prone to QRM, as I see it. > > I hope that DXPeditions from now on will not be using S/FH exclusively, > or myself and a lot of other small stations are going to be left out, > unless the SFX runs some power. Like 40-100w at least. I heard that > N5J was only running 10w. That could have been a big part of the > problem, if true. > > 73 de Gary - W9BS > > On 8/24/2024 4:50 PM, Jim Shorney via wsjt-devel wrote: > > It is a matter of deciBels as has been documented. Each time a stream is > > added in multi-stream there is a multi-dB power hit on all streams. This is > > what SF avoids but TANSTAAFL. You do lose some sensitivity but at that > > point you would probably not be hearing multi-stream FH either. It seems > > like you were a victim of propagation. If you were copying three streams of > > FH you probably could have copied SF under the same conditions. Just a > > guess as there are multiple factors in play. I was decoding SF more > > reliably than anything over 2 FH streams. It is rare that I am able to copy > > three or more FH streams when prop is poor. Signals have to be very good. > > That is not the fault of the software, it is the limitations imposed by my > > antenna system and local noise floor vs. propagation conditions. > > > > 73 > > > > -Jim > > NU0C > > > > On Sat, 24 Aug 2024 12:10:02 -0400 > > Gary via wsjt-devel <wsjt-devel@lists.sourceforge.net> wrote: > > > >> Perhaps it is a problem when running 5 streams. But my experience was > >> just the opposite. I could never hear them on SFox on any band. > >> Admittedly, I run a low profile station with an attic antennas. From > >> this perspective, SFox was a huge disappointment. It wasn't until they > >> switched to regular fox near the end of their deployment that I was able > >> to finally work them on 17 and 30. They were running 3 streams at the > >> time. Never decoded them once on Superfox on any band, after > >> frustratingly listening for them all week. > >> > >> 73 de Gary - W9BS > >> > >> On 8/24/2024 11:51 AM, Jim Shorney via wsjt-devel wrote: > >>> But the same can be said of multi-stream F/H. I can say with certainty > >>> that I have Super contacts in the log with N5J that I would not have if > >>> they had been using multi-stream. > >>> > >>> On Sat, 24 Aug 2024 13:48:02 +0000 (UTC) > >>> Greg Chartrand via wsjt-devel <wsjt-devel@lists.sourceforge.net> wrote: > >>> > >>>> The bottom line is SF mode IS self defeating when/where conditions are > >>>> marginal AND they are always marginal somewhere in the world > >> > >> _______________________________________________ > >> wsjt-devel mailing list > >> wsjt-devel@lists.sourceforge.net > >> https://lists.sourceforge.net/lists/listinfo/wsjt-devel > > > > > > > _______________________________________________ > wsjt-devel mailing list > wsjt-devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/wsjt-devel -- 73 -Jim NU0C _______________________________________________ wsjt-devel mailing list wsjt-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/wsjt-devel
