On Tue, Jan 14, 2020 at 11:32 PM David Gilbert <[email protected]>
wrote:

>
> What would that be?  FT8/FT4 uses a better detection scheme than RTTY
> precisely because of the constraints that FT8/FT4 require.  Those
> constraints are what allow the better decoding ... there is no "magic"
> involved here.
>
> No one suggested magic. The program has MFSK as one mode option. (It has
many modes, not just FT8.)  RTTY uses BFSK, and BFSK is a subset of MFSK.
In order to discriminate which tone is being sent, there is an algorithm
the program uses. Some algorithms are better than others. If you look at
the various RTTY programs available, you will see a considerable variation
in error rate vs. S/N. DM780, for example, requires an S5 signal to get
reasonably good print, and around S9 for 100% print. Others are slightly
better. My thought was that the frequency discrimination algorithm in
WSJT-X is better in the presence of noise than most of the other programs
available. (I've hear the word, "bins" used, so I suspect that FFT is at
the heart of the discriminator.) Using this as the heart of a RTTY program
might provide an improvement in performance on RTTY in the presence of
noise. I understand that RTTY doesn't use or allow the redundancy and error
correction techniques found in FT8, so the performance will never approach
that of FT8. But it could be improved. Many years ago, I used a T/U with
tuned circuits using 88 mH toroids, and Kleinschmidt teletype equipment,
and could get very good print even when the signal faded below my ability
to hear it in the headphones. I haven't found a single program currently
available that approaches that.

I understand that there is finite time available for programmers. For me,
as a user, WSJT-X works splendidly, and short of printing out QSL cards,
putting stamps on them, and taking them to the post office, I don't know
what else it could do for my operating practice. Since WSJT-X is now used
in RTTY contests, it seems like it would be a natural fit to add a RTTY
window to the program. The UI is completely different, but much simpler for
RTTY. An alternative might be for someone who already knows the code in
WSJT-X to write a small external program that is called when "RTTY" is
selected as the mode in WSJT-X. Since WSJT-X is used in RTTY contests, this
would make it more convenient. Is WSJT-X more of a test bed for
experimenting with new communications modes or a T/U program intended for
daily and contest use by hams? Of course, the developers make the choice as
to the mix. I was simply stating my preference as one ham among many.

73,
Frank
KF6E
_______________________________________________
wsjt-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Reply via email to