|
Hello Markus,
For 1:
extending the audio frequency range is possible, limitation is the fmax of 5kHz of the JT9 decoder. As a result the maximum audio frequency is 2.5kHz for JT9-2 and 1kHz for JT9-5. But even for this there is a workaroud possible by including pre-mixing of
the incoming audio before sending it to the JT9 decoder.
But before doing all this I just want to be sure that it is worth the effort:
- Do JT9-2 and JT9-5 "in the real world" (with ionoshepric instabilies, drifting transmitter and receivers) have a significant advantage over JT9?
- Is there sufficient interest of the LF/MF community to use thse modes?
For 2:
One of the advantages of JT9-2 is that you have 20 seconds time to type the free text (of maximum 13 characters). For JT9-5 you even have 50 seconds ;-)
But is should be possible to allow a "late start". It would just mean that the first symbols of the codes message are skiped at the cost of S/N loss. I think this is a better option that transmitting the entire codes message some seconds late (what probably
would lead to a fail to decode).
But as for 1, I would like to wait with this until I am sure that SlowJT9 is of any use.
3. I haven't had any crash so far, but it is better to be safe than sorry. Maybe I can add a "don't show this message again" option.
73, Rik ON7YD - OR7T
Hi Rik,
thanks for the great work on this nice piece of software!
Comments / questions from this end:
1. My LF equipment has a fixed 135.5 kHz LO, so audio will need to be around 1800 Hz or so. Such TX and RX frequencies can apparently be be set (albeit red), but the clickable spectrogram range ends hard at 1400 Hz. Would it be possible to allow a user defined
spectrogram shift? The alternative is external frequency conversion using SpecLab and virtual cables, but I'd rather avoid that due to extra latencies.
2. In the past I often found myself just missing the start of the next transmit slot (particularly when entering free text). Is it possible to start a transmission a few seconds late into the timeslot, or even change the message on the fly? I think the SNR
penalty of loosing a few symbols at the beginning may not be too severe.
3. The software seems to be fairly stable. Clicking the "save any work done on this computer" disclamer may not be necessary on every startup.
Best 73,
Markus (DF6NM)
-----Ursprüngliche Mitteilung-----
Von: Rik Strobbe < [email protected]>
An: [email protected] < [email protected]>; [email protected] < [email protected]>
Verschickt: Fr, 9. Nov. 2018 23:27
Betreff: LF: SlowJT9: 1st QSO and 2nd bug found
Dear all,
I just completed a JT9-2 QSO with Marco, DD7PC:
2141 -11 0.1 1104 2 OR7T DD7PC JN49
2145 -10 0.1 1106 2 OR7T DD7PC JN49
2153 -16 0.0 1105 2 OR7T DD7PC R-13
2157 -12 0.0 1105 2 OR7T DD7PC 73
2201 -16 0.1 1106 2 UFB RIK
During this QSO I discovered a 2nd bug:
As can be seen above the time (UTC) is given at odd minutes (eg. Marco's 21:40 - 21-42 transmission is reported 2141). This because the time is taken at the moment of decode.
For JT9(-1) this is not an issue, because decoding is in the same minute as the start of transmission, but for JT9-2 and JT9-5 it is because it results in a wrong calculation of the cycle (1st or 2nd). As the wrong cycle is given SlowJT9 will transmit in at
the same cycle as the station you received, very inconvenient!
I will fix this big after the weekend (sorry, to many other obligations on Saturday and Sunday).
For the time being the workaround is rather easy: if you get a decode you want to reply to (by double clicking on the call) just change the cycle manually (click once on the "TX 1st cycle" checkbox). The good thing is you have plenty of time to do so as there
are 20 seconds between the end of a transmission and the start of the next cycle.
To end with a positive note: during Marco's 2152-2154 transmission there was very deep QSB and his signal dissapeared from the waterfall display for almost 1 minute. But despite that still a good decode, just some dB weaker.
73, Rik ON7YD - OR7T
|
