Loading ...
Sorry, an error occurred while loading the content.
 

WSJT-X v1.1 beta release, r3487: Unexpected JT65A decodes

Expand Messages
  • g4wjs
    Hi, I received the following while trying to work CT3IB. The DTs look impossible to me and the messages possibly garbled. 1400 -4 1.7 743 # CT3IB RX1AB 73
    Message 1 of 4 , Jul 16, 2013
      Hi,

      I received the following while trying to work CT3IB. The DTs look impossible to me and the messages possibly garbled.

      1400 -4 1.7 743 # CT3IB RX1AB 73

      1400 -1 8.7 1521 # CQ CT3IB IM12

      1402 -1 4.7 1456 # CQ CT3IB IM12

      1404 -1 1.1 1456 # G4WJS CT3IB

      1406 -1 1.2 1457 # G4WJS CT3IB

      1408 -1 1.2 1457 # TU BILL 73

      73
      Bill.
    • Joe Taylor
      ... 1. Reports of what you believe to be abnormal decoder behavior can be useful, but only if you save the *.wav file and send it to us. 2. The messages you
      Message 2 of 4 , Jul 16, 2013
        g4wjs wrote:
        > I received the following while trying to work CT3IB. The DTs look impossible to me and the messages possibly garbled.
        >
        > 1400 -4 1.7 743 # CT3IB RX1AB 73
        > 1400 -1 8.7 1521 # CQ CT3IB IM12
        > 1402 -1 4.7 1456 # CQ CT3IB IM12
        > 1404 -1 1.1 1456 # G4WJS CT3IB
        > 1406 -1 1.2 1457 # G4WJS CT3IB
        > 1408 -1 1.2 1457 # TU BILL 73

        1. Reports of what you believe to be abnormal decoder behavior can be
        useful, but only if you save the *.wav file and send it to us.

        2. The messages you show are surely decoded exactly as they were sent.

        3. Looks like someone adjusted his computer clock?

        -- Joe, K1JT
      • Joe Taylor
        Hi Bill, ... Thanks for sending the *.wav files. The waterfall displays for these recordings make it clear that the CT3IB at 1400 signal does, indeed, start
        Message 3 of 4 , Jul 16, 2013
          Hi Bill,

          On 7/16/2013 10:12 AM, g4wjs wrote:
          > I received the following while trying to work CT3IB. The DTs look impossible to me and the messages possibly garbled.
          > 1400 -4 1.7 743 # CT3IB RX1AB 73
          > 1400 -1 8.7 1521 # CQ CT3IB IM12
          > 1402 -1 4.7 1456 # CQ CT3IB IM12
          > 1404 -1 1.1 1456 # G4WJS CT3IB
          > 1406 -1 1.2 1457 # G4WJS CT3IB
          > 1408 -1 1.2 1457 # TU BILL 73

          Thanks for sending the *.wav files.

          The waterfall displays for these recordings make it clear that the CT3IB
          at 1400 signal does, indeed, start nearly 9 seconds late (according to
          your computer), and the one at 1402 nearly 5 seconds late. (Set N Avg =
          1 for highest available time resolution.)

          He's probably using JT65-HF, which would send the message
          "G4WJS CT3IB" (two blanks between calls) as free text. The message is
          received by WSJT-X exactly as it was sent.

          WSJT-X is slightly smarter and would recognize that the two blanks were
          not really intended. The extra blank would be removed and the message
          sent as "G4WJS CT3IB", a normal JT65 structured (as opposed to free
          text) message.

          Nothing about the messages you received is garbled, and there's no
          indication of abnormal behavior of WSJT-X.

          -- Joe, K1JT
        • Joe Taylor
          ... The meaning of DT is the same as it has always been. It s the time offset between the sync pattern in a received signal and its expected time. The most
          Message 4 of 4 , Jul 16, 2013
            G4WJS wrote:
            > OK, I thought that the DT was a time offset of his clock not a start
            > time. So can I assume that JT65A can decode even with large time offsets
            > or is the meaning of DT different between JT65 and JT9?

            The meaning of DT is the same as it has always been. It's the time
            offset between the sync pattern in a received signal and its expected
            time. The most likely cause of non-zero DT is a difference between
            computer clocks at the transmitting and receiving stations.

            >> He's probably using JT65-HF, which would send the message
            >> "G4WJS CT3IB" (two blanks between calls) as free text. The message is
            >> received by WSJT-X exactly as it was sent.
            >>
            > OK, I never sent a report to him but he signed with a 73. I had assumed
            > that he had sent a report but it got mistranslated. From what you say he
            > must have given up on the QSO.

            He probably *thought* he sent you a report; but because of the two
            successive blanks JT65-HF sent his message as free text. The report
            extended the message beyond 13 characters, so it had to be truncated. A
            report was not actually sent.

            > OK, that's good. Clearly the JT65 protocol isn't as strict as JT9.

            No, that's not the right conclusion. We are talking *only* about JT65
            here -- the mode you were using.

            The JT65 protocol allows for both structured and free text messages.

            Because of two successive blanks between callsigns, JT65-HF decided the
            message composed by CT3IB must be free text. Characters beyond the 13th
            were therefore not sent.

            WSJT-X would have recognized that the two blanks were probably not
            intended. It would remove the extra blank and send the message as a
            standard, structured message -- presumably, including a signal report
            that he entered.

            -- 73, Joe, K1JT
          Your message has been successfully submitted and would be delivered to recipients shortly.