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

Latest Obs, V4.3.0

Expand Messages
  • John, VK5PO
    Conducting further test this afternoon on 40m proved to be an almost non event. Myself and Paul (VK3DPW) played around for an hour or more, but only a few
    Message 1 of 14 , Jul 31, 2011
    • 0 Attachment
      Conducting further test this afternoon on 40m proved to be an almost non event.


      Myself and Paul  (VK3DPW) played around for an hour or more, but only a few lines of text were exchanged.

      For keybord to keyboard qso's Winmor kills V4.   Needless to say, winmor has a wider bandwidth than v4,

      and data speeds ramp up quite nicely.

      We acheived speeds up to 5.7kb per minute....testing by dumping several text files.  Its incredible.!!!!

      Without dumping a tub of water on V4 chat, it has a long way to go before it could be considered reliable.



      I hear you all ask... How do you do keyboard to keyboard qso's with winmor...


      Ask, and I shall tell...
      Email me direct.
      vk5po@...

      73, de VK5PO   John McRae Lewiston South Australia

    • Rick Muething
      John, I would very much like to see your debug logs of V4 (V4 TNC debug). and if possible your WINMOR TNC debug logs near the same time. Some points in your
      Message 2 of 14 , Jul 31, 2011
      • 0 Attachment
        John,
        I would very much like to see your debug logs of V4 (V4 TNC debug).  and if possible your WINMOR TNC debug logs near the same time.
         
        Some points in your comparisons of V4 and WINMOR
         
        1) WINMOR is quite a bit more mature than V4 (I should know, I wrote both protocols). WINMOR is now going on 4 years old.
        2) I know the coding used and theory well for both WINMOR and V4 and V4 should be more robust once debugged. Why?
                WINMOR uses 94 baud vs 47 baud.... That is faster of course but half the energy per bit which is a lot in terms of affecting the Bit Error Rate.
                WINMOR does not use Viterbi encoding with interleaving on 4FSK.... Only RS FEC.  That is more susceptible to burst errors.
                WINMOR takes up more than twice the bandwidth (higher speed of course) but also < half the signal to noise (same power...twice the noise bandwidth)
        3) WINMOR is optimized for sending binary data (e.g. files and attachments) and has a much higher latency (less “slick”) so it is not as convenient in a back-and-forth keyboard type connection.
        4) The concept behind V4 was make a protocol as robust as possible that fits into 200 Hz and runs fast enough to keep up with a fairly good typist (about 40-50 WPM). 
         
        Perhaps its time to re think both again and perhaps put a mode into WINMOR that better fits keyboarding. One thing that would be certain.  It would give me a lot more time to concentrate on the effort!
         
        Please send me the debug logs if you have them.  They tell far more than an email could on what is really happening inside the protocols.
         
        73,
        Rick Muething, KN6KB
         
         
        Sent: Sunday, July 31, 2011 4:33 AM
        Subject: [V4Protocol] Latest Obs, V4.3.0
         
         

        Conducting further test this afternoon on 40m proved to be an almost non event.


        Myself and Paul  (VK3DPW) played around for an hour or more, but only a few lines of text were exchanged.

        For keybord to keyboard qso's Winmor kills V4.   Needless to say, winmor has a wider bandwidth than v4,

        and data speeds ramp up quite nicely.

        We acheived speeds up to 5.7kb per minute....testing by dumping several text files.  Its incredible.!!!!

        Without dumping a tub of water on V4 chat, it has a long way to go before it could be considered reliable.



        I hear you all ask... How do you do keyboard to keyboard qso's with winmor...


        Ask, and I shall tell...
        Email me direct.
        vk5po@...

        73, de VK5PO   John McRae Lewiston South Australia



        No virus found in this message.
        Checked by AVG - www.avg.com
        Version: 10.0.1390 / Virus Database: 1518/3799 - Release Date: 07/30/11

      • Ron Wenig
        Standing by on 14.073 and on http://www.obriensweb.com/sked/ for chat. 73, Ron ny3j
        Message 3 of 14 , Jul 31, 2011
        • 0 Attachment
          Standing by on 14.073 and on http://www.obriensweb.com/sked/ for chat.

          73, Ron ny3j
        • Curt Nixon
          Hi Rick and all: Where would you like us to send debug logs? To You direct? or here? NY3J and I, (KU8L) tried several attempts this AM and the pair of logs
          Message 4 of 14 , Jul 31, 2011
          • 0 Attachment
            Hi Rick and all:

            Where would you like us to send debug logs? To You direct? or here?

            NY3J and I, (KU8L) tried several attempts this AM and the pair of logs
            might help. What we observed was this.

            1. Kind of shaky connects but a couple of successful connections
            established but dropped out while trying to send info. 20M and I
            suspect that the conditions were not so good.

            2. Moved to 40M and got nearly instant connects a couple of times. But
            while it looked like data was transferring, the process stopped part way
            thru and then ISS timed out while sending repeats.

            From Ron NY3J end, he observed apparently that his system was trying to
            send while I was sending. Ths sounds like a QSB or signal condition
            problem to me but perhaps you could confirm what was happening by seeing
            both logs.?

            I am trying to recruit a local to try this all out at solid NVIS-type
            conditions. both 40 and 20 here have been less than ideal lately.

            THanks

            Curt
            KU8L
            >
            >
          • Ron Wenig
            Curt, Thanks for the test on V4Chat. I was sending you acks while you were transmitting until I timed out and you kept trying to send. Rick, I guess this is
            Message 5 of 14 , Jul 31, 2011
            • 0 Attachment
              Curt,

              Thanks for the test on V4Chat.  I was sending you acks while you were transmitting until I timed out and you kept trying to send.  Rick, I guess this is normal under ARQ mode but Curt probably wasn't copying my acks.

              I don't have enough experience to know if this is normal operation.

              Ron ny3j

              On 7/31/2011 9:39 AM, Curt Nixon wrote:
               

              Hi Rick and all:

              Where would you like us to send debug logs? To You direct? or here?

              NY3J and I, (KU8L) tried several attempts this AM and the pair of logs
              might help. What we observed was this.

              1. Kind of shaky connects but a couple of successful connections
              established but dropped out while trying to send info. 20M and I
              suspect that the conditions were not so good.

              2. Moved to 40M and got nearly instant connects a couple of times. But
              while it looked like data was transferring, the process stopped part way
              thru and then ISS timed out while sending repeats.

              >From Ron NY3J end, he observed apparently that his system was trying to
              send while I was sending. Ths sounds like a QSB or signal condition
              problem to me but perhaps you could confirm what was happening by seeing
              both logs.?

              I am trying to recruit a local to try this all out at solid NVIS-type
              conditions. both 40 and 20 here have been less than ideal lately.

              THanks

              Curt
              KU8L
              >
              >

            • Rick Muething
              Ron, As I have said many times before and is in the Help....Send logs directly to me as attachments. Your word description doesn’t tell me much. Was the ISS
              Message 6 of 14 , Jul 31, 2011
              • 0 Attachment
                Ron,
                 
                As I have said many times before and is in the Help....Send logs directly to me as attachments.
                 
                Your word description doesn’t tell me much.  Was the ISS really sending you data or IDLs?  If data and his timeout is anything reasonable (say 60 or more seconds) it is unlikely it will time out.  Remember a connect request frame is DATA ...its just formatted in a particular way to cram in both call signs into 16 characters.  If you can CONNECT you have already sent one data frame and that had to be ACKed in less than 30 seconds. 
                 
                If the other side was sending IDLs it was because nothing was in their outbound queue (Did they type anything after connecting?  Did they transfer the typed data to the outbound queue with a <Ctrl> Cr?  If not the ISS will keep repeating IDLs until it times out.  I don’t know how to fix that...maybe just make up some text and send the first few lines of the Tale of Two Cities??? 
                 
                The logs tell me all ...even the quality of the decode.  Without them debugging or improving is almost impossible.
                 
                Rick KN6KB
                 
              • John, VK5PO
                Message 7 of 14 , Jul 31, 2011
                • 0 Attachment
                  On 31/07/2011 11:09 PM, Curt Nixon wrote:
                  > Hi Rick and all:
                  >
                  > Where would you like us to send debug logs? To You direct? or here?
                  >
                  > NY3J and I, (KU8L) tried several attempts this AM and the pair of logs
                  > might help. What we observed was this.
                  >
                  > 1. Kind of shaky connects but a couple of successful connections
                  > established but dropped out while trying to send info. 20M and I
                  > suspect that the conditions were not so good.
                  >
                  > 2. Moved to 40M and got nearly instant connects a couple of times. But
                  > while it looked like data was transferring, the process stopped part way
                  > thru and then ISS timed out while sending repeats.
                  >
                  > From Ron NY3J end, he observed apparently that his system was trying to
                  > send while I was sending. Ths sounds like a QSB or signal condition
                  > problem to me but perhaps you could confirm what was happening by seeing
                  > both logs.?
                  >
                  > I am trying to recruit a local to try this all out at solid NVIS-type
                  > conditions. both 40 and 20 here have been less than ideal lately.
                  >
                  > THanks
                  >
                  > Curt
                  > KU8L
                  >>
                  >
                  > ------------------------------------
                  >
                  > Yahoo! Groups Links
                  >
                  >
                  >
                  >
                Your message has been successfully submitted and would be delivered to recipients shortly.