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

Re: New poll for oliviadata

Expand Messages
  • jhaynesatalumni
    ... before they ... fun I would settle for a nice easy-to-install Olivia. I have Windows 95. I ve tried several of the things offered on the group but none of
    Message 1 of 6 , Feb 15, 2005
    • 0 Attachment
      --- In oliviadata@yahoogroups.com, "N7HIY" <N7HIY@c...> wrote:
      > think everyone is waiting for a nice easy to use GUI for Olivia
      before they
      > jump in but give Olivia Aid a try. it actually works quite well. a
      fun

      I would settle for a nice easy-to-install Olivia. I have Windows 95.
      I've tried several of the things offered on the group but none of them
      have worked. I haven't tried downloading and installing all of
      Cygwin, because I have only a dialup connection and the Win95 machine
      is not on it. Maybe Olivia will never work on Win 95.

      I could also try running it under Linux, if suitably motivated.
      The only trouble with Linux is that right now the radio that is
      connected to the sound card is a Kachina, and there is no Linux
      program to run the Kachina.
    • Paul L Schmidt
      I ve got it running on a Panasonic CF-25 laptop (with crappy 8-bit sound card and all) under Windows 95. Attached is a zip file that should get you up and
      Message 2 of 6 , Feb 15, 2005
      • 0 Attachment
        I've got it running on a Panasonic CF-25 laptop (with crappy 8-bit
        sound card and all) under Windows 95. Attached is a zip file that
        should get you up and running -- it contains:
        cygwin/cygwin.reg
        cygwin/bin/cygwin1.dll
        cygwin/bin/cygncurses-8.dll
        cygwin/usr/share/terminfo/c/cygwin

        Unzip it in C:\, and use regedit to import the cygwin.reg file into
        the registry. The .reg file is in the old format so Win95 will read
        it fine.

        To get Olivia to run, there must be a path to the cygwin DLL's. You
        can either add c:\cygwin\bin to the path, or just load the Olivia
        binaries into c:\cygwin\bin to begin with and execute it that way.



        jhaynesatalumni wrote:
        >
        > I would settle for a nice easy-to-install Olivia. I have Windows 95.
        > I've tried several of the things offered on the group but none of them
        > have worked. I haven't tried downloading and installing all of
        > Cygwin, because I have only a dialup connection and the Win95 machine
        > is not on it. Maybe Olivia will never work on Win 95.
        >
        > I could also try running it under Linux, if suitably motivated.
        > The only trouble with Linux is that right now the radio that is
        > connected to the sound card is a Kachina, and there is no Linux
        > program to run the Kachina.
        >
        >
      • Rick Williams
        ... and polar ... Could someone outline how much difference there is between Olivia and MT63 except that Olivia allows for changing the parameters? When I did
        Message 3 of 6 , Feb 15, 2005
        • 0 Attachment
          --- In oliviadata@yahoogroups.com, "N7HIY" <N7HIY@c...> wrote:
          > a fun
          > mode and good rag chew potential with no repeats. looking for my first
          > polar zone contact to study the effects of polar phase distortion
          and polar
          > zone fading on Olivia.

          Could someone outline how much difference there is between Olivia and
          MT63 except that Olivia allows for changing the parameters?

          When I did try it, it did not seem as robust as I expected it to when
          we lowered the number of tones. This was across the state in the
          evening on 80 meters. Seems like 8 tones gave us problems.

          From my perspective, the throughput is too slow in the default speed.

          Has anyone been able to compare it to MFSK16 which thus far seems to
          be the best mode for very weak signals and moderate keyboard speed?

          I am hoping to hear some return comments from the FCC to those who
          received the notices so that I feel comfortable operating the mode again.

          SCAMP has been taking up some time too. No one can say that mode is
          weak signal though:)

          73,

          Rick, KV9U

          73,

          Rick, KV9U
        • Paul L Schmidt
          ... Both MT63 and Olivia use a Walsh-Hadamard function for FEC (as opposed to MFSK16 which uses Viterbi FEC) - for MT63 and Olivia, a 7-bit input word produces
          Message 4 of 6 , Feb 16, 2005
          • 0 Attachment
            Rick Williams wrote:

            > Could someone outline how much difference there is between Olivia and
            > MT63 except that Olivia allows for changing the parameters?

            Both MT63 and Olivia use a Walsh-Hadamard function for FEC (as opposed
            to MFSK16 which uses Viterbi FEC) - for MT63 and Olivia, a 7-bit input
            word produces an array of 64 bits which describe the 7-bit word; for
            MFSK16, the Viterbi encoding expands a 4-bit word to 8 bits.

            In all cases, the encoded stream of bits is interleaved -- MT63 uses a
            straightforward interleaver that spreads the 64 bits across both time
            and frequency so that it takes either 32 or 64 symbol intervals to get
            them all out. Olivia takes blocks of encoded characters (5 for 32-tone
            MFSK, 6 for 64-tone MFSK, etc) and spreads them out over time to produce
            a symbol to transmit during each time interval. MFSK16 uses "diagonal
            interleaving" which places the bit patterns in arrays by column, taking
            them out across the diagonals, and running through multiple iterations
            to spread the data out over time. 4 bits are transmitted by each tone
            (or 5 bits for MFSK8, which uses 32 tones).

            Since MFSK16 uses a small symbol size (4 bits for 16 tones), it uses
            varicode encoding to convert incoming data into a simple stream of
            bits -- and a character keeps going until two "0's" are seen back-to-
            back. MT63 and Olivia used a fixed character size of 7 bits, which
            requires other schemes (e.g. "escape-127") to transmit 8-bit data.

            > When I did try it, it did not seem as robust as I expected it to when
            > we lowered the number of tones. This was across the state in the
            > evening on 80 meters. Seems like 8 tones gave us problems.

            The relationship between tones and bandwidth has several consequences.
            If you're running 1000 Hz bandwidth, and reduce to 8 tones, the spacing
            between tones is 125 Hz. This also means you're running 125 baud -
            which is getting a bit fast for reliable communications in a noisy
            environment. On the other hand, if you go to 256 tones at 1000 Hz,
            you're going to be running about 4 Hz between tones, and only going
            to have about 4 baud symbol rate. While that is going to give you
            excellent weak-signal characteristics, it's going to be S...L...O...W.

            Work done 50 years or so ago in the field of Information Theory came
            up with some interesting stuff about gettin information thorough
            noise -- in a given bandwidth, the amount of information you can get
            through is dependent on the signal-to-noise ratio. You can get by
            with a lot of noise if you are willing to reduce your throughput,
            but if you want more throughput, you've got to keep your noise down.
            That's why our landline modems have so much more throughput than we
            see on the radio, and why someone with a quiet phone line that is
            close to the phone company's central office gets a faster modem
            connection than someone who's 10 miles away on a noisy line.

            The FEC in Olivia is "stronger" than that in MFSK16, at the expense
            of throughput. Try running Olivia at 250 Hz - 16 tones -- the same
            bandwidth and baud rate as MFSK16. Olivia will be slower, but
            I would expect it to dig farther down into the noise for a signal.

            You'll also find that the tuning for Olivia is much less critcial,
            as it has the ability to search (within limits) for a valid signal.

            >>From my perspective, the throughput is too slow in the default speed.
            >
            > Has anyone been able to compare it to MFSK16 which thus far seems to
            > be the best mode for very weak signals and moderate keyboard speed?
            >
            > I am hoping to hear some return comments from the FCC to those who
            > received the notices so that I feel comfortable operating the mode again.

            I received a response from the FCC asking for more details about what
            exactly Ron received -- one sentence in the response says "Given
            the latitude in this rule, I can't imagine why an FCC field office would
            send someone an order." The reply continues on that additional info
            is needed (exactly what was sent, from what office, and the particulars
            cited about the operation). Until they get all the questions answered,
            they aren't going to go on record saying all is well -- but unless
            something unexpected pops up, I expect "all is well" to be the end result.

            By the way, I didn't get a notice from the FCC -- I inquired because I'm
            playing around with adding an ARQ wrapper around MT63, and wanted to stay
            out of hot water with the Feds when doing on-the-air testing. From the
            reply I received on that issue, I'm in the clear.

            > SCAMP has been taking up some time too. No one can say that mode is
            > weak signal though:)

            Again, we get to pick weak-signal or high-throughput. "Both" isn't
            an answer. The best answer for amateur use would be "adaptive" -
            smart enough to increase throughput if there's enough signal, and
            smart enough to drop it down if there isn't.

            Next time someone asks "Why are there so many digital modes?", we can
            answer "because the radio environment isn't always the same".

            > 73,
            >
            > Rick, KV9U
            >
            > 73,
            >
            > Rick, KV9U
            >
            >
            >
            >
            >
            >
            > Yahoo! Groups Links
            >
            >
            >
            >
            >
            >
          • Chris Gerber
            Well done Paul. A very good comment. Regarding adaptieve. Thats is done with Pactor 2 and 3. They even go further and control the output power and reduce
            Message 5 of 6 , Feb 16, 2005
            • 0 Attachment
              Well  done Paul. A very good comment.
              Regarding adaptieve. Thats is done with Pactor 2 and 3. They even go further and control the output power
              and reduce them, when the troughput is going up.

              73 Chris HB9BDM

              Paul L Schmidt schrieb:
              Rick Williams wrote:

              > Could someone outline how much difference there is between Olivia and
              > MT63 except that Olivia allows for changing the parameters?

              Both MT63 and Olivia use a Walsh-Hadamard function for FEC (as opposed
              to MFSK16 which uses Viterbi FEC) - for MT63 and Olivia, a 7-bit input
              word produces an array of 64 bits which describe the 7-bit word; for
              MFSK16, the Viterbi encoding expands a 4-bit word to 8 bits.

              In all cases, the encoded stream of bits is interleaved -- MT63 uses a
              straightforward interleaver that spreads the 64 bits across both time
              and frequency so that it takes either 32 or 64 symbol intervals to get
              them all out.  Olivia takes blocks of encoded characters (5 for 32-tone
              MFSK, 6 for 64-tone MFSK, etc) and spreads them out over time to produce
              a symbol to transmit during each time interval.  MFSK16 uses "diagonal
              interleaving" which places the bit patterns in arrays by column, taking
              them out across the diagonals, and running through multiple iterations
              to spread the data out over time. 4 bits are transmitted by each tone
              (or 5 bits for MFSK8, which uses 32 tones).

              Since MFSK16 uses a small symbol size (4 bits for 16 tones), it uses
              varicode encoding to convert incoming data into a simple stream of
              bits -- and a character keeps going until two "0's" are seen back-to-
              back.  MT63 and Olivia used a fixed character size of 7 bits, which
              requires other schemes (e.g. "escape-127") to transmit 8-bit data.

              > When I did try it, it did not seem as robust as I expected it to when
              > we lowered the number of tones. This was across the state in the
              > evening on 80 meters. Seems like 8 tones gave us problems.

              The relationship between tones and bandwidth has several consequences.
              If you're running 1000 Hz bandwidth, and reduce to 8 tones, the spacing
              between tones is 125 Hz.  This also means you're running 125 baud -
              which is getting a bit fast for reliable communications in a noisy
              environment.  On the other hand, if you go to 256 tones at 1000 Hz,
              you're going to be running about 4 Hz between tones, and only going
              to have about 4 baud symbol rate.  While that is going to give you
              excellent weak-signal characteristics, it's going to be S...L...O...W.

              Work done 50 years or so ago in the field of Information Theory came
              up with some interesting stuff about gettin information thorough
              noise -- in a given bandwidth, the amount of information you can get
              through is dependent on the signal-to-noise ratio.  You can get by
              with a lot of noise if you are willing to reduce your throughput,
              but if you want more throughput, you've got to keep your noise down.
              That's why our landline modems have so much more throughput than we
              see on the radio, and why someone with a quiet phone line that is
              close to the phone company's central office gets a faster modem
              connection than someone who's 10 miles away on a noisy line.

              The FEC in Olivia is "stronger" than that in MFSK16, at the expense
              of throughput.  Try running Olivia at 250 Hz - 16 tones -- the same
              bandwidth and baud rate as MFSK16.  Olivia will be slower, but
              I would expect it to dig farther down into the noise for a signal.

              You'll also find that the tuning for Olivia is much less critcial,
              as it has the ability to search (within limits) for a valid signal.

              >>From my perspective, the throughput is too slow in the default speed.
              >
              > Has anyone been able to compare it to MFSK16 which thus far seems to
              > be the best mode for very weak signals and moderate keyboard speed?
              >
              > I am hoping to hear some return comments from the FCC to those who
              > received the notices so that I feel comfortable operating the mode again.

              I received a response from the FCC asking for more details about what
              exactly Ron received -- one sentence in the response says "Given
              the latitude in this rule, I can't imagine why an FCC field office would
              send someone an order."  The reply continues on that additional info
              is needed (exactly what was sent, from what office, and the particulars
              cited about the operation).  Until they get all the questions answered,
              they aren't going to go on record saying all is well -- but unless
              something unexpected pops up, I expect "all is well" to be the end result.

              By the way, I didn't get a notice from the FCC -- I inquired because I'm
              playing around with adding an ARQ wrapper around MT63, and wanted to stay
              out of hot water with the Feds when doing on-the-air testing.  From the
              reply I received on that issue, I'm in the clear.

              > SCAMP has been taking up some time too. No one can say that mode is
              > weak signal though:)

              Again, we get to pick weak-signal or high-throughput.  "Both" isn't
              an answer.  The best answer for amateur use would be "adaptive" -
              smart enough to increase throughput if there's enough signal, and
              smart enough to drop it down if there isn't.

              Next time someone asks "Why are there so many digital modes?", we can
              answer "because the radio environment isn't always the same".

              > 73,
              >
              > Rick, KV9U
              >
              > 73,
              >
              > Rick, KV9U
              >
              >
              >
              >
              >

              > Yahoo! Groups Links
              >
              >
              >

              >
              >


            Your message has been successfully submitted and would be delivered to recipients shortly.