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

GeniusBeaconing Forecast Error

Expand Messages
  • Bob Burns W9RXR
    Could someone take a crack at explaining the GeniusBeaconing Forecast Error feature to me? I have read the Wiki entries on this topic, but it s not sinking in
    Message 1 of 19 , Oct 29, 2012
    • 0 Attachment
      Could someone take a crack at explaining the GeniusBeaconing Forecast
      Error feature to me?

      I have read the Wiki entries on this topic, but it's not sinking in
      to my thick head. The other settings all make sense and I can predict
      in my mind when a beacon would be sent based on time, distance, or
      heading change. The impact of Forecast Error, though, eludes me.

      Thanks!

      Bob...
    • James Ewen
      ... Sit down on the couch and pull off your sick Bob... we re going to get into some tricky math... 8) Aw heck put those socks back on! Think of forecast error
      Message 2 of 19 , Oct 29, 2012
      • 0 Attachment
        On Mon, Oct 29, 2012 at 6:04 PM, Bob Burns W9RXR <w9rxr_@...> wrote:

        > Could someone take a crack at explaining the GeniusBeaconing Forecast
        > Error feature to me?
        >
        > I have read the Wiki entries on this topic, but it's not sinking in
        > to my thick head. The other settings all make sense and I can predict
        > in my mind when a beacon would be sent based on time, distance, or
        > heading change. The impact of Forecast Error, though, eludes me.

        Sit down on the couch and pull off your sick Bob... we're going to get
        into some tricky math... 8)

        Aw heck put those socks back on!

        Think of forecast error like this. You're driving at 60 mph due north.
        You're set up to send a position report every mile, so once a minute
        you'll be making noise. Now, jam on the brakes and come to a stop.
        When's your next position report going to happen? You're no longer
        zooming along at 60 mph, so it's not going to happen 1 mile down the
        road from your last report. However, the program is forecasting your
        location due to the last reported speed and direction. Once that
        prediction forecast exceeds the threshold, the program will fire off a
        position report to let people know that you have stopped. The
        prediction threshold is adjustable in the settings.

        Same thing goes for making corners, etc... once the predicted location
        based on the last reported course and speed gets too far from reality,
        a new position report gets fired off. Turn on the meatball and the
        circle on the map display, and watch what happens! Once the meatball
        rolls off your plate, a new position report is fired off.

        --
        James
        VE6SRV
      • Randy Love
        Once the meatball rolls off your plate... That s priceless, James, but pretty darned accurate! Randy WF5X
        Message 3 of 19 , Oct 30, 2012
        • 0 Attachment
          "Once the meatball rolls off your plate..."

          That's priceless, James, but pretty darned accurate!

          Randy
          WF5X

        • Lynn W Deffenbaugh (Mr)
          And it goes to show why the Red Dot exists. For those that have no clue what we re talking about, read http://aprsisce.wikidot.com/doc:red-dot and make sure
          Message 4 of 19 , Oct 30, 2012
          • 0 Attachment
            And it goes to show why the "Red Dot" exists. 

            For those that have no clue what we're talking about, read http://aprsisce.wikidot.com/doc:red-dot and make sure you have Configure / Screen / RedDot checked.  It's only interesting while mobile with a true GPS feed, though.

            Lynn (D) - KJ4ERJ - Author of APRSISCE for Windows Mobile and Win32

            On 10/30/2012 12:22 PM, Randy Love wrote:
            "Once the meatball rolls off your plate..."

            That's priceless, James, but pretty darned accurate!

            Randy
            WF5X

          • Bob Burns W9RXR
            ... Sick? Now, I really confused. ... This threshold you refer to is the Forecast Error setting in the GeniusBeaconing window? Here s what I don t understand
            Message 5 of 19 , Oct 30, 2012
            • 0 Attachment
              At 08:35 PM 10/29/2012, James Ewen wrote:

              >Sit down on the couch and pull off your sick Bob.

              Sick? Now, I really confused.

              >However, the program is forecasting your location due to the last
              >reported speed and direction. Once that prediction forecast exceeds
              >the threshold, the program will fire off a position report to let
              >people know that you have stopped.

              This threshold you refer to is the Forecast Error setting in the
              GeniusBeaconing window?

              Here's what I don't understand about what you just said. If I'm
              driving down the road at 60 mph and sending a beacon every mile, I
              understand that I'm beaconing every 60 seconds. If I stop, then I no
              longer accumulate distance. Sixty seconds after my last beacon, the
              forecast error part of GeniusBeaconing thinks I should beacon, but
              the distance calculator says "no, you haven't gone far enough yet".
              That leads me to believe that the forecast error calculates the next
              beacon as a single fixed location based on my last speed and
              direction. How, can that prediction keep moving to exceed this
              threshold? This is my mental hang-up with this concept. I think the
              prediction is a fixed point, but you're telling me that it's moving.

              >Turn on the meatball and the circle on the map display, and watch
              >what happens!

              I know how to turn on the red dot, but what circle are you referring to?


              At 12:24 PM 10/30/2012, Lynn W Deffenbaugh (Mr) wrote:

              >It's only interesting while mobile with a true GPS feed, though.

              Agreed. The downside is that it's hard to play with the settings
              while mobile. And, it's even harder to look at the Wiki while mobile
              without an Internet connection.

              Bob...
            • g6enu
              The red dot is continuously updating. Its instantaneous calculated position is given by + *
              Message 6 of 19 , Oct 30, 2012
              • 0 Attachment
                The red dot is continuously updating. Its instantaneous calculated position is given by

                <Position of last beacon> + <time since last beacon> * <velocity at last beacon>

                Your locally-held GPS position is also updating semi-continuously, and almost certainly much more often than you are beaconing.

                If the calculated instantaneous red dot position differs from the locally-held GPS position by more than a threshold amount (the exact details of which someone more knowledgeable than I can explain) a beacon is forced even if it is not due on a pure time interval basis.

                (Note that this explanation is somewhat simplified, to match the brain and knowledge of the person posting it.)


                --- In aprsisce@yahoogroups.com, Bob Burns W9RXR <w9rxr_@...> wrote:
                >
                > At 08:35 PM 10/29/2012, James Ewen wrote:
                >
                > >Sit down on the couch and pull off your sick Bob.
                >
                > Sick? Now, I really confused.
                >
                > >However, the program is forecasting your location due to the last
                > >reported speed and direction. Once that prediction forecast exceeds
                > >the threshold, the program will fire off a position report to let
                > >people know that you have stopped.
                >
                > This threshold you refer to is the Forecast Error setting in the
                > GeniusBeaconing window?
                >
                > Here's what I don't understand about what you just said. If I'm
                > driving down the road at 60 mph and sending a beacon every mile, I
                > understand that I'm beaconing every 60 seconds. If I stop, then I no
                > longer accumulate distance. Sixty seconds after my last beacon, the
                > forecast error part of GeniusBeaconing thinks I should beacon, but
                > the distance calculator says "no, you haven't gone far enough yet".
                > That leads me to believe that the forecast error calculates the next
                > beacon as a single fixed location based on my last speed and
                > direction. How, can that prediction keep moving to exceed this
                > threshold? This is my mental hang-up with this concept. I think the
                > prediction is a fixed point, but you're telling me that it's moving.
                >
                > >Turn on the meatball and the circle on the map display, and watch
                > >what happens!
                >
                > I know how to turn on the red dot, but what circle are you referring to?
                >
                >
                > At 12:24 PM 10/30/2012, Lynn W Deffenbaugh (Mr) wrote:
                >
                > >It's only interesting while mobile with a true GPS feed, though.
                >
                > Agreed. The downside is that it's hard to play with the settings
                > while mobile. And, it's even harder to look at the Wiki while mobile
                > without an Internet connection.
                >
                > Bob...
                >
              • Lynn W Deffenbaugh (Mr)
                Simple answer: When the red dot hits the circle, a beacon is sent. This is an indicator of the Forecast Error feature of GeniusBeaconing. ... Typo: he meant
                Message 7 of 19 , Oct 30, 2012
                • 0 Attachment
                  Simple answer: When the red dot hits the circle, a beacon is sent.
                  This is an indicator of the Forecast Error feature of GeniusBeaconing.

                  On 10/30/2012 8:23 PM, Bob Burns W9RXR wrote:
                  > At 08:35 PM 10/29/2012, James Ewen wrote:
                  >
                  >> Sit down on the couch and pull off your sick Bob.
                  > Sick? Now, I really confused.

                  Typo: he meant to say pull off your SOCKS Bob...

                  >> However, the program is forecasting your location due to the last
                  >> reported speed and direction. Once that prediction forecast exceeds
                  >> the threshold, the program will fire off a position report to let
                  >> people know that you have stopped.
                  > This threshold you refer to is the Forecast Error setting in the
                  > GeniusBeaconing window?

                  Yes.

                  > Here's what I don't understand about what you just said. If I'm
                  > driving down the road at 60 mph and sending a beacon every mile, I
                  > understand that I'm beaconing every 60 seconds. If I stop, then I no
                  > longer accumulate distance. Sixty seconds after my last beacon, the
                  > forecast error part of GeniusBeaconing thinks I should beacon, but
                  > the distance calculator says "no, you haven't gone far enough yet".
                  > That leads me to believe that the forecast error calculates the next
                  > beacon as a single fixed location based on my last speed and
                  > direction. How, can that prediction keep moving to exceed this
                  > threshold? This is my mental hang-up with this concept. I think the
                  > prediction is a fixed point, but you're telling me that it's moving.

                  The prediction is moving, yes. Every time APRSISCE/32 beacons, it makes
                  a note of the location, time, speed, and heading. Every second after
                  that, it predicts where an outside observer would think you are based on
                  that information. It compares that predicted point (which moves every
                  second) with your actual position from the GPS. The difference between
                  those two is a distance and a bearing. If the distance is greater than
                  the Forecast Error, then a new beacon is sent because the outside world
                  thinks you're not close enough to where you really are.

                  The distance is also scaled such that the Configure / Screen / Circle
                  (the screen scale circle) is equal to the Forecast Error setting. The
                  red dot is drawn at this scaled distance between the center and the
                  circle in the direction of the error bearing.

                  So, if you beaconed 60mph due North and subsequently slowed down, you
                  can expect to see the red dot move towards the top until it hits the
                  circle at which point a new beacon triggered by the Forecast Error will
                  tell the world that you've slowed down.

                  If you beaconed 30mph due North and subsequently speed up, you can
                  expect the red dot to move towards the bottom of the circle until it
                  hits which triggers a beacon to tell the world that you're going faster.

                  Turns are interesting because if you turn right, the world thinks you're
                  still going straight ahead so the red dot will move LEFT until it hits
                  the circle. It goes this direction because the world thinks you're left
                  of where you really are until your next beacon tells them your new heading.

                  >> Turn on the meatball and the circle on the map display, and watch
                  >> what happens!
                  > I know how to turn on the red dot, but what circle are you referring to?

                  Configure / Screen / Circle which puts the range circle on the screen.
                  This same circle is scaled to the Forecast Error for the purposes of the
                  red dot indicating the error between a forecast (prediction) based on
                  your last transmitted information and the reality as known from your
                  local GPS.

                  > At 12:24 PM 10/30/2012, Lynn W Deffenbaugh (Mr) wrote:
                  >
                  >> It's only interesting while mobile with a true GPS feed, though.
                  > Agreed. The downside is that it's hard to play with the settings
                  > while mobile. And, it's even harder to look at the Wiki while mobile
                  > without an Internet connection.

                  You can even see the effect walking down the road. Walk at a steady
                  slow pace and hit Transmit. Then pick up speed and you'll see the red
                  dot begin to go towards the bottom of the screen. My neighbors must
                  have thought I was going crazy the few nights that I was out there
                  jogging and crawling alternately down the street...

                  Or just get someone else to drive!

                  Lynn (D) - KJ4ERJ - Author of APRSISCE for Windows Mobile and Win32


                  >
                  > Bob...
                  >
                  >
                  >
                  > ------------------------------------
                  >
                  > Yahoo! Groups Links
                  >
                  >
                  >
                  >
                • Lynn W Deffenbaugh (Mr)
                  No, the red dot does NOT indicate the calculated, forecasted, projected position! If I did that, the scaling gets all funky. Been there, did that, and it
                  Message 8 of 19 , Oct 30, 2012
                  • 0 Attachment
                    No, the red dot does NOT indicate the calculated, forecasted, projected
                    position! If I did that, the scaling gets all funky. Been there, did
                    that, and it really didn't help except at really close in zooms.
                    Although I guess I could put a pink dot there and add even more to the
                    confusion?

                    The red dot, relative to the scale circle on the screen, represents the
                    DIFFERENCE (or Error) between the calculated, forecasted, projected
                    position and your actual current position as a vector (magnitude and
                    direction).

                    If the red dot is moving out in front of you, then the outside world
                    thinks you're further along than you actually are. You must have slowed
                    down.

                    If the red dot is lagging behind you, then the outside world thinks
                    you're not as far along as you actually are. You must have sped up.

                    If the red dot is to your left, then you must have turned right but the
                    world thinks you're left of where you actually are.

                    And vice versa.

                    Lynn (D) - KJ4ERJ - Author of APRSISCE for Windows Mobile and Win32

                    PS. Can you tell that Florida roads go at 90 degree angles, N, S, E, W?

                    On 10/30/2012 8:49 PM, g6enu wrote:
                    >
                    >
                    > The red dot is continuously updating. Its instantaneous calculated position is given by
                    >
                    > <Position of last beacon> + <time since last beacon> * <velocity at last beacon>
                    >
                    > Your locally-held GPS position is also updating semi-continuously, and almost certainly much more often than you are beaconing.
                    >
                    > If the calculated instantaneous red dot position differs from the locally-held GPS position by more than a threshold amount (the exact details of which someone more knowledgeable than I can explain) a beacon is forced even if it is not due on a pure time interval basis.
                    >
                    > (Note that this explanation is somewhat simplified, to match the brain and knowledge of the person posting it.)
                    >
                    >
                    > --- In aprsisce@yahoogroups.com, Bob Burns W9RXR <w9rxr_@...> wrote:
                    >> At 08:35 PM 10/29/2012, James Ewen wrote:
                    >>
                    >>> Sit down on the couch and pull off your sick Bob.
                    >> Sick? Now, I really confused.
                    >>
                    >>> However, the program is forecasting your location due to the last
                    >>> reported speed and direction. Once that prediction forecast exceeds
                    >>> the threshold, the program will fire off a position report to let
                    >>> people know that you have stopped.
                    >> This threshold you refer to is the Forecast Error setting in the
                    >> GeniusBeaconing window?
                    >>
                    >> Here's what I don't understand about what you just said. If I'm
                    >> driving down the road at 60 mph and sending a beacon every mile, I
                    >> understand that I'm beaconing every 60 seconds. If I stop, then I no
                    >> longer accumulate distance. Sixty seconds after my last beacon, the
                    >> forecast error part of GeniusBeaconing thinks I should beacon, but
                    >> the distance calculator says "no, you haven't gone far enough yet".
                    >> That leads me to believe that the forecast error calculates the next
                    >> beacon as a single fixed location based on my last speed and
                    >> direction. How, can that prediction keep moving to exceed this
                    >> threshold? This is my mental hang-up with this concept. I think the
                    >> prediction is a fixed point, but you're telling me that it's moving.
                    >>
                    >>> Turn on the meatball and the circle on the map display, and watch
                    >>> what happens!
                    >> I know how to turn on the red dot, but what circle are you referring to?
                    >>
                    >>
                    >> At 12:24 PM 10/30/2012, Lynn W Deffenbaugh (Mr) wrote:
                    >>
                    >>> It's only interesting while mobile with a true GPS feed, though.
                    >> Agreed. The downside is that it's hard to play with the settings
                    >> while mobile. And, it's even harder to look at the Wiki while mobile
                    >> without an Internet connection.
                    >>
                    >> Bob...
                    >>
                    >
                    >
                    >
                    > ------------------------------------
                    >
                    > Yahoo! Groups Links
                    >
                    >
                    >
                    >
                  • Bob Burns W9RXR
                    At 08:59 PM 10/30/2012, Lynn W Deffenbaugh (Mr) wrote: I will digest the rest of your reply later today, but one comment you made really has me confused. ...
                    Message 9 of 19 , Oct 31, 2012
                    • 0 Attachment
                      At 08:59 PM 10/30/2012, Lynn W Deffenbaugh (Mr) wrote:

                      I will digest the rest of your reply later today, but one comment you
                      made really has me confused.

                      >Configure / Screen / Circle which puts the range circle on the screen.
                      >This same circle is scaled to the Forecast Error for the purposes of
                      >the red dot indicating the error between a forecast (prediction)
                      >based on your last transmitted information and the reality as known
                      >from your local GPS.

                      Which implies that as you change the Forecast Error setting, this
                      circle should change. But it doesn't. The only circle I see is the
                      one which extends from top to bottom of the screen which I thought
                      indicated the scale of the map. How can this circle also indicate the
                      Forecast Error?

                      Bob...
                    • g6enu
                      Question. Is the centre of the circle the last beaconed position or the current GPS position? Because if it s the latter, then the only difference between what
                      Message 10 of 19 , Oct 31, 2012
                      • 0 Attachment
                        Question. Is the centre of the circle the last beaconed position or the current GPS position? Because if it's the latter, then the only difference between what I said and what you said (assuming the scales of the map and the vector are the same) is how you are describing things, not what things actually are. (And I should know, because I spent quite a bit of time studying vector arithmetic during my math degree.)



                        --- In aprsisce@yahoogroups.com, "Lynn W Deffenbaugh (Mr)" <kj4erj@...> wrote:
                        >
                        > No, the red dot does NOT indicate the calculated, forecasted, projected
                        > position! If I did that, the scaling gets all funky. Been there, did
                        > that, and it really didn't help except at really close in zooms.
                        > Although I guess I could put a pink dot there and add even more to the
                        > confusion?
                        >
                        > The red dot, relative to the scale circle on the screen, represents the
                        > DIFFERENCE (or Error) between the calculated, forecasted, projected
                        > position and your actual current position as a vector (magnitude and
                        > direction).
                        >
                        > If the red dot is moving out in front of you, then the outside world
                        > thinks you're further along than you actually are. You must have slowed
                        > down.
                        >
                        > If the red dot is lagging behind you, then the outside world thinks
                        > you're not as far along as you actually are. You must have sped up.
                        >
                        > If the red dot is to your left, then you must have turned right but the
                        > world thinks you're left of where you actually are.
                        >
                        > And vice versa.
                        >
                        > Lynn (D) - KJ4ERJ - Author of APRSISCE for Windows Mobile and Win32
                        >
                        > PS. Can you tell that Florida roads go at 90 degree angles, N, S, E, W?
                        >
                        > On 10/30/2012 8:49 PM, g6enu wrote:
                        > >
                        > >
                        > > The red dot is continuously updating. Its instantaneous calculated position is given by
                        > >
                        > > <Position of last beacon> + <time since last beacon> * <velocity at last beacon>
                        > >
                        > > Your locally-held GPS position is also updating semi-continuously, and almost certainly much more often than you are beaconing.
                        > >
                        > > If the calculated instantaneous red dot position differs from the locally-held GPS position by more than a threshold amount (the exact details of which someone more knowledgeable than I can explain) a beacon is forced even if it is not due on a pure time interval basis.
                        > >
                        > > (Note that this explanation is somewhat simplified, to match the brain and knowledge of the person posting it.)
                        > >
                        > >
                        > > --- In aprsisce@yahoogroups.com, Bob Burns W9RXR <w9rxr_@> wrote:
                        > >> At 08:35 PM 10/29/2012, James Ewen wrote:
                        > >>
                        > >>> Sit down on the couch and pull off your sick Bob.
                        > >> Sick? Now, I really confused.
                        > >>
                        > >>> However, the program is forecasting your location due to the last
                        > >>> reported speed and direction. Once that prediction forecast exceeds
                        > >>> the threshold, the program will fire off a position report to let
                        > >>> people know that you have stopped.
                        > >> This threshold you refer to is the Forecast Error setting in the
                        > >> GeniusBeaconing window?
                        > >>
                        > >> Here's what I don't understand about what you just said. If I'm
                        > >> driving down the road at 60 mph and sending a beacon every mile, I
                        > >> understand that I'm beaconing every 60 seconds. If I stop, then I no
                        > >> longer accumulate distance. Sixty seconds after my last beacon, the
                        > >> forecast error part of GeniusBeaconing thinks I should beacon, but
                        > >> the distance calculator says "no, you haven't gone far enough yet".
                        > >> That leads me to believe that the forecast error calculates the next
                        > >> beacon as a single fixed location based on my last speed and
                        > >> direction. How, can that prediction keep moving to exceed this
                        > >> threshold? This is my mental hang-up with this concept. I think the
                        > >> prediction is a fixed point, but you're telling me that it's moving.
                        > >>
                        > >>> Turn on the meatball and the circle on the map display, and watch
                        > >>> what happens!
                        > >> I know how to turn on the red dot, but what circle are you referring to?
                        > >>
                        > >>
                        > >> At 12:24 PM 10/30/2012, Lynn W Deffenbaugh (Mr) wrote:
                        > >>
                        > >>> It's only interesting while mobile with a true GPS feed, though.
                        > >> Agreed. The downside is that it's hard to play with the settings
                        > >> while mobile. And, it's even harder to look at the Wiki while mobile
                        > >> without an Internet connection.
                        > >>
                        > >> Bob...
                        > >>
                        > >
                        > >
                        > >
                        > > ------------------------------------
                        > >
                        > > Yahoo! Groups Links
                        > >
                        > >
                        > >
                        > >
                        >
                      • Lynn W Deffenbaugh (Mr)
                        The center of the circle depends on what you have selected. In the normal mode, it is Centered on ME and therefore is your current position, but only the
                        Message 11 of 19 , Oct 31, 2012
                        • 0 Attachment
                          The center of the circle depends on what you have selected. In the
                          "normal" mode, it is "Centered on ME" and therefore is your current
                          position, but only the GPS position if you have the GPS enabled.
                          Otherwise, you may have panned around and ME is somewhere else, but the
                          center of the circle has nothing to do with the behavior of the red dot
                          because the red dot isn't showing a location on the map, but an
                          error/difference vector.

                          Think of the "ball" in an aircraft carrier landing. It doesn't where
                          the glide path is nor where the airplane is, but where the airplane is
                          relative to the flight path. High and right, low and left, dead center
                          on. (I may have this wrong because I'm not a navy pilot, but I think
                          that's how it works). The red dot is the same, it doesn't show where an
                          observer would think you are, but how far off that observer would be
                          from where you really are.

                          But the difference in what you said and what I said is that you said
                          that the red dot shows where an outside observer would think you are.
                          If it did that, and you're zoomed out to say about a 5 mile circle
                          radius, the red dot won't even come out from underneath ME's station symbol.

                          I said that the red dot shows the magnitude and direction of the
                          difference between where you are and where an observer would think you
                          are scaled such that the Forecast Error value is the radius (in pixels,
                          not miles) of the circle. So that even with a 5 mile zoom radius and a
                          1/10 mile Forecast Error, the red dot will move out across the map
                          showing the difference on a completely different (Forecast Error vs zoom
                          scale) scale.

                          The difference is between the red dot showing a position vs the red dot
                          showing an difference/error vector.

                          Lynn (D) - KJ4ERJ - Author of APRSISCE for Windows Mobile and Win32

                          PS. Put a different way, your original equation for the red dot's
                          "instantaneous calculated position" is correct as the position that an
                          outside observer would think you are at, but the red dot itself is the
                          difference between that and reality scaled such that the Forecast Error
                          value represents the full circle radius.

                          On 10/31/2012 6:10 AM, g6enu wrote:
                          > Question. Is the centre of the circle the last beaconed position or the current GPS position? Because if it's the latter, then the only difference between what I said and what you said (assuming the scales of the map and the vector are the same) is how you are describing things, not what things actually are. (And I should know, because I spent quite a bit of time studying vector arithmetic during my math degree.)
                          >
                          >
                          >
                          > --- In aprsisce@yahoogroups.com, "Lynn W Deffenbaugh (Mr)" <kj4erj@...> wrote:
                          >> No, the red dot does NOT indicate the calculated, forecasted, projected
                          >> position! If I did that, the scaling gets all funky. Been there, did
                          >> that, and it really didn't help except at really close in zooms.
                          >> Although I guess I could put a pink dot there and add even more to the
                          >> confusion?
                          >>
                          >> The red dot, relative to the scale circle on the screen, represents the
                          >> DIFFERENCE (or Error) between the calculated, forecasted, projected
                          >> position and your actual current position as a vector (magnitude and
                          >> direction).
                          >>
                          >> If the red dot is moving out in front of you, then the outside world
                          >> thinks you're further along than you actually are. You must have slowed
                          >> down.
                          >>
                          >> If the red dot is lagging behind you, then the outside world thinks
                          >> you're not as far along as you actually are. You must have sped up.
                          >>
                          >> If the red dot is to your left, then you must have turned right but the
                          >> world thinks you're left of where you actually are.
                          >>
                          >> And vice versa.
                          >>
                          >> Lynn (D) - KJ4ERJ - Author of APRSISCE for Windows Mobile and Win32
                          >>
                          >> PS. Can you tell that Florida roads go at 90 degree angles, N, S, E, W?
                          >>
                          >> On 10/30/2012 8:49 PM, g6enu wrote:
                          >>>
                          >>> The red dot is continuously updating. Its instantaneous calculated position is given by
                          >>>
                          >>> <Position of last beacon> + <time since last beacon> * <velocity at last beacon>
                          >>>
                          >>> Your locally-held GPS position is also updating semi-continuously, and almost certainly much more often than you are beaconing.
                          >>>
                          >>> If the calculated instantaneous red dot position differs from the locally-held GPS position by more than a threshold amount (the exact details of which someone more knowledgeable than I can explain) a beacon is forced even if it is not due on a pure time interval basis.
                          >>>
                          >>> (Note that this explanation is somewhat simplified, to match the brain and knowledge of the person posting it.)
                          >>>
                          >>>
                          >>> --- In aprsisce@yahoogroups.com, Bob Burns W9RXR <w9rxr_@> wrote:
                          >>>> At 08:35 PM 10/29/2012, James Ewen wrote:
                          >>>>
                          >>>>> Sit down on the couch and pull off your sick Bob.
                          >>>> Sick? Now, I really confused.
                          >>>>
                          >>>>> However, the program is forecasting your location due to the last
                          >>>>> reported speed and direction. Once that prediction forecast exceeds
                          >>>>> the threshold, the program will fire off a position report to let
                          >>>>> people know that you have stopped.
                          >>>> This threshold you refer to is the Forecast Error setting in the
                          >>>> GeniusBeaconing window?
                          >>>>
                          >>>> Here's what I don't understand about what you just said. If I'm
                          >>>> driving down the road at 60 mph and sending a beacon every mile, I
                          >>>> understand that I'm beaconing every 60 seconds. If I stop, then I no
                          >>>> longer accumulate distance. Sixty seconds after my last beacon, the
                          >>>> forecast error part of GeniusBeaconing thinks I should beacon, but
                          >>>> the distance calculator says "no, you haven't gone far enough yet".
                          >>>> That leads me to believe that the forecast error calculates the next
                          >>>> beacon as a single fixed location based on my last speed and
                          >>>> direction. How, can that prediction keep moving to exceed this
                          >>>> threshold? This is my mental hang-up with this concept. I think the
                          >>>> prediction is a fixed point, but you're telling me that it's moving.
                          >>>>
                          >>>>> Turn on the meatball and the circle on the map display, and watch
                          >>>>> what happens!
                          >>>> I know how to turn on the red dot, but what circle are you referring to?
                          >>>>
                          >>>>
                          >>>> At 12:24 PM 10/30/2012, Lynn W Deffenbaugh (Mr) wrote:
                          >>>>
                          >>>>> It's only interesting while mobile with a true GPS feed, though.
                          >>>> Agreed. The downside is that it's hard to play with the settings
                          >>>> while mobile. And, it's even harder to look at the Wiki while mobile
                          >>>> without an Internet connection.
                          >>>>
                          >>>> Bob...
                          >>>>
                          >>>
                          >>>
                          >>> ------------------------------------
                          >>>
                          >>> Yahoo! Groups Links
                          >>>
                          >>>
                          >>>
                          >>>
                          >
                          >
                          >
                          > ------------------------------------
                          >
                          > Yahoo! Groups Links
                          >
                          >
                          >
                          >
                        • Lynn W Deffenbaugh (Mr)
                          ... No, the circle goes along with the scale of the map as you zoom in and out. The scale value just below the +/- zoom indicator is the scale of the radius
                          Message 12 of 19 , Oct 31, 2012
                          • 0 Attachment
                            On 10/31/2012 6:02 AM, Bob Burns W9RXR wrote:
                            > At 08:59 PM 10/30/2012, Lynn W Deffenbaugh (Mr) wrote:
                            >
                            > I will digest the rest of your reply later today, but one comment you
                            > made really has me confused.
                            >
                            >> Configure / Screen / Circle which puts the range circle on the screen.
                            >> This same circle is scaled to the Forecast Error for the purposes of
                            >> the red dot indicating the error between a forecast (prediction)
                            >> based on your last transmitted information and the reality as known
                            > >from your local GPS.
                            >
                            > Which implies that as you change the Forecast Error setting, this
                            > circle should change. But it doesn't. The only circle I see is the
                            > one which extends from top to bottom of the screen which I thought
                            > indicated the scale of the map. How can this circle also indicate the
                            > Forecast Error?

                            No, the circle goes along with the scale of the map as you zoom in and
                            out. The scale value just below the +/- zoom indicator is the scale of
                            the radius of the circle. The circle itself is always sized to the max
                            that will fit within the smallest of your window dimensions.

                            What changes as you change the Forecast Error is where the red dot will
                            appear within that circle. If the magnitude of the error between the
                            forecast position and your actual position is 1/10 miles and the
                            Forecast Error is 1/10 mile, the red dot will be at the circle. If you
                            change the Forecast Error to 2/10 mile, then the Red dot will be half
                            way between the center and the edge of the circle.

                            I "borrow" the circle as the visual indicator of the configured limit of
                            the Forecast Error and scale the location of the red dot against that
                            configured value such that when the red dot hits the circle the Forecast
                            Error has been reached.

                            The math behind it isn't really important, IMHO. The fact is that the
                            Red Dot gives you a visual representation, when compared to the circle,
                            of how far away from your current position an outside observer would be
                            estimating with respect to the maximum that you want them to be in error
                            (the configured Forecast Error value).

                            Again, the short answer, when the red dot hits the circle, the Forecast
                            Error has been reached.

                            Lynn (D) - KJ4ERJ - Author of APRSISCE for Windows Mobile and Win32
                          • Fred Hillhouse
                            The downside is that it s hard to play with the settings while mobile. And, it s even harder to look at the Wiki while mobile without an Internet connection.
                            Message 13 of 19 , Oct 31, 2012
                            • 0 Attachment
                              "The downside is that it's hard to play with the settings while mobile. And,
                              it's even harder to look at the Wiki while mobile without an Internet
                              connection."

                              Gaaack! You not supposed to use this mobile! Just what were you thinking?!
                              Just kidding! For some reason I was struck with a brief moment of sanity!
                              Er, um, insanity!

                              Watch out for gremlins, spooks and the like!

                              Best regards,
                              Fred, N7FMH
                            • Steve Daniels
                              A couple of people have queried the fact that they are still receiving emails after unsubscribing. This Is down to yahoo being very slow at processing
                              Message 14 of 19 , Oct 31, 2012
                              • 0 Attachment

                                A couple of people have queried the fact that they are still receiving emails after unsubscribing.

                                This Is down to yahoo being very slow at processing unsubscriptions, up to a couple of days, your account gets disabled from a moderation point of view straight away.

                                But unfortunately the emails will still arrive for a couple of days.

                                 

                                I suggest if you wish to unsubscribe, to go into your account  and select no emails for the group, which should happen instantly, barring perhaps whats in the queue. Then unsubscribe

                                It will still take a couple of days for your account to fully clear from the group, but hopefully you won’t get emails whilst it happens.

                                 

                                There really is nothing the moderation team can do about it, short of complaining to yahoo, and I know what the response to that will beJ

                                 

                                Steve Daniels

                                G6UIM

                                Torbay Freecycle Moderator http://http://uk.groups.yahoo.com/group/torbay_freecycle

                                APRSISCE/32 Beta tester and WIKI editor http://aprsisce.wikidot.com._,___

                              • g6enu
                                Aha! That s what I was missing - the circle is being borrowed on a different scale for the red dot processing. You have also explained to me something I (as
                                Message 15 of 19 , Oct 31, 2012
                                • 0 Attachment
                                  Aha! That's what I was missing - the circle is being "borrowed on a different scale" for the red dot processing.

                                  You have also explained to me something I (as a non-American, non-Navy non-Pilot) kept hearing and not understanding when watching episodes of JAG - the interchange "I have the ball", "Roger ball" between pilot and deck ops.


                                  --- In aprsisce@yahoogroups.com, "Lynn W Deffenbaugh (Mr)" <kj4erj@...> wrote:
                                  >
                                  > The center of the circle depends on what you have selected. In the
                                  > "normal" mode, it is "Centered on ME" and therefore is your current
                                  > position, but only the GPS position if you have the GPS enabled.
                                  > Otherwise, you may have panned around and ME is somewhere else, but the
                                  > center of the circle has nothing to do with the behavior of the red dot
                                  > because the red dot isn't showing a location on the map, but an
                                  > error/difference vector.
                                  >
                                  > Think of the "ball" in an aircraft carrier landing. It doesn't where
                                  > the glide path is nor where the airplane is, but where the airplane is
                                  > relative to the flight path. High and right, low and left, dead center
                                  > on. (I may have this wrong because I'm not a navy pilot, but I think
                                  > that's how it works). The red dot is the same, it doesn't show where an
                                  > observer would think you are, but how far off that observer would be
                                  > from where you really are.
                                  >
                                  > But the difference in what you said and what I said is that you said
                                  > that the red dot shows where an outside observer would think you are.
                                  > If it did that, and you're zoomed out to say about a 5 mile circle
                                  > radius, the red dot won't even come out from underneath ME's station symbol.
                                  >
                                  > I said that the red dot shows the magnitude and direction of the
                                  > difference between where you are and where an observer would think you
                                  > are scaled such that the Forecast Error value is the radius (in pixels,
                                  > not miles) of the circle. So that even with a 5 mile zoom radius and a
                                  > 1/10 mile Forecast Error, the red dot will move out across the map
                                  > showing the difference on a completely different (Forecast Error vs zoom
                                  > scale) scale.
                                  >
                                  > The difference is between the red dot showing a position vs the red dot
                                  > showing an difference/error vector.
                                  >
                                  > Lynn (D) - KJ4ERJ - Author of APRSISCE for Windows Mobile and Win32
                                  >
                                  > PS. Put a different way, your original equation for the red dot's
                                  > "instantaneous calculated position" is correct as the position that an
                                  > outside observer would think you are at, but the red dot itself is the
                                  > difference between that and reality scaled such that the Forecast Error
                                  > value represents the full circle radius.
                                  >
                                  > On 10/31/2012 6:10 AM, g6enu wrote:
                                  > > Question. Is the centre of the circle the last beaconed position or the current GPS position? Because if it's the latter, then the only difference between what I said and what you said (assuming the scales of the map and the vector are the same) is how you are describing things, not what things actually are. (And I should know, because I spent quite a bit of time studying vector arithmetic during my math degree.)
                                  > >
                                  > >
                                  > >
                                  > > --- In aprsisce@yahoogroups.com, "Lynn W Deffenbaugh (Mr)" <kj4erj@> wrote:
                                  > >> No, the red dot does NOT indicate the calculated, forecasted, projected
                                  > >> position! If I did that, the scaling gets all funky. Been there, did
                                  > >> that, and it really didn't help except at really close in zooms.
                                  > >> Although I guess I could put a pink dot there and add even more to the
                                  > >> confusion?
                                  > >>
                                  > >> The red dot, relative to the scale circle on the screen, represents the
                                  > >> DIFFERENCE (or Error) between the calculated, forecasted, projected
                                  > >> position and your actual current position as a vector (magnitude and
                                  > >> direction).
                                  > >>
                                  > >> If the red dot is moving out in front of you, then the outside world
                                  > >> thinks you're further along than you actually are. You must have slowed
                                  > >> down.
                                  > >>
                                  > >> If the red dot is lagging behind you, then the outside world thinks
                                  > >> you're not as far along as you actually are. You must have sped up.
                                  > >>
                                  > >> If the red dot is to your left, then you must have turned right but the
                                  > >> world thinks you're left of where you actually are.
                                  > >>
                                  > >> And vice versa.
                                  > >>
                                  > >> Lynn (D) - KJ4ERJ - Author of APRSISCE for Windows Mobile and Win32
                                  > >>
                                  > >> PS. Can you tell that Florida roads go at 90 degree angles, N, S, E, W?
                                  > >>
                                  > >> On 10/30/2012 8:49 PM, g6enu wrote:
                                  > >>>
                                  > >>> The red dot is continuously updating. Its instantaneous calculated position is given by
                                  > >>>
                                  > >>> <Position of last beacon> + <time since last beacon> * <velocity at last beacon>
                                  > >>>
                                  > >>> Your locally-held GPS position is also updating semi-continuously, and almost certainly much more often than you are beaconing.
                                  > >>>
                                  > >>> If the calculated instantaneous red dot position differs from the locally-held GPS position by more than a threshold amount (the exact details of which someone more knowledgeable than I can explain) a beacon is forced even if it is not due on a pure time interval basis.
                                  > >>>
                                  > >>> (Note that this explanation is somewhat simplified, to match the brain and knowledge of the person posting it.)
                                  > >>>
                                  > >>>
                                  > >>> --- In aprsisce@yahoogroups.com, Bob Burns W9RXR <w9rxr_@> wrote:
                                  > >>>> At 08:35 PM 10/29/2012, James Ewen wrote:
                                  > >>>>
                                  > >>>>> Sit down on the couch and pull off your sick Bob.
                                  > >>>> Sick? Now, I really confused.
                                  > >>>>
                                  > >>>>> However, the program is forecasting your location due to the last
                                  > >>>>> reported speed and direction. Once that prediction forecast exceeds
                                  > >>>>> the threshold, the program will fire off a position report to let
                                  > >>>>> people know that you have stopped.
                                  > >>>> This threshold you refer to is the Forecast Error setting in the
                                  > >>>> GeniusBeaconing window?
                                  > >>>>
                                  > >>>> Here's what I don't understand about what you just said. If I'm
                                  > >>>> driving down the road at 60 mph and sending a beacon every mile, I
                                  > >>>> understand that I'm beaconing every 60 seconds. If I stop, then I no
                                  > >>>> longer accumulate distance. Sixty seconds after my last beacon, the
                                  > >>>> forecast error part of GeniusBeaconing thinks I should beacon, but
                                  > >>>> the distance calculator says "no, you haven't gone far enough yet".
                                  > >>>> That leads me to believe that the forecast error calculates the next
                                  > >>>> beacon as a single fixed location based on my last speed and
                                  > >>>> direction. How, can that prediction keep moving to exceed this
                                  > >>>> threshold? This is my mental hang-up with this concept. I think the
                                  > >>>> prediction is a fixed point, but you're telling me that it's moving.
                                  > >>>>
                                  > >>>>> Turn on the meatball and the circle on the map display, and watch
                                  > >>>>> what happens!
                                  > >>>> I know how to turn on the red dot, but what circle are you referring to?
                                  > >>>>
                                  > >>>>
                                  > >>>> At 12:24 PM 10/30/2012, Lynn W Deffenbaugh (Mr) wrote:
                                  > >>>>
                                  > >>>>> It's only interesting while mobile with a true GPS feed, though.
                                  > >>>> Agreed. The downside is that it's hard to play with the settings
                                  > >>>> while mobile. And, it's even harder to look at the Wiki while mobile
                                  > >>>> without an Internet connection.
                                  > >>>>
                                  > >>>> Bob...
                                  > >>>>
                                  > >>>
                                  > >>>
                                  > >>> ------------------------------------
                                  > >>>
                                  > >>> Yahoo! Groups Links
                                  > >>>
                                  > >>>
                                  > >>>
                                  > >>>
                                  > >
                                  > >
                                  > >
                                  > > ------------------------------------
                                  > >
                                  > > Yahoo! Groups Links
                                  > >
                                  > >
                                  > >
                                  > >
                                  >
                                • Bob Burns W9RXR
                                  ... This is huge, Lynn. I ve seen explanations before that imply that the red dot is a projection of where the program thinks you should be based on last
                                  Message 16 of 19 , Oct 31, 2012
                                  • 0 Attachment
                                    On Wed, Oct 31, 2012 at 8:38 AM, Lynn W Deffenbaugh (Mr) <kj4erj@...> wrote:
                                    I said that the red dot shows the magnitude and direction of the
                                    difference between where you are and where an observer would think you
                                    are scaled such that the Forecast Error value is the radius (in pixels,
                                    not miles) of the circle.

                                    This is huge, Lynn. I've seen explanations before that imply that the red dot is a projection of where the program thinks you should be based on last reported position, speed, and direction. But, what you are saying here is the forehead-slapping moment. The red dot is the vector _difference_ between where you actually are (based on GPS input) and where the program thinks you should be.

                                    Ah-ha! <slap>

                                    Bob...

                                    P.S. Anybody remember the old Night Court sit-com? The Bull Shannon character did the best forehead slaps I've ever seen.
                                  • Bob Burns W9RXR
                                    ... More, ah-ha! It s starting to sink in. And, I didn t have to take my shoes off once. Bob...
                                    Message 17 of 19 , Oct 31, 2012
                                    • 0 Attachment
                                      On Wed, Oct 31, 2012 at 8:44 AM, Lynn W Deffenbaugh (Mr) <kj4erj@...> wrote:
                                      What changes as you change the Forecast Error is where the red dot will
                                      appear within that circle.  If the magnitude of the error between the
                                      forecast position and your actual position is 1/10 miles and the
                                      Forecast Error is 1/10 mile, the red dot will be at the circle.  If you
                                      change the Forecast Error to 2/10 mile, then the Red dot will be half
                                      way between the center and the edge of the circle.

                                      More, ah-ha!

                                      It's starting to sink in. And, I didn't have to take my shoes off once.

                                      Bob...
                                    • Bob Burns W9RXR
                                      ... OK, smarty-pants. Let s talk about proper settings of the parameters in the GeniusBeaconing window. The Min Time, Max Time, Time Only, and
                                      Message 18 of 19 , Oct 31, 2012
                                      • 0 Attachment
                                        On Wed, Oct 31, 2012 at 9:50 AM, Fred Hillhouse <fmhillhouse@...> wrote:
                                        Gaaack! You not supposed to use this mobile! Just what were you thinking?!

                                        OK, smarty-pants. <grin>  Let's talk about proper settings of the parameters in the GeniusBeaconing window. 

                                        The Min Time, Max Time, Time Only, and Start/Stop parameters all seem to be straightforward.

                                        Heading Change -- Lynn has mentioned recently that he thinks the Forecast Error works better than the Heading Change, so he's recommending setting this to 180 which essentially disables Heading Change as a factor. I'm good with that as long as Forecast Error works as well as he thinks it does.

                                        Max Distance -- My daily routine might have me on the freeway driving 60-70 MPH with few turns or on city thoroughfares driving 30-45 MPH with few turns or on city or residential streets driving 20-30 MPH with lots of turns. If I wanted to drop a beacon at least every 2 miles, then Max Distance would be set to 20, right? Alternatively, if I'm the tail-end vehicle for a marathon (which I will be this Saturday), I'll only be driving about 4-5 MPH while following the last runner (walker) on the marathon route. Net Control would like to see a beacon from me roughly every half mile. So, Max Distance should be set to 5, right?

                                        Forecast Error -- Here's where I'm clueless. How does one determine what a good number is for Forecast Error? What are other people using?

                                        Bob...
                                      • Lynn W Deffenbaugh (Mr)
                                        Just as a point of reference, here s the Configure / Genius dialog (Note: this is NOT my mobile): ... Yep. Made to be that way. Min Time has some interesting
                                        Message 19 of 19 , Oct 31, 2012
                                        • 0 Attachment
                                          Just as a point of reference, here's the Configure / Genius dialog (Note: this is NOT my mobile):



                                          On 10/31/2012 12:02 PM, Bob Burns W9RXR wrote:
                                          The Min Time, Max Time, Time Only, and Start/Stop parameters all seem to be straightforward.
                                          Yep.  Made to be that way.  Min Time has some interesting effects on the others though.  It trumps everything and even the manual Transmit menu option will warn you about doing it too soon if you are less than this time.


                                          Heading Change -- Lynn has mentioned recently that he thinks the Forecast Error works better than the Heading Change, so he's recommending setting this to 180 which essentially disables Heading Change as a factor. I'm good with that as long as Forecast Error works as well as he thinks it does.

                                          Easiest way to find out is to set it to 180, check Configure / Beacon / Why, and drive around for a while and check your track later.

                                          Remember also that the track points for ME are recorded when APRSISCE/32 transmits, so your local track view should match what an outside observer (say aprs.fi) would see if they received 100% of your transmissions.


                                          Max Distance -- My daily routine might have me on the freeway driving 60-70 MPH with few turns or on city thoroughfares driving 30-45 MPH with few turns or on city or residential streets driving 20-30 MPH with lots of turns.

                                          That's why I use Max Distance and not any sort of slow/fast sliding window.  APRSISCE/32 has the luxury of compute power so it's (relatively) cheap to calculate the actual distance since your last beacon.  Any slow/fast slider between high/low speeds is hard to predict and even worse to have work properly with various speeds.  You ask APRSISCE/32 for a beacon every 1 mile max, and you'll get a beacon every mile.  But remember, that's still not a mile traveled (like an odometer), but a mile away from your last beaconed position.

                                          If I wanted to drop a beacon at least every 2 miles, then Max Distance would be set to 20, right?

                                          Yes.

                                          Alternatively, if I'm the tail-end vehicle for a marathon (which I will be this Saturday), I'll only be driving about 4-5 MPH while following the last runner (walker) on the marathon route. Net Control would like to see a beacon from me roughly every half mile. So, Max Distance should be set to 5, right?

                                          Yes again.


                                          Forecast Error -- Here's where I'm clueless. How does one determine what a good number is for Forecast Error? What are other people using?

                                          Personally, I like 1/10 mile.  In conjunction with a 15 or 30 second Min Time, it seems to perform quite to my liking.  But as always, YMMV (pun intended).

                                          Lynn (D) - KJ4ERJ - Author of APRSISCE for Windows Mobile and Win32

                                          PS.  I'm also thinking of adding an accumulator that counts how many beacons were triggered for what reasons just to make this a bit easier to see after a trip.  Of course, I'm also trying to figure out an easy way to clear these accumulators so that you can start a given trip "clean".



                                          Bob...

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