[Xastir-Dev] question regarding my track

[Chris Bell]

> On Fri, 21 Feb 2003, Chris Bell wrote:
> 
> > The fundamental issue seems to be that track points generated by
> > moving my station (by whatever means, in the usual case GPS, either
> > local/tnc/gpsd) are treated differently than the a point received from
> > an echo (digipeat).  Should they?
> 
> I don't think they should be any different.  I believe what you are
> seeing is that until the packets go through the decode routines in
> db.c, they aren't seen by Xastir.
> 
> The code should send out packet _and_ fling a packet into the decode
> routines as well (kind of like a local loopback interface).  Perhaps
> these locally flung packets aren't getting flung?  Also perhaps
> they're not getting flung fast enough.
> 

When the gps updates our position, it does not generate a packet. 


> 
> > Operational example:   I have a GPS on the radio port of my kpc-3+.
> > Fire up xastir, drive down the road.  My icon (which track-me turned
> > on, but does not seem to make a difference)
> 
> It does make a difference as you get close to the map edge.  If it's
> not on, you won't flip to the next map view.
> 

That is only what you see, it (should) generate a track even if we are
not on the screen at the time.  That is why I say it makes no
difference.  (BTW, I **LOVE** the way it scrolls the moving map on a
tracked station!  It does it exactly how I wanted it to work!)

> 
> > hops along down the road
> > with each GPS poll.  I have no colored track though.  Once I finally
> > get an echo back from a digi, I suddenly start drawing a trail,
> > following the gps updates.
> 
> Seems to be that our local loopback is broken?
> 

more than that. I think it might be easy to fix though.

> 
> > Things get even stranger with DR turned on.  My icon gets
> > dead-reconned, but the DR ghost is drawn with the current speed and
> > heading, starting from the current icon position (which are all tracking
> > the GPS).  Only after a successful digi echo does the DR track get
> > reset to 0.  [so even though the ghost position has no meaning at all
> > in a DR context, it did show me how long "ago" I had a good echo.
> 
> We should probably disable DR for our own station.
> 

I tried updating the sec_heard and that makes DR to do the right
thing.  In testing it, I realized we could actually DR our own
position if we lost GPS for some reason... I am looking at how easy
that would be.  I will add a config toggle for it though.  

> 
> > I first noticed this testing some gpsd related issues, if I was not
> > connected to the TNC (or net) I never drew a track.  I could fix it by
> > connecting to something and getting a digipeat.  Another way to fix it
> > was to load a log, either from disk or findu that had my station
> > moving around.  Then the gps track would follow me around.
> 
> I might take a look at this sometime this weekend as well.  Putting
> a Garmin in simulator mode and driving myself around the map is a
> pretty good way to do it.  Let me know if you find something so that
> we don't duplicate efforts.  I'll do the same.
> 

I think I have it fixed.  Will check in after I complete some more
testing.  The DR part is done, I am looking at the track part.  [hard
to test comprehensively while driving, and telling my brother to poke
at things...  :) ]

Chris.