> It's late. Details to follow when I'm bored tomorrow. :-)
That's a good attitude!
> I think that actually is a legal (albeit tacky) waypoint name for a
> D109, but I'll have to pore over the specs for it.
> > The RINO then stores that waypoint with no name, and then when you fetch
> > it back with gpsbabel you get an empty name:
> I'd want to stare a the dumps coming and going on that before issuing
> a final judgement.
Makes sense...I guess I'm only looking at the gpx file that gets
produced. I'm having a good time looking at these dumps and actually
seeing what is really happening.
> Let's focus on this code path: is a waypoint named %20 really legal?
> do we do the right thing on a read? do we do the right thing on a write?
> This is known silliness in the Garmin protocol. There are floating
> point conversions involved in the encoding of the lat/lon/alt to any of
> the various Garmin protocols and there is some rounding error involved
> in much the same way as rounding 2/3 to hundredths == .67. If you do
> that and back a few times, you can accumulate an error, even though it's
> way beyond the precision of your unit.
I understand all of this at the 'oh christ, _that_ problem again'
level. The lat/long is one issue, the elevation is dropping a meter per
conversion. It looks like the unit is displaying in feet, so maybe that
is where it goes.
> (Wasn't that the plot of a really bad superman movie in the late 70's?)
Wait...you aren't implying that there were also _good_ superman movies are
> Look in GPS_D109_Get(), GPS_D109_Send(), and the Garmin protocol spec
> and be prepared to learm more about IEEE math and GPS_Math_Deg_to_Semi()
> than you probably ever wanted to know.
> If you're an FP guy that actually does understand the conversions to
> degree->semicircle->IEEE and back, you have the opportunity to earn a
> gold star in all this.)
hmmmm....gold star vs early brain death. Let me sleep on that decision
I have a deadline for another project, and then I'm out for a few days, so
my attention may be split.
On the other hand, I found myself fascinated today by searching out why I
was getting segfaults, so who knows what kinky code I'll want to explore