GPS field trips
werner at openmoko.org
Wed Apr 16 08:45:04 CEST 2008
FYI, reposting from an internal discussion:
In response to some truly devastating GPS results Will reported from
a field trip he did recently (basically: not a single fix in two
days, under sometimes ideal conditions), I gave my GTA02 a try as
The main question was how long it would take for GPS to find the
position after a cold-start and how this compared to "typical" other
devices. Here's what I found:
>From my office, downtown Buenos Aires, indoors, 21th floor, facing
north, clear field about 120 deg horizontally, 45 deg vertically,
horizon at +5 deg, foggy afternoon, 15C, 50% humidity.
All devices cold-started.
Garmin eTrex (2001):
~45s sees the first satellite
~75s gets a firm lock on it
5+ min still no position
It does get the correct time.
TomTom GO (2004-ish ?)
<20s gets time
<20s sees satellites
7-8min later, gets a fix, but comes and goes
~10s sees 1st sat (in GSV sentence *)
~20s we have the time
~180s gets position
(*) GPS sentences are explained here:
>From my rooftop terrace, outdoors, 22nd floor, horizon at -5...0 deg
for 240 deg hor, +10 deg very for 70 deg, +45 deg for 50 deg (yours
truly standing in the way), i.e., almost as good as standing in the
middle of the pampa in terms of clearance. Same weather conditions.
15s sees satellites
50s gets first firm satellite fix
110s gets position
50s gets position
45s gets first firm satellite fix (in GSA sentence)
60s gets position
Full disclosure: the GTA02 had a slight advantage because I left the
top off, so it could also see satellites that would normally be
covered by the earpiece.
Excerpts from a sample session:
# echo 1 >/sys/devices/platform/s3c2440-i2c/i2c-adapter/i2c-0/0-0073/neo1973-pm-gps.0/pwron
# tr -s '\012' </dev/ttySAC1 | grep GPGSA
^ no fix
^ 2D fix (the best I got indoors, outdoors got 3D right away)
I also have a few GTA01 and HXD8 laying around, but they're not in a
state fit for quick GPS testing.
More information about the openmoko-devel