bienemelker
Geocacher
Hallo,
bei den Developers gb es folgenden Beitrag.
So i resume what i found around this problem. First my rom is genuine htc 1.96.710.3 wwe. my radio is 1.05.25.bs14.
through htc gps tool 1.1 i've looked a lot to nmea message, and it appears that static navigation is achieved throug the $gpvtg sentence only. $gpvtg tells to software what is current speed and direction. It seems that under 5km/h $gpvtg always report 0km/h and maybe 0°.
BUT: gps coordinates still get updated, and show the gps coordinates slowly moving beecause of position error OR real slow moving of the gps itself. So i guess we have two situations here: some software use $gpvtg to report movementand some don't. The later ones use gps coordinates only and calculate speed and direction themselves. Those software aren't affected by static navigation.
How should we disable static navigation now ?
Let's find the piece of software/driver which compute $gpvtg, because it's there that low speeds get filtered and create static navigation. Does the file gpsdriver.dll contain the string $gpvtg ? then what test on speed should we hack (through disassembling) to prevent driver from filtering low speeds.
Another way to fix static nav is to develop another layer of driver which recompute $gpvtg using gps coordinate. I think of a program listening to gps through com4, computing $gpvtg, then resending it through another com. Then as a user you'll have to choose com5 for pedestrian mode and com4 for car navigation.
Kann es nicht sein, wenn man in Cachewolf die GPS-Routine abändert, damit er nur die GPS-Koordinaten nimmt, dass man zumindest eine genaue Positionierung auch unter 3 km/h bekommt?
Gruß
bei den Developers gb es folgenden Beitrag.
So i resume what i found around this problem. First my rom is genuine htc 1.96.710.3 wwe. my radio is 1.05.25.bs14.
through htc gps tool 1.1 i've looked a lot to nmea message, and it appears that static navigation is achieved throug the $gpvtg sentence only. $gpvtg tells to software what is current speed and direction. It seems that under 5km/h $gpvtg always report 0km/h and maybe 0°.
BUT: gps coordinates still get updated, and show the gps coordinates slowly moving beecause of position error OR real slow moving of the gps itself. So i guess we have two situations here: some software use $gpvtg to report movementand some don't. The later ones use gps coordinates only and calculate speed and direction themselves. Those software aren't affected by static navigation.
How should we disable static navigation now ?
Let's find the piece of software/driver which compute $gpvtg, because it's there that low speeds get filtered and create static navigation. Does the file gpsdriver.dll contain the string $gpvtg ? then what test on speed should we hack (through disassembling) to prevent driver from filtering low speeds.
Another way to fix static nav is to develop another layer of driver which recompute $gpvtg using gps coordinate. I think of a program listening to gps through com4, computing $gpvtg, then resending it through another com. Then as a user you'll have to choose com5 for pedestrian mode and com4 for car navigation.
Kann es nicht sein, wenn man in Cachewolf die GPS-Routine abändert, damit er nur die GPS-Koordinaten nimmt, dass man zumindest eine genaue Positionierung auch unter 3 km/h bekommt?
Gruß
