Re: 2 separate positions pass through gpsd

[Gary E. Miller]

Yo Filip!

On Fri, 14 Oct 2022 08:29:10 +0200
Filip Kubicz <filip.kubicz@tier.app> wrote:

> > Every other DR receiver does that for you.  Quectel should not make
> > the  
> client do that.
> 
> I'm aware of it - their team is working on a fix for this.

I'll not hold my breath...

> > $PQTMINS, $PQTMIMU and $PQTGPS do not return time!  So no way to
> > include  
> thme in a TPV report.  There may be a way to showhorn this mess into
> an ATT report.
> 
> This hurts, but the time must come from $GNRMC or $GNGNS -
> unfortunately you cannot base only on the $PQTMINS.

I'll bet the $PQTMINS measurement instant is not the same as the $GNRMC
one.  u-blox does the same thing, the IMU has its own, unsynced, clock.

It would be nice if you could proved a 10 second raw capture:
        gpspipe -R -x 10 > raw.log

Then we can all see this mess up closer.

> Looks like gpsd is
> merging the time into TPV coming after each $PQTMINS pretty well.

I have no way to know wwithout a raw cap.

> > We can add new variables in gps_fix_t if needed.  But if you are
> > getting  
> rounding errors on doubles, you got big problems.
> 
> The rounding was coming from gpsd_json.c: it was putting these doubles
> into json with just 1 decimal place (for the wing angle variables).

"wing angle"??  You mean Wind Angle Magnetic (wanglem)?  You are measuring
wind to better than 0.1 degrees??

> For now, I worked it around by putting 9 decimal places, the same as
> lat/lon. I'm happy with this solution - until a FW fix arrives.

9 decimal places of fractional wind angle???
 
> > Go ahead an make Merge Request to gpsd with your code.  It will
> > need a  
> lot of work.  Easier for me to add and track comments in the MR
> process. If nothing else just getting the decodes to the log output
> would be helpful.
> 
> Sure - for the $PQTMINS. The $PQTMGPS is a temporary quick fix and I
> would not include this into gpsd. As soon as Quectel releases a new
> firmware, it will not be needed.

gpsd wants to decode evertyhing NMEA that it can, even if only to the
logs.  Users complain when they get "Unknown NMEA message" warnings.
Rather than worry about the deetails, get an MR up that has stubs, just
decodes to logs, for both messages.  After that is merged, and we see the
data, we can decide what to do with it.

RGDS
GARY
---------------------------------------------------------------------------
Gary E. Miller Rellim 109 NW Wilmington Ave., Suite E, Bend, OR 97703
        gem@rellim.com  Tel:+1 541 382 8588

            Veritas liberabit vos. -- Quid est veritas?
    "If you can't measure it, you can't improve it." - Lord Kelvin

pgpbxg0USSeJu.pgp
Description: OpenPGP digital signature