Re: Time precision in gps_fix_t

[Greg Troxel]

I think PPS is a different thing from what Marco might be asking about.

Many receivers now can provide fixes at faster than 1 Hz.  I have an
M8030 which is doing 2 Hz fixes (I have not tried to change this, but
it's configurable for a variety of 1-10 Hz rates depending on enabled
constellations).  Each fix of course has position and time, but after
being processed through gpsd the json output has lost the fractional
time (probably 0.500s), and every other fix has the same time.  I think
the question is about carrying a time-of-fix through the system without
losing precision.

There will be some latency from measurement at the fix time to
calculation in the chipset to reception  of UBX binary at the computer,
etc.  But I don't think that's a good reason to round.

I am guessing this is fairly easy to fix.