Re: [gpsd-users] gpsrinex: collecting one 24h GPS-day?

[Greg Troxel]

"Gary E. Miller" <address@hidden> writes:

> Yo Greg!
>
> On Wed, 25 Sep 2019 12:58:26 -0400
> Greg Troxel <address@hidden> wrote:
>
>> "Gary E. Miller" <address@hidden> writes:
>> 
>> >>  I am unclear on if "-i 30" lines up to gps times N, such that N %
>> >> 30 is reliably 0.  
>> >
>> > Yes, many of the PPP services are old and crusty FORTRAN that fails
>> > if that is not true.  
>> 
>> It would be nice if the man page said this clearly.  Maybe I missed
>> it, but I read it to try to find the answer and didn't.
>
> The number of critical details in RINEX is huge, service dependent.  And
> undocumented anywhere.  When in doubt, read the code, then test.
>
> Doc patches welcome.

Sorry, I didn't mean the complexity of RINEX.

I meant that if "-i 30" means "sample every 30s, aligned to integral
multiples of 30s in GPS time", it would be nice to say that.  vs "sample
right away, and then every 30s" vs "integral mulitples of 30s in UTC
(assuming leap seconds don't exist, as is UNIX tradition).

I have so far not gone down the gpsd/RINEX path, so am reluctant to
change docs based on my fuzzy understanding.