time-nuts@lists.febo.com

Discussion of precise time and frequency measurement

View all threads

Re: [time-nuts] Galileo service currently degraded

HM
Hal Murray
Sat, Jul 13, 2019 7:54 PM

The library I used for the U-blox UBX protocol frankly sucks, and was a major
source of issues.

What library are you using and/or did you look at gpsd?

--
These are my opinions.  I hate spam.

kevin@kevincroissant.com said: > The library I used for the U-blox UBX protocol frankly sucks, and was a major > source of issues. What library are you using and/or did you look at gpsd? -- These are my opinions. I hate spam.
KC
Kevin Croissant
Sat, Jul 13, 2019 9:16 PM

This is the library I am using:
https://github.com/jkua/ubx

Basically the only applicable files are ublox2.py and ubloxMessage.py.
There's some bug in how it handles its byte buffer and the buffer grows out
of control, which causes all sorts of issues. I tried to fix it but failed
due to time constraints. Highly recommend against touching that library...

I started off by using NMEA but found that there is was more resolution in
the UBX messages, so that's why I went that route. I thought there must be
a way to get the same resolution, but did not figure it out, and eventually
I had to just get it online so I could move on. Looking at gpsd, I'm still
not sure that it supports UBX, and I think that was why I disregarded it
before. Do you know if it does? The only UBX messages my project truly
needs are NAV-PVT and NAV-DOP, though I look at other ones for data
validation and debugging.

Kevin

On Sat, Jul 13, 2019 at 5:06 PM Hal Murray hmurray@megapathdsl.net wrote:

The library I used for the U-blox UBX protocol frankly sucks, and was a

major

source of issues.

What library are you using and/or did you look at gpsd?

--
These are my opinions.  I hate spam.


time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe, go to
http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
and follow the instructions there.

--
Kevin Croissant

This is the library I am using: https://github.com/jkua/ubx Basically the only applicable files are ublox2.py and ubloxMessage.py. There's some bug in how it handles its byte buffer and the buffer grows out of control, which causes all sorts of issues. I tried to fix it but failed due to time constraints. Highly recommend against touching that library... I started off by using NMEA but found that there is was more resolution in the UBX messages, so that's why I went that route. I thought there must be a way to get the same resolution, but did not figure it out, and eventually I had to just get it online so I could move on. Looking at gpsd, I'm still not sure that it supports UBX, and I think that was why I disregarded it before. Do you know if it does? The only UBX messages my project truly needs are NAV-PVT and NAV-DOP, though I look at other ones for data validation and debugging. Kevin On Sat, Jul 13, 2019 at 5:06 PM Hal Murray <hmurray@megapathdsl.net> wrote: > > kevin@kevincroissant.com said: > > The library I used for the U-blox UBX protocol frankly sucks, and was a > major > > source of issues. > > What library are you using and/or did you look at gpsd? > > > -- > These are my opinions. I hate spam. > > > > > _______________________________________________ > time-nuts mailing list -- time-nuts@lists.febo.com > To unsubscribe, go to > http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com > and follow the instructions there. > -- Kevin Croissant
GE
Gary E. Miller
Sun, Jul 14, 2019 12:13 AM

Yo Kevin!

On Sat, 13 Jul 2019 17:16:37 -0400
Kevin Croissant kevin@kevincroissant.com wrote:

Looking at gpsd, I'm still not sure that it supports UBX,

gpsd has supported u-blox since the SiRF2-ublox TIM chip in 2007.

gpsd now support all the way up to the ZED series.

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
Yo Kevin! On Sat, 13 Jul 2019 17:16:37 -0400 Kevin Croissant <kevin@kevincroissant.com> wrote: > Looking at gpsd, I'm still not sure that it supports UBX, gpsd has supported u-blox since the SiRF2-ublox TIM chip in 2007. gpsd now support all the way up to the ZED series. 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
MW
Michael Wouters
Sun, Jul 14, 2019 12:15 AM

Hello Kevin

There is a Perl script to configure and log the ublox that is
available as part of OpenTTP
https://github.com/openttp/openttp/tree/master/software/gpscv/ublox

Cheers
Michael

On Sun, Jul 14, 2019 at 8:03 AM Kevin Croissant
kevin@kevincroissant.com wrote:

This is the library I am using:
https://github.com/jkua/ubx

Basically the only applicable files are ublox2.py and ubloxMessage.py.
There's some bug in how it handles its byte buffer and the buffer grows out
of control, which causes all sorts of issues. I tried to fix it but failed
due to time constraints. Highly recommend against touching that library...

I started off by using NMEA but found that there is was more resolution in
the UBX messages, so that's why I went that route. I thought there must be
a way to get the same resolution, but did not figure it out, and eventually
I had to just get it online so I could move on. Looking at gpsd, I'm still
not sure that it supports UBX, and I think that was why I disregarded it
before. Do you know if it does? The only UBX messages my project truly
needs are NAV-PVT and NAV-DOP, though I look at other ones for data
validation and debugging.

Kevin

On Sat, Jul 13, 2019 at 5:06 PM Hal Murray hmurray@megapathdsl.net wrote:

The library I used for the U-blox UBX protocol frankly sucks, and was a

major

source of issues.

What library are you using and/or did you look at gpsd?

--
These are my opinions.  I hate spam.


time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe, go to
http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
and follow the instructions there.

--
Kevin Croissant


time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
and follow the instructions there.

Hello Kevin There is a Perl script to configure and log the ublox that is available as part of OpenTTP https://github.com/openttp/openttp/tree/master/software/gpscv/ublox Cheers Michael On Sun, Jul 14, 2019 at 8:03 AM Kevin Croissant <kevin@kevincroissant.com> wrote: > > This is the library I am using: > https://github.com/jkua/ubx > > Basically the only applicable files are ublox2.py and ubloxMessage.py. > There's some bug in how it handles its byte buffer and the buffer grows out > of control, which causes all sorts of issues. I tried to fix it but failed > due to time constraints. Highly recommend against touching that library... > > I started off by using NMEA but found that there is was more resolution in > the UBX messages, so that's why I went that route. I thought there must be > a way to get the same resolution, but did not figure it out, and eventually > I had to just get it online so I could move on. Looking at gpsd, I'm still > not sure that it supports UBX, and I think that was why I disregarded it > before. Do you know if it does? The only UBX messages my project truly > needs are NAV-PVT and NAV-DOP, though I look at other ones for data > validation and debugging. > > Kevin > > On Sat, Jul 13, 2019 at 5:06 PM Hal Murray <hmurray@megapathdsl.net> wrote: > > > > > kevin@kevincroissant.com said: > > > The library I used for the U-blox UBX protocol frankly sucks, and was a > > major > > > source of issues. > > > > What library are you using and/or did you look at gpsd? > > > > > > -- > > These are my opinions. I hate spam. > > > > > > > > > > _______________________________________________ > > time-nuts mailing list -- time-nuts@lists.febo.com > > To unsubscribe, go to > > http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com > > and follow the instructions there. > > > > > -- > Kevin Croissant > _______________________________________________ > time-nuts mailing list -- time-nuts@lists.febo.com > To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com > and follow the instructions there.
KC
Kevin Croissant
Sun, Jul 14, 2019 12:23 AM

Thanks for the info Gary! I will check this out -- I am using the F9P for
another project now, so that might be useful.

Unrelated to gpsd...
I forgot about this until now, but I actually encountered a WNRO bug in the
U-Blox M8N (or possibly my library?) when my data collection was running. I
only have this screenshot now from the incident, but you can also download
data from the website around the WNRO timestamp and see that all
constellations stopped collecting simultaneously.
Screenshot: https://gnssperformancemonitor.com/time-nuts/wnro.png -- look
at "Server Timestamp" to see what time the glitches all actually happened
at.
It seems to me that the cause was that the u-blox M8N looked at GPS for
timestamps, even when you explicitly turn off GPS. I never did any more
investigation into it since the 9th gen was coming out and it shouldn't be
a problem again for a while.

Did anyone else see WNRO glitches with the 8th gen?

Kevin

On Sat, Jul 13, 2019 at 8:13 PM Gary E. Miller gem@rellim.com wrote:

Yo Kevin!

On Sat, 13 Jul 2019 17:16:37 -0400
Kevin Croissant kevin@kevincroissant.com wrote:

Looking at gpsd, I'm still not sure that it supports UBX,

gpsd has supported u-blox since the SiRF2-ublox TIM chip in 2007.

gpsd now support all the way up to the ZED series.

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

--
Kevin Croissant

Thanks for the info Gary! I will check this out -- I am using the F9P for another project now, so that might be useful. Unrelated to gpsd... I forgot about this until now, but I actually encountered a WNRO bug in the U-Blox M8N (or possibly my library?) when my data collection was running. I only have this screenshot now from the incident, but you can also download data from the website around the WNRO timestamp and see that all constellations stopped collecting simultaneously. Screenshot: https://gnssperformancemonitor.com/time-nuts/wnro.png -- look at "Server Timestamp" to see what time the glitches all actually happened at. It seems to me that the cause was that the u-blox M8N looked at GPS for timestamps, even when you explicitly turn off GPS. I never did any more investigation into it since the 9th gen was coming out and it shouldn't be a problem again for a while. Did anyone else see WNRO glitches with the 8th gen? Kevin On Sat, Jul 13, 2019 at 8:13 PM Gary E. Miller <gem@rellim.com> wrote: > Yo Kevin! > > On Sat, 13 Jul 2019 17:16:37 -0400 > Kevin Croissant <kevin@kevincroissant.com> wrote: > > > Looking at gpsd, I'm still not sure that it supports UBX, > > gpsd has supported u-blox since the SiRF2-ublox TIM chip in 2007. > > gpsd now support all the way up to the ZED series. > > 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 > -- Kevin Croissant
BK
Bob kb8tq
Sun, Jul 14, 2019 1:00 AM

Hi

One alternative, if you only need two lines is to write a parser just for them.
There’s not a whole lot to the protocol and the uBlox doc’s are pretty good at
describing it. Yes, it’s a binary protocol so there will be a bit of this and that
involved.

Bob

On Jul 13, 2019, at 5:16 PM, Kevin Croissant kevin@kevincroissant.com wrote:

This is the library I am using:
https://github.com/jkua/ubx

Basically the only applicable files are ublox2.py and ubloxMessage.py.
There's some bug in how it handles its byte buffer and the buffer grows out
of control, which causes all sorts of issues. I tried to fix it but failed
due to time constraints. Highly recommend against touching that library...

I started off by using NMEA but found that there is was more resolution in
the UBX messages, so that's why I went that route. I thought there must be
a way to get the same resolution, but did not figure it out, and eventually
I had to just get it online so I could move on. Looking at gpsd, I'm still
not sure that it supports UBX, and I think that was why I disregarded it
before. Do you know if it does? The only UBX messages my project truly
needs are NAV-PVT and NAV-DOP, though I look at other ones for data
validation and debugging.

Kevin

On Sat, Jul 13, 2019 at 5:06 PM Hal Murray hmurray@megapathdsl.net wrote:

The library I used for the U-blox UBX protocol frankly sucks, and was a

major

source of issues.

What library are you using and/or did you look at gpsd?

--
These are my opinions.  I hate spam.


time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe, go to
http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
and follow the instructions there.

--
Kevin Croissant


time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
and follow the instructions there.

Hi One alternative, if you only need two lines is to write a parser just for them. There’s not a whole lot to the protocol and the uBlox doc’s are pretty good at describing it. Yes, it’s a binary protocol so there will be a bit of this and that involved. Bob > On Jul 13, 2019, at 5:16 PM, Kevin Croissant <kevin@kevincroissant.com> wrote: > > This is the library I am using: > https://github.com/jkua/ubx > > Basically the only applicable files are ublox2.py and ubloxMessage.py. > There's some bug in how it handles its byte buffer and the buffer grows out > of control, which causes all sorts of issues. I tried to fix it but failed > due to time constraints. Highly recommend against touching that library... > > I started off by using NMEA but found that there is was more resolution in > the UBX messages, so that's why I went that route. I thought there must be > a way to get the same resolution, but did not figure it out, and eventually > I had to just get it online so I could move on. Looking at gpsd, I'm still > not sure that it supports UBX, and I think that was why I disregarded it > before. Do you know if it does? The only UBX messages my project truly > needs are NAV-PVT and NAV-DOP, though I look at other ones for data > validation and debugging. > > Kevin > > On Sat, Jul 13, 2019 at 5:06 PM Hal Murray <hmurray@megapathdsl.net> wrote: > >> >> kevin@kevincroissant.com said: >>> The library I used for the U-blox UBX protocol frankly sucks, and was a >> major >>> source of issues. >> >> What library are you using and/or did you look at gpsd? >> >> >> -- >> These are my opinions. I hate spam. >> >> >> >> >> _______________________________________________ >> time-nuts mailing list -- time-nuts@lists.febo.com >> To unsubscribe, go to >> http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com >> and follow the instructions there. >> > > > -- > Kevin Croissant > _______________________________________________ > time-nuts mailing list -- time-nuts@lists.febo.com > To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com > and follow the instructions there.