[Thumbgps-devel] Progress on Plain Jane

[Thumbgps-devel] Progress on Plain Jane

[Eric S. Raymond]

NaviSys has agreed to send me three SiRFIII+PL2303 GPSes with a blue-wire
1PPS to DCD_N patch for $30 each.

UniTraq has agreed to send me a SkyTraq+CP-2012 GPS with 1PPS to DCD
connected.  I think this will be blue-wired, but the company rep seems
a bit confused about what is going on and I think she may have
misunderstood her engineers - she said there was "no room in the
enclosure for another PCB trace" and they'd have to send me a bare
board.  I told her that's OK for a test unit but we'll need the
production version to be enclosed.

Both companies seem actively interested in fabricating a production
version. Plane Jane is looking pretty promising.
-- 
		<a href="http://www.catb.org/~esr/">Eric S. Raymond</a>

Re: [Thumbgps-devel] Progress on Plain Jane

[Gary E. Miller]

[-- Attachment #1: Type: text/plain, Size: 478 bytes --]

Yo Eric!

On Tue, 20 Mar 2012 08:17:24 -0400
"Eric S. Raymond" <esr@thyrsus.com> wrote:

> NaviSys has agreed to send me three SiRFIII+PL2303 GPSes with a
> blue-wire 1PPS to DCD_N patch for $30 each.

At that price I'll buy 3 from you.  Then I can test and integrate.

RGDS
GARY
---------------------------------------------------------------------------
Gary E. Miller Rellim 109 NW Wilmington Ave., Suite E, Bend, OR 97701
	gem@rellim.com  Tel:+1(541)382-8588

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

Re: [Thumbgps-devel] Progress on Plain Jane

[Eric S. Raymond]

[-- Attachment #1: Type: text/plain, Size: 691 bytes --]

Gary E. Miller <gem@rellim.com>:
> Yo Eric!
> 
> On Tue, 20 Mar 2012 08:17:24 -0400
> "Eric S. Raymond" <esr@thyrsus.com> wrote:
> 
> > NaviSys has agreed to send me three SiRFIII+PL2303 GPSes with a
> > blue-wire 1PPS to DCD_N patch for $30 each.
> 
> At that price I'll buy 3 from you.  Then I can test and integrate.

Hm.  Well, why don't you buy them from NaviSys, then?  That'll simplify
things, they didn't have to ship to three different locations anyway.

I don't actually need to get one before we have to sign off on the
production design with the actual PCB trace.

I'll pass you the contact info.
-- 
		<a href="http://www.catb.org/~esr/">Eric S. Raymond</a>

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 190 bytes --]

Re: [Thumbgps-devel] Progress on Plain Jane

[Eric S. Raymond]

Gary E. Miller <gem@rellim.com>:
> As you can see the NMEA/USB short term jitter is about 2 milliSec and
> the PPS/USB time is about 0.5 milliSec jitter.  Long term NMEA/USB drift
> is quite a bit larger, maybe 100 milliSec or more.
> 
> So the jitter of PPS over USB is about 50x worse than PPS over GigE, but
> I think for the purposes at hand pretty good.  Of course I expect the
> results to be worse when run on a single core router.

That's excellent news.  0.5msec jitter is as good as we could have
expected and exactly what one would predict from a max error of 1msec
due to polling interval. The long-term NMEA/USB drift is consistent
with previous observations by Hal Murray and myself.

Do you consider the GR-601 with this mod to be ready for production?
Should we give Navisys the green light on GPSD compatibility?  I see
no reason not to at this point, but you're doing the testing so you
make the call.
-- 
		<a href="http://www.catb.org/~esr/">Eric S. Raymond</a>

Re: [Thumbgps-devel] Progress on Plain Jane

[Eric S. Raymond]

[-- Attachment #1: Type: text/plain, Size: 753 bytes --]

Gary E. Miller <gem@rellim.com>:
> > That's excellent news.  0.5msec jitter is as good as we could have
> > expected and exactly what one would predict from a max error of 1msec
> > due to polling interval. The long-term NMEA/USB drift is consistent
> > with previous observations by Hal Murray and myself.
> 
> Nice when theory and practice agree.  I wonder if there is a USB 2.0
> equivalent of the PL2303 which would give us a 125 microSec poll?
> 
> Might be worth asking navisys about that.

Maybe.  But that would require that they qualify a new and probably more
expensive adapter part. We should get this puppy shipping and let them
see some ROI before we push that.
-- 
		<a href="http://www.catb.org/~esr/">Eric S. Raymond</a>

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 190 bytes --]

Re: [Thumbgps-devel] Progress on Plain Jane

[tz]

[-- Attachment #1: Type: text/plain, Size: 2104 bytes --]

If you start to go too exotic, then an Arduino with ethernet shield
emitting a packet right at the pps starts to look good, and the parts are
at Radio Shack.  But any such not in quantity would be more expensive.
There are plenty of generic usb 2.0 chips and acm or cdc is just a set of
bit and endpoint specs.  And any 2.0 chip would need to use an existing
linux driver unless someone plans to do a lot of libusb coding.

The base problem is hardware lameness.  Polling instead of an interrupt pin
that can immediately respond, much less something like an input capture.  I
could probably sync a 2ppm dallas semiconductor i2c real time clock's 1Hz
output, but even then there is nothing to sync that with the internal
clocks.  (I was already inspired to use pps to create an accurate synced
timestamp for j1850 data off my motorcycle - oh, and batchpcb.com can do
small quantities of small boards themselves cheap, but you still need to
build them).

If you have a 2nd usb port and want to get tricky, the pnp/insert/remove
event is likely higher priority, so with a FET...
 On May 1, 2012 9:32 PM, "Eric S. Raymond" <esr@thyrsus.com> wrote:

> Gary E. Miller <gem@rellim.com>:
> > > That's excellent news.  0.5msec jitter is as good as we could have
> > > expected and exactly what one would predict from a max error of 1msec
> > > due to polling interval. The long-term NMEA/USB drift is consistent
> > > with previous observations by Hal Murray and myself.
> >
> > Nice when theory and practice agree.  I wonder if there is a USB 2.0
> > equivalent of the PL2303 which would give us a 125 microSec poll?
> >
> > Might be worth asking navisys about that.
>
> Maybe.  But that would require that they qualify a new and probably more
> expensive adapter part. We should get this puppy shipping and let them
> see some ROI before we push that.
> --
>                <a href="http://www.catb.org/~esr/">Eric S. Raymond</a>
>
> _______________________________________________
> Thumbgps-devel mailing list
> Thumbgps-devel@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/thumbgps-devel
>
>

[-- Attachment #2: Type: text/html, Size: 2767 bytes --]