[Beowulf] precise synchronization of system clocks

Lux, James P james.p.lux at jpl.nasa.gov
Tue Sep 30 09:22:22 PDT 2008

On 9/30/08 8:55 AM, "Vincent Diepeveen" <diep at xs4all.nl> wrote:

> Hmm,
> The European Union waited a bit long introducing its own equivalent
> and it should
> be online within a few years, it is called Galileo and it is 1 meter
> accurate for everyone.

1 meter without differential corrections of some sort would be quite
1 meter with differential corrections (e.g. WAAS for GPS, or any of a number
of other schemes) is pretty straightforward.

For a differential measurement case (e.g. Surveying, where you want to know
how far apart two things are) the typical performance is a few mm+few ppm.
So over a 1km distance, you can easily know distance between two points to
<1cm. (easily, not cheaply!.. The receivers will set you back a few tens of

This is true for GPS, Glonass, or Galileo.
(BTW, some of the precision service for Galileo is planned to be "fee
based", I understand.. That public/private partnership thing.  For GPS, if
you qualify to use the precision signal, it's free for the taking.)

> Also with the garantuee of it not getting switched off; which happens
> so easily with GPS,
> even the smallest threat of war is enough, and it is difficult to
> follow logics in Washington there.

Almost inconceivable that this would happen these days.  Too many commercial
applications would squawk if it were to be shut down.  Selective
Availability (degraded precision for non-differential measurements) to be
sure, but I think that as a policy thing, they've pretty much abandoned it,
it's too easy to get around.

> On GPS technical data i'm not sure what i'm allowed to quote, so
> let's not do it.
> Let's not even comment how far off or on Robert is.

The original GPS design assumed one could build a receiver to track within 1
PN chip, and at a rate of about 1 Megachip/second, that means 300m
uncertainty. The precise code (P/Y) is 10 times faster, for 30m accuracy
(sufficient for midcourse correction of an ICBM aimed at a hardened target).
Before the spacecraft were even up, tracking to tenths of a chip was readily
possible.  By the 80s, commercial mfrs had figured out ways to exploit the
fact that the two carrier frequencies (L1 carries the C/A code at 1Mchip/sec
and the P/Y at 10Mchip/sec, L2 carries just the P/Y) are phase locked (by
design) to get mm scale precision in surveying applications (just need many
seconds of integration, so there's no real strategic threat.. ICBMs can't
reside at a location for 10 seconds to integrate)

Today, one can spend <$100 to get a GPS receiver that exceeds the
performance of original intent of the design by several orders of magnitude.

For instance, in my garage, I have a surplus Z8301 frequency standard which
is an ovenized quartz oscillator disciplined by GPS so that it's accurate to
around 1E-11 or 1E-12. This was all of a $250 purchase.  As a side effect, I
can use the observables from the GPS receiver inside to potentially measure
the movement of my house as the Pacific plate drifts northward at a couple

Rest assured that I am far from at the limit on homebrew timing fanaticism
(check out the time-nuts list.. Haven't you and the kids always wanted to
experimentally verify General Relativity over the weekend with stuff in your

More information about the Beowulf mailing list