[time-nuts] GPSDO for my rubber duckie

shalimr9 at gmail.com shalimr9 at gmail.com
Wed Aug 3 02:06:26 UTC 2011


You are entitled to your opinions, but opinions that cannot be expressed with civility like those in the message below have no place on this list.
You are welcome to keep them for yourself.

Didier KO4BB

Sent from my BlackBerry Wireless thingy while I do other things...

-----Original Message-----
From: msokolov at ivan.Harhan.ORG (Michael Sokolov)
Sender: time-nuts-bounces at febo.com
Date: Tue, 2 Aug 2011 19:33:17 
To: <time-nuts at febo.com>
Reply-To: Discussion of precise time and frequency measurement
	<time-nuts at febo.com>
Subject: Re: [time-nuts] GPSDO for my rubber duckie

Chris Albertson <albertson.chris at gmail.com> wrote:

> NTP software can keep system time within a few  milliseconds of UTC.
> No custom hardware, no FPGA or PCB.

But I _*REFUSE*_ to do it that way.  You've mentioned UTC: that's one
thing I'm taking great care to avoid in my solution.  I want my system
to be completely insulated from whatever evil things the ITU may do to
UTC and leap seconds.  By starting from GPS time coming *directly* out
of a GPS receiver via its native EIA-232 port, I can take untampered GPS
time in the week number + time-of-week format, convert it to TAI by
adding a constant 19 s offset, and then convert from TAI to UTR.
GPS time -> TAI -> UTR; there is no UTC involved at any intermediate

Standard NTP software is another thing I wish to avoid like the plague
in this project.  That software has been touched by the hands of people
like PHK and Warner Losh, the same criminals whose handprints are on the
axe that is about to sever most of the world's civil time from the
millennia-old tradition of mean solar time.  Those people are criminals
of the highest degree in my book, and I do not want to use any software
that has been touched by them.

> If the requirement were for nano second level accuracy [...]

I don't care for that level of accuracy, but I do very much care about
the philosophical puriry of the system, end to end and at every
intermediate step.

> but software using Internet pool servers is OK for Milliseconds

I have no idea / don't want to think about what will happen to those NTP
servers when/if leap seconds are killed.  I wish to *insulate* myself
and all computer systems under my care from that insanity.  And even now
while the leap seconds are still with us, NTP does an utter mess in the
vicinity of one.  Once again, I wish to insulate myself from it.

Although my rubber duckie will never act as an NTP client, i.e., will
never ask another NTP server for the time, it *will* act as an NTP
server itself, i.e., it will serve my UTR timescale to the public
Internet.  For as long as the leap seconds are still with us, UTR will
agree with UTC except in the immediate vicinity of one.  However, if and
when the ITU/PHK bastards kill the leap second, the NTP timescale will
fork.  My NTP servers will serve UTR, linked to Earth's rotation just
like the current leap-enabled UTC is.  Don't know / don't care about NTP
servers operated by muggles.

I hope this clarifies a little better what I am after.


time-nuts mailing list -- time-nuts at febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

More information about the time-nuts mailing list