Re: DragonflyBSD kernel clock improvements

In message <20040125183335.B30359_at_gamplex.bde.org>, Bruce Evans writes:

>relevant sysctl so that they are all as accurate as possible.  For the
>i8254 clock, the best possible is 0.5 / 1193182 seconds or about 0.5
>ppm, since fractional frequencies can't be set.  0.5 ppm is a good
>practical accuracy to aim for since the frequencies of all clocks on
>all my machines drift with the temperature by about +- 2 ppm every
>day.

It is indeed worth noting that while quartz crystals has just the
right combination of properties to be our only affordable timekeeping
device of any precision (in fact: no other crystals have the right
combination of properties), it is even better at measuring temperature.

For anybody venturing into timekeeping I would really urge you to
start by reading John. R Vig's tutorial "Introduction to Quartz
Frequency Standards" before you attempt anything:

	http://www.ieee-uffc.org/fcmain.asp?view=review

Also, you have to realize that you need _three_ identically good
oscillators for measuring anything, unless you have a GPS receiver
with precision PPS output.

It's an expensive hobby, but very rewarding, for people with an
inclination to physics:

	http://www.leapsecond.com/

-- 
Poul-Henning Kamp       | UNIX since Zilog Zeus 3.20
phk_at_FreeBSD.ORG         | TCP/IP since RFC 956
FreeBSD committer       | BSD since 4.3-tahoe    
Never attribute to malice what can adequately be explained by incompetence.