[ntp:questions] Is 24PPM an Excessive Real-Time Clock Correction?

Hal Murray hal-usenet at ip-64-139-1-69.sjc.megapath.net
Tue Jul 10 20:06:45 UTC 2007

>A question, please: how can I pick up and localize the clock the ntpd is
>based on (in order to measure the temperature of its crystal). The systems
>used are Linux or FreeBSD.

You could look at the kernel sources.

Most PCs have 2 crystals.  One at 32 KHz is connected to
the battery-backed time-of-day clock chip.  It's also
called the RTC or TOY clock.

The other is 14.xxx MHz (4x color burst, leftover from
one of the very early IBM PCs).  It goes to various PLLs
to make all the other clocks.  Usually there is one
magic clock generator chip that does all the work.

You can also measure the temperature of both crystals and
see which one matches your drift best.

I have an old Linux system with a 2.4 kernel.  It uses
the RTC clock.  A 2.6 kernel on another system uses the 14 MHz
crystal via the TSC.

The latter has a buggy calibration routine.  The drift jumps
around when you reboot.  That might be another test.  Just grep
your system log files for a line like this and see if the CPU speed
is stable:
Jun 30 23:26:15 localhost kernel: [    0.000000] Detected 431.255 MHz processor
Jul  1 14:25:22 localhost kernel: [    0.000000] Detected 431.244 MHz processor
Jul  1 17:41:10 localhost kernel: [    0.000000] Detected 431.243 MHz processor
Jul  7 12:44:16 localhost kernel: [    0.000000] Detected 431.222 MHz processor

These are my opinions, not necessarily my employer's.  I hate spam.

More information about the questions mailing list