On Mon, 28 Mar 2005, Chuck Silvers wrote:

> I tried this too just now, and it didn't seem to make any difference.
> my test machine loses about 10 microseconds per second both when that
> dd pipeline is running and when the machine is idle.

This represents a clock source stability of -10 PPM; that is indeed 
extraordinary for a non-compensated crystal-controlled clock source.

Typical crystals show a +/- 40 PPM short-term performance (not taking aging
or temperature into account, because that would push it even more.)

Good sources, such as the Digitally Compensated Oscillators, can achieve
+/- 0.3 PPM, but I have never seen those used as computer clocks.

In short, 10 PPM seems mighty good!

-Mike