3 ms·
> You are describing a problem where two clocks drift relative to each other, not a problem where one clock drifts away from the actual time The latter is simp
by grkvlt 13y ago
> You are describing a problem where two clocks drift relative to each other, not a problem where one clock drifts away from the actual time
The latter is simply a specific case of the former. The second clock is the one measuring 'actual' time, and the drift is relative to it.
- danbruc 13y agoOf course, but the point was, that a drift relative to actual time does not cause problems as long as all clocks in the system drift at the same rate while the failure was caused by different clocks in the system drifting relative to each other and therefore with different rates relative to actual time. Therefore I treated actual time as a special clock.