[RE] Reference to the crystal clock accuracy
As I mentioned on telco, ax excerpt
about crystal clock accuracy from Stefano Bregni, Synchronization in Digital
Communication Networks, p 209, last sentence in the section 5.3.2 "Accuracy".
"Quartz-oscillators can be very
stable in the short term, but they drift in frequency and do not feature
the frequency accuracy of atomic clocks. [5.9]"
Something to keep in mind though : clocks
derived from the Crystal oscillators can be deliberately added with jitter
to meet the emission standard requirements. This essentially spreads generated
emission power over a larger frequency range. But I believe that this jitter
is less considerable then the one we get when transmitting timestamps over
the Ethernet.
Max.