Accuracy means how far the oscillator might be off from the spec'ed frequency (long-term). A wrong frequency in the master clock will effect the pitch of the audio signal (a recorded 1kHz tone might be reproduced as 999.9999 Hz), but even the cheap standard crystal's specs are just too good to be concerned.
The jitter spec provides an information about very short-time inaccuracies (phase noise), which is much more audible. There are some very good threads with lots of valuable information on jitter. Do a search.
There are different kinds of jitter, and the total RMS jitter spec is usually worthless for evaluation. A CD player clock should be optimized by listening tests, as Elso Kwak did it. There is a schematic of his "Kwak clock 6" on the forum, and I suggest you consider it. It doesn't need exotic parts, and Elso even provides a kit and PCB (as I recall).
The jitter spec provides an information about very short-time inaccuracies (phase noise), which is much more audible. There are some very good threads with lots of valuable information on jitter. Do a search.
There are different kinds of jitter, and the total RMS jitter spec is usually worthless for evaluation. A CD player clock should be optimized by listening tests, as Elso Kwak did it. There is a schematic of his "Kwak clock 6" on the forum, and I suggest you consider it. It doesn't need exotic parts, and Elso even provides a kit and PCB (as I recall).
AMT-freak said:
lower jitter = better
What is lacking in rms spec is
- sigma value
- spectral content
It is not easy to measure though
optimising clocks on perception may lead to a satisfying clock, but it does not have to have lowest jitter - it is just what you are after.......
cheers
Guido
www.TentLabs.com
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.