On Sat, 24 Jan 2004 09:42:02 +0000 (GMT)
Jim Lesurf wrote:

Yes. In that what really matters is that the data samples are converted
into analogue output at (ideally) perfectly regular intervals. Otherwise
you would be distorting the output by adding a form of phase modulation. If
you can't do this, then all else is a bit futile.

Of course.

No. In that there is a slight risk that jitter might be so bad as to cause
a bit to be lost (or duplicated) on the way to the DAC somewhere.

I really doubt this happens with any regularity, even on spdif ;-)

In practice I assume that decent players/DACs employ suitable buffering,
etc, to be able to ensure that the data is read from the disc reliably.

Wel, an outboard (spdif based) DAC has the problem that it cant feed back to ensure the data is reliable. but yes, for a CD player thats true.

Can cable length increase jitter?

What really matters here is the non-dispersive bandwidth of the cables.

Ok.

The S/PDIF waveform mixes together both the data pattern and a clock
signal.

Yes, I know that.

*IF* the receiver uses the timing of the 'edges' of the waveform as
it clock indication the result is that the apparent clock will then jitter
with the data pattern *even if the source had a perfect clock*.

Ugh. surely no-one tries to use the raw waveform as the clock?

In the event described above, the receiver might then produce irregular
output samples, and the sound becomes modulated by the effect of the finite
bandwidth or dispersion of the cables used, via the daft use of the signal
edges as the clock.

Surely only the cheapest DACs do this?

In most cases in practice, though, I doubt this should matter much. :-)

agreed.

If not, then I doubt it matters where you place the clock.

In fact, for audio use, Im sure it doesnt matter, as Jodrell bank does
radio interferometry using much higher frequencies over far greater
distances, which also requires a synchronised time source.

That is different for various reasons. The key one is that the clock is
kept *seneparate* to any data.

Yes, quite - I was referring at that point to Stewarts 'synchronous bus' system, which also seperates the clock.

The reference clock is sent between antenna
stations in a way that is not 'corrupted' by data transfer.

Would be a bit pointless if it wasnt ;-)

The S/PDIF
system (and various comms systems) puts the clock and data together to use
a single channel.

Yes, I know.

In theory, it does not matter where, physically, you locate the master
clock. However this is 'provided you can pipe it around with no problems
due to noise, etc'.

Yes, that was my question - as to wether cabling

FWIW the longest baselines I know of were back in the Pioneer/Voyager days.
The spacecraft transmitter was phase locked to the signals it received from
Earth (I think the ratio was 222/221 or some-such.) Thus here the phase
lock was acheived over a loop distance greater than the diameter of the
major planets of the solar system! :-)

That is rather cool

or via carrying 'atomic' clocks about (which has
other problems).

Such as relatavistic effects due to being on a plane ;-)

Surely if you design a good atomic clock you have a stable frequency reference, and so all you need to do is build two clocks and adjust the phase?

What strategies do they use to synch atomic clocks?

Or, more to the point, whats the highest frequency they can work with now - I assume the degree of synchronisation defines the effective precision of the baseline, and thus the minimum wavelength they can reliably use for interferometry ?

There the baselines are typically a few thousand km.

I had heard GPS could be used as a frequency reference for this - but have no idea how good it is, beyond knowing its good enough to have to care about relatavistic effects.



--
Spyros lair: http://www.mnementh.co.uk/ |||| Maintainer: arm26 linux

Do not meddle in the affairs of Dragons, for you are tasty and good with ketchup.