To understand why it's complex we need to remind ourselves of what jitter actually is. It's pitch modulation, basically like analogue wow&flutter, off-centered records, slightly untrue pulleys and vibrating motors in vinyl players, dirt on tape machine capstans etc.. It's nothing more than that in concept.
The only thing about digital timing jitter is that it can spread over a larger rate range. Whereas a vinyl player might modulate as 50 - 60Hz (for motors) + 3Hz or so (eccentricity), because digital is not 'mechanical' and has no 'inertia' it can have a far higher rate of change of frequency range.. Thinking about it a bit you realise immediately that analogue systems typically have dramatically (100's of times) more timing modulation than even the cheapest basic digital system (due to mechanical inaccuracy as such like), it's just that it's rate was slower and therefore less complex - it could wow by a whole musical cents every 3 secs during rotation of an LP and most casual listeners wouldn't even notice - LOL.. Ok so if we were to start drawing a graph of rate of change of timing and annoyance value we would see that it rises as the rate of change increases - it does not matter at all at DC, but matters much more as freq of modulation rise towards the mid and HF ranges.. Ok now there is double whammy; the higher the frequencies in the music itself the greater we are likely to notice the jitter, up until the frequencies are so high we can hardly hear them. So the sensitivity goes up, peaks around the 2KHz mark (highest music notes) and then falls again. So the total effect is a combination of BOTH curves; the rate of the jitter modulation - and the frequencies in the music programme itself.. Hope you get that - visualise the sort graph in your mind. So the amount of peak jitter you can tolerate in normal music drops from loads for LF modulation and Bass instruments down to very little for HF modulation and female vocals etc. This means that the common discussion about hard peak numbers for timing jitter does NOT tell you how damaging it may or may not be at all. It's another example of 'hot air' - not due to massive over-complication (like last time), but in this case massive over-simplification! The figures are therefore all but meaningless :-( Ok so to go back to my DAC recommendations with re-clocking etc, the good DAC which employs a high quality phase locked loop to follow the clock rate, act's in a similar concept to a collosal flywheel (if were on an analogue system). The higher the rate of change of jitter timing - the less it will respond - just exactly what is needed to filter timing jitter in a way where it will NEVER become damaging - whatever rubbish timing jitter gets fed to the DAC :-) It will perform exactly similarly to an SPDIF feed from a cheap consumer player full of jitter as it will for an AES feed from the most expensive system available - it will sound exactly the same :- -- TheOctavist Vortexbox>SBT(TT 3.0)>>Forssell MDAC-2>>>Klein and Hummell 0300D Sota Sapphire/Lyra Kleos>>Bespoke Valve Phono Stage>>Mastersound Due Venti>>Link Audio K100 ------------------------------------------------------------------------ TheOctavist's Profile: http://forums.slimdevices.com/member.php?userid=52700 View this thread: http://forums.slimdevices.com/showthread.php?t=93232 _______________________________________________ audiophiles mailing list audiophiles@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/audiophiles
