Home Tweakers' Asylum

Tweaks for systems, rooms and Do It Yourself (DIY) help. FAQ.

[ Follow Ups ] Thread: [ Display  All  Email ] [ Tweakers' Asylum ]
[ Alert Moderator ]

RE Error Rates and Jitter

[ If the transport mechanism & chipset are capable of delivering a reasonable error rate, the quality of the sound will depend almost completely on the clock & power supply. ]

Pretty big claim you are making. I believe that your understanding of CD playback in the real world is incomplete.

[ SPDIF without a global clock should generally be avoided ]

This describes all the real world CD transport/DAC combo's out there with few exceptions. They do not use a global clock.

[ If the DAC is in the same case, then the transport CAN effect the sound in other ways, by coupling noise and such through the power supply. ]

Indeed. However, with LSI type LIM taking place, even a perfect power supply, a master crystal oscillator locked DAC, etc. there can be jitter artifacts.

[ It should be easy to check the DAC's supply with a scope to verify its quality, though.) ]

And how low is low enough? ALL power supplies have SOME local voltage transients/ground bounces. Is a 0.1 volt deviation acceptable? Why would you think that if you did?

Many CD players have voltage transients that exceed 0.5 volts for the 5 volt supply. I have seen some as high as 1 volt.

See:
http://www.geocities.com/jonrisch/jitter.htm(Note: you may need to go to this URL first, in order for Geocities to allow access:
http://www.geocities.com/jonrisch )

and the reference and citations contained therin.

I will grant that modern CD players have robust enough error correction to avoid actualy changing or dropping bits/words in almost all cases, with most all CD's.

However, the amount of jitter that exsists within a system is not going to be below the theorectical limits of the 16 bit/44kHz playback system. According to Ken Pohlman's tome on Digital Audio, less than 250 pS RMS of jitter is necessary to reach 16 bit resolution at 20 kHz.

According to the various measurements of CD playback systems, only a few high end units approach or exceed this number.

If we look for resolving power that extends beyond the classic text book 16 bit/44kHz limits, as when we use dither, or higher 'sampling' frequencies or bit depths, jitter must be much lower, down in the 10-20 pS range. No currently existing CD playback equipment is capable of jitter levels that low.

Now these are for simple THD/sinewave/first order measurement criteria, this does not include what we might be capable of hearing as sonic degradation due to jitter.

Signal correlated jitter would tend to be self-IMing in nature, rather than causing a simple noise floor limitation as would random jitter. This would make it much more of a problem, more noticable and annoyig, and would depend to some extent on the musical single complexity at any given moment. A simple static sinewave would not cause the same jitter pattern as a complex changing musical signal would.

As I note in the jitter page referred to, everything feeds through the power supply. Even use of a separate transport does not avoid digital signal jitter within the transport from feeding through to the DAC, via the digital interconnect, and the firing pattern of the receiver circuits.

The better systems reduce this kind of feedthrough, but it is never completely eliminated.

Jon Risch


Jon Risch

This post is made possible by the generous support of people like you and our sponsors:
  VH Audio  

Follow Ups Full Thread
Follow Ups
You can not post to an archived thread.

[ Top Top of Page ] [ Contact Us ] [ Support/Wish List ] [ Copyright © 2025, Audio Asylum, all rights reserved ]
[ General ] [ Speakers ] [ Tubes ] [ Vinyl ] [ Digital ] [ Hi-Rez ] [ Video Asylum ] [ Cables ] [ Tweaks/DIY ] [ Music ] [ Films ]