View RSS Feed


Jitter in the digital recording-storage-playback chain.

Rate this Entry
This blog post sums up the influence of jitter in the recording-storage-playback chain in three simple but sound statements. [Some additional insights and reasoning are given in brackets.]

Note: Statement two rules out systematic differences in the playback of bit identical files due to differences in noise stored along with the information (e.g. through varying strengths of the physical representations of 1s and 0s) when playing from RAM.

1) Jitter during analog to digital conversion is embedded into the digital data.

[That means the digital data is not an accurate representation of the analog signal. Nothing can be done (short of trying to second guess the jitter and correct for that - at best a statistical game), once the conversion has been done. And there are other reasons the digital data is not an accurate representation (fx the bit depth necessarily leads to rounding).]

2) As long as no errors in the data are introduced (which is rare and can easily be checked), copying, streaming, uploading, downloading, moving, ... the digital data does not have any influence on jitter.

[Even though the signal is stored by analog levels of something on the media, the reconstruction process will read data into many levels of memory hierarchy, obliterating any systematic differences due to potentially differing analog levels on the media. By design, this type of noise is ignored when moving data in the digital domain.]

3) During playback, many factors can influence the timing of how the data packets are delivered to the DAC, i.e., can introduce jitter.

[How much influence this has on the resulting sound depends to a good degree on how well-designed the DAC and its power supply are.]