Originally Posted by machajew
Finally, what exactly is "jitter" ? - this seems to come up quite a bit
Looks like your other questions were answered well, so I won't belabor them.
This one, not so much. ;-)
What is jitter?
Jitter is to me better described as FM distortion. It is a distortion that artificially changes the frequencies contained in an audio signal, as opposed to AM distortion AKA Harmonic and IM distortion, which artificially change the amplitude of an audio signal. In both cases the changes are usually short term although they don't have to be.
One example of FM distortion is Doppler distortion that causes a change in the pitch of a train whistle as it goes by. Another example of FM distortion is the vibrato that musicians apply to their voices as they sing or the instruments that they play.
If you want to hear audible jitter, listen to a LP. It is nearly impossible to play back a LP record without adding audible amounts of flutter and wow which are kinds of jitter. Flutter is a rapid pulsating change in the pitch. Wow is a slower change that is more like what happens to the train whistle I just mentioned except it always comes back to the same frequency.
There are two frequencies that are associated with jitter. One is the frequency of the signal that the jitter is applied to, and the other is the frequency of the artificial change in frequency. For example a violinist applies vibrato to his playing by rapidly moving his hand in such a way as to make quick changes to the tension of the string. The frequency of the signal is the note that he is playing, and the frequency of the jitter is the rate at which he moves his hand this way.
At low and medium frequencies, jitter makes music sound watery. At high frequencies jitter actually adds additional tones that make the music sound more complex or full, but do so in ways that are not harmonically related to the music so it often ends up making the music sound sour. Another term that is used to describe the audible effects of mid and high frequency jitter is roughness which I think is related to the sourness that I previously mentioned.
In digital audio, jitter is due to artificial alterations of the time base of the samples that make up the music. This time base is usually called the clock. One interesting aspect of digital audio is that jitter can always be reduced to vanishing
amounts by regenerating a new clock that is a smoothed out version of the existing clock signal which has some how become unsteady. Of course if jitter is truly massive, the whole signal may be lost.
The current concern over jitter is probably due to the errors in judgement that come from doing sighted evaluations of audio gear. Sighted evaluations, that is comparisons where the listener knows what he is listening to at any moment are well-known to be strongly affected by the state of mind of the listener. If you are feeling bad it is less likely that things are going to sound right. If you are told that something is going to sound a certain way you may believe that it does, even when it actually sounds differently.
Back when digital audio was first being introduced, the first thing that people noticed is that digital recordings generally sounded significantly different from the analog recordings of the day. Compare and contrast how quickly and profoundly CD and DVD took over the marketplace for recordings of music from analog LPs and tapes and video tapes as compared to what is happening with SACD and Blu Ray. SACD completely failed to gain traction in the mainstream, while Blu Ray is only relatively slowly taking over the marketplace for video. In fact physical media itself is in question. This is signficant because transmission of audio and video over networks inherently causes massive amounts of jitter. However this jitter is usually dealt with quite efficiently.
However, people are creatures of risk aversion and resistance to change. A tiny vocal minority starting perceiving bad sound from digtial recordings. They started looking for reasons why digtial which measured to orders of magnitude better performance than analog sounded bad to them. It was easy to show with blind tests at the time that most digital equipment of the day reproduced music in ways that were either far more perfect than analog, or just plain indistinguishable from the source.
However, the doubters continued to complain and it was thought that some previously unmeasured form of distortion was the problem. Fact is that measuring jitter had been done for decades up until this time, particularly as flutter and wow and it was also known that even the best analog equipment either have audible jitter or are close enough to have audible jitter that a little slipup in the ongoing need of analog equipment to be maintained would lead to audible jitter. If anything, the fact that even the digital equipment of the day had very low jitter made it sound different than the cherished and trusted analog gear that people had been using for years.
High end audio had a big problem with digital equipment because it worked too well and digital equipment was quickly becoming too inexpensive to profit from. This cut their flow of cash. Their response was to create a myth about the difficulty of making audio work well in digital environments which they purported to cure by separating the DACs from the transports in high end music players. They further made this myth apparently truthful by often doing this separation of function very badly, technically speaking. A self-fulfilling prophecy was created.
Today, many people who want to sell overpriced new or retro technology use jitter as a justification for more complex equipment and higher prices. If fact the original Sony CDP 101 from 1983, one of the first two digital players marketed to the mainstream in the world has very low jitter. It does have some very minor possibly audible problems, but jitter is not usually one of them.
One current Audiophile Myth relates to the use of optical connections as compared to using coax. Hands are waved and words about reflections in the fiber are thrown about. We hear the J-word (jitter). In fact, all things considered, optical connections are more likely to be trouble free because they are both immune to being sensitive to or they themselves creating electrical interference. Coax connections can do both and in clearly audible ways.