What's jitter? (1 Viewer)

[PhillJones]

I come accross this term from time to time, often as a justification for expensive digital audio cables. I've tried googling it and checking forums but it seems that the debate on jitter always seems to develop into an EE/transmission line theory, ahem... contest. With the person who knows the most buzz words obviously being right.

So, in simple terms, forget I went to college, what's jitter and why should I care?

Cheers,
Phill

 

[JeremyErwin]

Jitter is clock noise.

CD audio is sampled 44100 times a second, and accurate reproduction of the signal requires a clock that can "tick" precisely every 1/44100 second. A Jittery clock signal is one that deviates from this standard. PCM sent over SPDIF is more vulnerable to jitter than other protocols, such as firewire, hdmi, or usb-audio.

 

[PhillJones]

Thanks the the reply, that's pretty much the idea that I had.

Is this right?

So device A wants to tell device B a number and so sendsa timing pulse and a block of 8 bits (is that right?)

The receiveing device sees the timing pulse, starts counting and then samples the bit stream. But if the clock is jittery it may sample the same bit twice and miss another, eg

10110100 -> 10010100

This results in a slightly incorrect reconstruction and somehow this causes a slight loss of dynamic range when playing back music?

Dolby digital is less sucseptible than PCM for example because of built in error correction.

The bit (no pun intended) that I don't get is if you got any miss-timed bits in your bitstream, it seems to me that, the resulting audio interference wouldn't be a subtle loss of quality but a massive glitch like a very loud cracking noise or something. I must have some missunderstanding somewhere in my idea of how this works.

 

[Jeff Gatie]

As explained, this affects timing, not data loss. How much the effect of "jitter" actually effects what one hears is another "science vs. golden ears" argument. Proof would require a double blind test with a confirmed 'jittery' transport vs. a 'non-jittery' transport. However, that is something that is only going to convince scientists and we'd be back to square one.

MHO, it's about as important to worry about as keeping the same length for all speaker wires. In other words - Don't worry.

 

[PhillJones]

I can see how a jittery clock in your receiver would be bad, as it would cause distortion of the waveform thereby including extra high frequency information.

What I don't get is how jitter as a result of factors external to the clocks in both the receiver and source could cause a problem. Unless resultant jitter is big enough to cause a bit to get completely missed, wouldn't the internal clock of the reciever just result in a reconstructed waveform with the level of distortion associated with the clock miss-match?

Unless the timing of the sample isn't governed by a clock in the receiver but is based on the time at whihc the receiver receives the packet?

Even in that case, I don't see how things like cables could affect it, surely the dispersion curve of a cable is set in stone and so you'd get the same broadening, distortion and delay for every pulse?

Or is that not so?

 

[Jeff Gatie]

Only the most ardent cable defenders use jitter as an excuse for buying "designer" small diameter, plastic clad, stranded copper. Most use it as a reason to spend $10,000+ on a transport.

 

[Dick Knisely]

"Jitter is the deviation in or displacement of some aspect of the pulses in a high-frequency digital signal. As the name suggests, jitter can be thought of as shaky pulses. The deviation can be in terms of amplitude, phase timing, or the width of the signal pulse. Another definition is that it is "the period frequency displacement of the signal from its ideal location." Among the causes of jitter are electromagnetic interference (EMI) and crosstalk with other signals. Jitter can cause a display monitor to flicker; affect the ability of the processor in a personal computer to perform as intended; introduce clicks or other undesired effects in audio signals, and loss of transmitted data between network devices. The amount of allowable jitter depends greatly on the application." "whatis.com"

Cables, like any component that carries or processes the signal, can affect the signal in measurably bad ways. Introducing jitter is theoretically possible and might (I don't know this one for sure) be measureable. The debate is mostly about whether any specific type of change is audible by humans. That debate has raged for as long as there has been "hi fi" gear and will continue to rage on despite any number of double blind tests. IMHO, the majority of these subtle effects come down to the listener believing there's an effect -- perception as reality.

 

[PhillJones]

Cheers guys, I think I'm up to speed on this now.

I guess if a cable was sussceptible to interference, this cause jitter. Or if the pulses weren't perfectly the same each time, then if you do a fourier decomposition, you'd get a slgihtly different spectrum which could propegate differently depending on the dispersion characteristics of the cable. This could I guess amplify the jitter or cause imperfections in pulse shape to cause jitter.

I think I believe that but I'm still not going to buy an expensive cable.

 

[Cees Alons]

In fact, one still can't see how interference or cross-talk could cause "jitter" - and no other audible disturbances.

Digital cables in our homes are relatively short and under pretty normal circumstances, even an inexpensive shielded audio cable cannot possibly have its transmission speeds periodically altered such as to lead to audible jitter.

IMO.


Cees

 

[JeremyErwin]

Two stereophile explanations of jitter.
Jitter and the Digital Interface
The Jitter Game

Before anyone accuses me of hawking overpriced cables, might I quote the first article?


As for the coathanger cable, the experiment measured dolby digital error rates, and not, iirc, pulse code modulated audio. SInce dolby digital is either immune to, or less plagued by jitter related problems, those of us who haven't yet dicarded CDs for high fidelity ac-3 dualdiscs might still find 75-ohm cables to be of some value.