What does jitter sound like?

[tfarney]

I believe in jitter, but I have yet to see any published figures that put it firmly within the range of audibility when involving well-implentead DACs of reasonably contemporary design. More importantly, I don't hear it. And that is a small wonder, given that expert descriptions of what to listen for range from glare in the treble to dullness in the treble, to the kinds of esoteric audio descriptions that could be attributed to almost anything if they ever rose above the completely subjective.

 

It would be naive to assume that jitter does not exist. It would be even more naive to assume that most of us, even those of us with pretty highly resolving systems, should not put it way down our list of concerns.

 

Tim

 

[sastusbulbas]

For me in my current set up, jitter can be heard as affecting the bass performance, softening, poor pitch definition, Cymbals and such are rather splashy, with depth of image suffering and things being a little fatiguing. I have addressed this in other systems with variable results, to be honest, less expensive DAC's do seem to perform better and it does seem to vary with better recordings providing better results?

 

Of course this instance is with a rather old PC running spdif via the motherboard, the above is rectified to some degree by running the signal through a Genesis Digital Lense, which makes it a lot better. I could probably improve matters by spending some cash and removing certain items I guess.

 

Quote,

There are too many variables to determine how much jitter is created within a computer.

 

And that is how I feel things are at the moment, in the past I have preferred well constructed CD transports which measure well, but of course my preference was also before newer DACs were manufactured to address jitter and such, these days it seems the opinion is that source matters not if the DAC sorts it out, or that all digital is the same unless you have a rubbish DAC, I think that is naive.

 

My experiment with spdif direct from motherboard was due to an opinion I read stating this would involve less stages so be beneficial, I am unsure how sound this idea is, but my results with this PC combination are not great, and I cannot seeem to change the output frequency from 48khz to 44.4khz which may be detrimental.

I guess adding a reasonable soundcard will improve things, so may fit a M-Audio 192 or such to this project, as I will not need a flying lead for the spdif.

 

Personally I would rather get reasonable jitter,clock, and audio performance from the PC without relying on the Genesis, before deciding to change the DAC which provides satisfactory results with all my other CD equipment.

 

Guesswork, opinion, wheat and chaff come to mind with digital audio these days, as it did in the days of Hi Fi World's computer audio articles.

 

I guess Mac users have it better in some ways, as I said there are less variables and they seem well suited to getting the basics right for audio, with consistent results. No real need to worry about what motherboard MAY affect what and such other trivia. But again I don't have the money to spend on a Mac, and don't prefer Mac, as I can build a far higher spec PC for less to fulfil certain requirements.

 

[tfarney]

Call it guesswork if you like, but I've asked the questions many times in many different ways and contexts, and the view I've received from professionals on both sides of the jitter argument (it is/is not an audible problem) is that the the on/off, zero/one signal of digital audio cannot carry computer system noise or distortion in any form but jitter, so given the hardware and software needed to keep the zeros and ones bit perfect up to conversion, there is absolutely no reason to believe that there would be any difference (beyond jitter) in the signal post conversion to analog. And, yes, I think that loosely translates into "all digital is the same unless you have a rubbish DAC," or at least all bit-perfect digital is the same until you get to the DAC.

 

I don't think that is naive. I think that is the nature of the technology and the opposite of it is applying the common characteristics and occasional myths of analog audio where they do not apply.

 

Tim

 

[Roseval]

I believe in jitter, but I have yet to see any published figures that put it firmly within the range of audibility when involving well-implentead DACs of reasonably contemporary design.

 

I agree 180%

Unfortunately it is very hard to find these figures.

There is a study by Ashihara (2005) claiming 250 ns as the audible threshold.

Others like Adams (1994) states that there is no absolute threshold because different DAC architectures differs in sensibility to jitter. He states that jitter has to be between 1 ns – 20 ps.

Ivar Løkken (2005, probably a lecturer at Oslo university) states: audibility threshold decreases from 500ns at low frequencies to as little as 20ps at 20kHz.

I think it likely that if 250 ns is the threshold, a lot of DACs qualify. If 20 ps is correct, few will do.

Links to the publications can be found here: http://thewelltemperedcomputer.com/KB/BitPerfectJitter.htm

 

 

[sastusbulbas]

 

Quote,

""Call it guesswork if you like, but I've asked the questions many times in many different ways and contexts, and the view I've received from professionals on both sides of the jitter argument (it is/is not an audible problem) is that the the on/off, zero/one signal of digital audio cannot carry computer system noise or distortion in any form but jitter, so given the hardware and software needed to keep the zeros and ones bit perfect up to conversion, there is absolutely no reason to believe that there would be any difference (beyond jitter) in the signal post conversion to analog. And, yes, I think that loosely translates into "all digital is the same unless you have a rubbish DAC," or at least all bit-perfect digital is the same until you get to the DAC.

 

I don't think that is naive. I think that is the nature of the technology and the opposite of it is applying the common characteristics and occasional myths of analogue audio where they do not apply.""

 

 

 

So you conclude jitter can be an issue in a computer, but cannot be audible, unless your DAC is poor?

 

As you state all digital is the same, must we conclude that the ripping, Media player, and source matter not, therefore a Mac is an overpriced piece of FOO? That the Lynx AES16 is a complete rip off for two channel audio and in no way better than an M-Audio, or indeed spdif direct from the motherboard?

 

As I said NAIVE, if jitter was not audible, why would it need addressed?, and why would you need a DAC which addressed it? If my CD transport with low jitter and an accurate clock sounds better to my ears than the computer, why take some backward step and change a DAC I am happy with?

Indeed why did DAC manufacturers do anything about it at all? Why do digital sources not all sound identical? I guess any perceived audible difference between a CD transport playing a CD and a computer playing an EAC wav file of that same CD is purely imaginary?

 

I myself am sure many out there have perfectly acceptable DACs for CD replay, and that many others find variations and preferences between many of the newer DAC's out there. Should I buy a Dacmagic instead of the Apogee Mini Dac as it apparently sounds and measures better, or just because it makes most digital sources sound the same, even though I prefer my old legacy DAC which I think sounds audibly better with well constructed CD transports? Or should I continue to make an effort to improve the specification or apparent sound quality from my PC?

 

To my ears, even with jitter removed, my CD transport outperforms my computers (Macs and PC's), jitter may or may not be the most important thing to consider, but in many systems and with many ancillaries, it may well be a contributing factor to detrimental performance? I am also confident that the computers performance can be improved without having to resort to some mediocre DAC making it all sound the same and hiding the issue.

 

 

 

 

[Roseval]

I don't think that is naive.

I’m afraid I agree again 180%

Pretty often knowledge valid in the analogue domain is applied to the digital domain with highly amusing results.

 

Electrical signals are analogue by definition.

Therefore zero jitter don’t exist (there is no 100% accurate clock)

But in the digital domain this doesn’t matter as long as the jitter is low enough (to high means a timing error, a bit is misread, this is rare).

Copy your files to another disk won’t alter a single bit but the data stream does have jitter.

 

Where it al amount to is indeed the DA conversion.

This of course is again a matter of a clock so jitter is inevitably.

Only and only if the sample rate jitter is to high it will become audible in the analogue domain.

 

 

[tfarney]

"So you conclude jitter can be an issue in a computer, but cannot be audible, unless your DAC is poor?

 

As you state all digital is the same, must we conclude that the ripping, Media player, and source matter not, therefore a Mac is an overpriced piece of FOO?"

 

I conclude, to this point, and the evidence that is not subjective seems to indicate that a pc, configured to be bit-perfect, (and I admit I don't know what needs to be done to achieve this as it has not been necessary for me) will "sound" the same as a Mac. More accurately, I conclude that until it goes through a DAC, it does not "sound" at all. Does that make a Mac FOO, whatever that is? I don't know. I do know that in my business I worked on both platforms and chose the Mac for personal use long before I considered it as a music source. I like the operating system much better. I even liked it much better before Vista.

 

"...if jitter was not audible, why would it need addressed?, and why would you need a DAC which addressed it?"

 

I believe that jitter is not audible in most contemporary systems because it has already been addressed. I'm not sure if this is true of vintage NOS DACs, however.

 

"If my CD transport with low jitter and an accurate clock sounds better to my ears than the computer, why take some backward step and change a DAC I am happy with?"

 

If your cdp and old DAC sound better to you, no you should not change them, regardless of why it sounds better to you. I believe that answers the rest of your questions.

 

Tim

 

[sastusbulbas]

Cheers Tim, LOL, but I want my computer to cost little and sound as good as, if not better than, my CD transports. (Smiley would be here if there were emoticons)

 

 

 

[Tog]

This is getting surreal ...no really ..this reminds me of first year Uni Philosophy classes ..without the cheap Student Union bar prices or the attractive girls. Please remind me we are talking about hifi and of course that we all have qualifications in electronic engineering so that we can be absolutely clear that no one is making this up.

 

I think therefore I am going to talk about something that we can't agree can really be measured or if measured heard? " I may as well walk into Swanky Bros Audiophile emporium and discuss soundstage and my expanding midrange..

 

yours exasperated, tog

 

 

 

[audioengr]

"Perhaps I have oversimplified and overlooked a significant technical issue, can anyone here explain why this approach wouldn't work, or isn't widely used?"

 

This approach works and it is used in a few products, including the Pace-car 2 from Empirical Audio.

 

It isnt widely used for these reasons:

1) There are very few designers in high-end audio or just audio for that matter that have much digital experience

2) Implementation of the circuit is everything. The circuit design, parts choices and even the board layout are critical.

3) transmission-line effects are poorly understood by this design community. Just look at how many poor S/PDIF interface implementations there are. I have modded dozens of thes over the last 8 years.

4) It's expensive to do this right because low-jitter clocks are not cheap, and power systems that enable them to have low jitter are not cheap. Board designs with low-noise are usually not just 2-layer, adding more cost. Most companies are in it for the profit, so they dont go to the ends of the earth to get the best sound quality possible.

 

Steve N.

Empirical Audio

 

[Al]

tog!!!

 

[audioengr]

Tfarney wrote:

"It would be even more naive to assume that most of us, even those of us with pretty highly resolving systems, should not put it way down our list of concerns."

 

If you believe this strongly, then why are you here? Why waste your time? Dont you have better things to do?

 

Steve N.

 

[sastusbulbas]

So it is a load of rubbish! many of you are saying if a digital output has a high level of correlated jitter removing this cannot be audibly beneficial, it cannot be heard. It is nothing to worry about and need not be addressed?

 

 

 

 

 

[DavidJPettifor]

Tfarney wrote:

"It would be even more naive to assume that most of us, even those of us with pretty highly resolving systems, should not put it way down our list of concerns."

 

A perfectly reasonable comment.

 

audioengr wrote:

If you believe this strongly, then why are you here? Why waste your time? Dont you have better things to do?

 

Tim has been answering peoples questions, some directed specifically towards him. That's why he's been participating in this thread. Again, a perfectly reasonable thing to do. It is a forum, after all.

 

--

djp

 

 

 

 

[Ashley James]

Martin Grindrod, my business partner is probably one of the most experienced digital electronics engineers in Audio, but that hasn't stopped much of what I've said being ignored as we continue on a Monster Jitterthon that may well qualify for the Guinness book of Records!

 

He's designed Digital transceivers for Military Aircraft and other systems that used the very first 8 Bit monolithic DACs from Burr Brown nearly thirty years ago! One of his comments was that if you want your car to go faster, you're better off putting a bigger engine in it that reducing its weight by scraping all the paint off. In his view Jitter is the equivalent of paint removal and merely one of many factors to take into consideration when you design a DAC. He also pointed out that the clever people are the Chip designers and that AVI, Emprical Audio, Cambridge Audio and all the others are simply applications engineers whose job is to follow the Chip manufacturers instructions.

 

I'm rapidly approaching the time when I consider a call to Samaritans although Exit is still on a back burner!

 

Ash

 

PS. I've got an MP3 of a Sophie Tucker recording made in 1911 and restored by a friend and it sounds better than some of Amy Winehouse's stuff, so where does this put jitter in the 0-100 scoring system of effect on sound quality.

 

PS. Tim is on here because everyone loves him. He is cheerful friendly, witty, eloquent and an astute observer of human nature as well as being an expert at calming ruffled feathers.

 

[sastusbulbas]

Hi Ashley,

 

It does not change the fact that that removing excessive correlated jitter can improve matters with certain ancillaries. It may not make a huge difference, but once that difference is addressed and the benefits heard it may warrant some attention.

 

Your analogy is pointless, jitter exists. Sadly it is nothing like the paint on a car, nor a car engine representative of an audio system. The capability of a Dac to address jitter does not change the fact that a digital source may or may not benefit from jitter removal, nor indicate the perceived quality of that Dac.

 

My computer benefits from jitter removal, It sounds better. It's a joke to state the Dac must be rubbish due to such. Many Dac manufacturers seem to think it worthwhile addressing an issue many claim to not exist, yet we still see plenty of opinion regarding their performance.

 

Cambridge Audio, along with DCS are I believe two companies that use bespoke software in their Dacs, many Dacs use different implementations with the Dacs they use. I am sure not all Dac's get pulled out a blue skip and put together like a mechano set, pretty sure some Dac manufacturers make an effort in their designs.

 

I agree with your opinions of Tim, LOL, and your OK too. Not sure if this thread would be as much a reason to call the Samaritans as being forced to put up with Amy Whinehouse though!

 

 

 

 

 

 

 

[audioengr]

Ash wrote:

"Empirical Audio, Cambridge Audio and all the others are simply applications engineers whose job is to follow the Chip manufacturers instructions."

 

I am frankly offended and insulted by this comment. Tell me, how do did you become familiar with the way that I design anyway? There is a LOT more to good design than just following the databook recommendations, as you assumed we do here. Do the best chefs in the world just follow the published cookbooks?

 

The way that I design my products and do business is not similar to Cambridge Audio in any way.

 

I have invented more technology than you are apparently aware of. I invented source-synchronous logic signalling and developed many techniques for self-timed logic design. I taught courses on self-timed logic at Intel Corp. I developed the first high-performance massively parallel scalable computer. I was a design lead on the Pentium II. My 22 patents are a matter of record:

 

http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&p=2&u=%2Fnetahtml%2FPTO%2Fsearch-bool.html&r=77&f=G&l=50&co1=AND&d=PTXT&s1=nugent.INNM.&OS=IN/nugent&RS=IN/nugent

 

http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&p=3&u=%2Fnetahtml%2FPTO%2Fsearch-bool.html&r=130&f=G&l=50&co1=AND&d=PTXT&s1=nugent.INNM.&OS=IN/nugent&RS=IN/nugent

 

http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&p=3&u=%2Fnetahtml%2FPTO%2Fsearch-bool.html&r=149&f=G&l=50&co1=AND&d=PTXT&s1=nugent.INNM.&OS=IN/nugent&RS=IN/nugent

 

http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&p=4&u=%2Fnetahtml%2FPTO%2Fsearch-bool.html&r=156&f=G&l=50&co1=AND&d=PTXT&s1=nugent.INNM.&OS=IN/nugent&RS=IN/nugent

 

http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&p=4&u=%2Fnetahtml%2FPTO%2Fsearch-bool.html&r=169&f=G&l=50&co1=AND&d=PTXT&s1=nugent.INNM.&OS=IN/nugent&RS=IN/nugent

 

http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&p=4&u=%2Fnetahtml%2FPTO%2Fsearch-bool.html&r=190&f=G&l=50&co1=AND&d=PTXT&s1=nugent.INNM.&OS=IN/nugent&RS=IN/nugent

 

http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&p=4&u=%2Fnetahtml%2FPTO%2Fsearch-bool.html&r=197&f=G&l=50&co1=AND&d=PTXT&s1=nugent.INNM.&OS=IN/nugent&RS=IN/nugent

 

http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&p=5&u=%2Fnetahtml%2FPTO%2Fsearch-bool.html&r=212&f=G&l=50&co1=AND&d=PTXT&s1=nugent.INNM.&OS=IN/nugent&RS=IN/nugent

 

http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&p=5&u=%2Fnetahtml%2FPTO%2Fsearch-bool.html&r=214&f=G&l=50&co1=AND&d=PTXT&s1=nugent.INNM.&OS=IN/nugent&RS=IN/nugent

 

http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&p=5&u=%2Fnetahtml%2FPTO%2Fsearch-bool.html&r=220&f=G&l=50&co1=AND&d=PTXT&s1=nugent.INNM.&OS=IN/nugent&RS=IN/nugent

 

http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&p=5&u=%2Fnetahtml%2FPTO%2Fsearch-bool.html&r=224&f=G&l=50&co1=AND&d=PTXT&s1=nugent.INNM.&OS=IN/nugent&RS=IN/nugent

 

http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&p=5&u=%2Fnetahtml%2FPTO%2Fsearch-bool.html&r=228&f=G&l=50&co1=AND&d=PTXT&s1=nugent.INNM.&OS=IN/nugent&RS=IN/nugent

 

http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&p=5&u=%2Fnetahtml%2FPTO%2Fsearch-bool.html&r=235&f=G&l=50&co1=AND&d=PTXT&s1=nugent.INNM.&OS=IN/nugent&RS=IN/nugent

 

http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&p=6&u=%2Fnetahtml%2FPTO%2Fsearch-bool.html&r=262&f=G&l=50&co1=AND&d=PTXT&s1=nugent.INNM.&OS=IN/nugent&RS=IN/nugent

 

http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&p=7&u=%2Fnetahtml%2FPTO%2Fsearch-bool.html&r=309&f=G&l=50&co1=AND&d=PTXT&s1=nugent.INNM.&OS=IN/nugent&RS=IN/nugent

 

Steve N.

Empirical Audio

 

[mpmct]

 

http://en.wikipedia.org/wiki/Agnotology

 

 

[cfmsp]

 

 

"but that hasn't stopped much of what I've said being ignored as we continue on a Monster Jitterthon that may well qualify for the Guinness book of Records!"

 

Ashley,

 

there is a consistent sub-theme in your posts, whereby you express (seeming?) frustration that your opinions are not respected to the extent that you desire. You've spoken often of differing levels of knowledge here on CA with the not so subtle message that you think your opinions (and/or those that you agree with) trump all others.

 

You seem to believe that others here should just accept whatever you say and give up their own opinions to the contrary. You also seem to disregard opinions that are contrary to your own.

 

In my experience, the quickest way to gain respect for one's own opinions is to respect the opinions of others.

 

just my opinion, of course, but it seems unrealistic that anyone, let alone a manufacturer, should expect his/her opinions to be accepted by all when the bulk of one's posts (about other's opinions) totally and absolutely reject the opinions being offered here by others.

 

 

clay

 

 

 

 

 

 

 

 

[Ashley James]

Clay

Where possible I try to stick to facts supported by data and anything contentious I run by Martin and or the engineers at LS Design. In the case of jitter I merely explained that it hadn't been a significant issue for some time.

 

Recently someone calling himself CS said the same on this Forum. He is head of research in a significant multinational and based in the UK. He had computer modelled some modern DACs, found that jitter wasn't a significant factor and then modelled the then new Wolfson WM8741 with it's dedicated receiver chip, which he thought a better compromise than SRC. He felt the latter (common practice in Pro Audio) was a sledgehammer to crack a walnut.

 

To take this a stage further we need to consider points made by Tog and Tim. Tim has observed that DACs don't differ much these days and Tog said that he'd bought a more expensive and later one and wasn't sure it was better than its predecessor. Tog's DAC and the Cambridge Audio 840CD are pretty widely regarded as being amongst the best sounding you can get. I suspect you'd struggle to hear much difference between these two and the M-Audio Transit too.

 

These are problems facing hi end in particular and why the audio industry in the UK is having a hard time. CD players barely sell any more and DACs, even in a review in the latest Hi Fi Choice magazine, don't vary very much. Cambridge Audio also make a better power amplifier than many of the specialist manufacturers. As Tim said, the amount of power an Amp has makes the biggest difference and Tog may have noticed that too. It's also important to understand that they nearly all measure and sound worse than the best DACs and that speakers are worse still.

 

So I'd say that there is some subjective consensus to support the facts that I and others (Drdigital as well) have presented and also that the value of this information is in steering audiophiles away from areas of their system where only tiny differences can be made towards the big issues.

 

Let's face it a PS3 is as good a source/streamer as anything you can buy and lots of people already are. Computers are exciting and new in this context and are bringing all our media together in one aesthetically pleasing package if we do it right. Why bog down in minutae like other Forums where months are spent arguing over whether or not a £400 mains cable can improve sound quality.

 

Some fact, some opinion hopefully constructive.

 

Ash

 

 

 

 

[Tog]

I have no patents to my name but I am bright enough to spot bad manners when I see them. Ash and Tim quite clearly have a breadth of experience between them coupled with a jovial approach to this forum. They are simply trying to point out the futility of chasing something that may have as much impact on the sound of your music as miasma had on the spread of disease. It is a sure sign of desperation when some of you resort to rudeness rather than wit.

 

If you really must indulge in such loutish posting or desperate ego massage leave Ash and the good guys alone and hang with the sterophile obsessives at 6moons or enjoythemadness.com.

 

Anyway my wife has solved my immediate jitter issues (if I have any) by spending the dosh I could have used on an empirical pacsetter recloaking device on £200 worth of silk cushions. Still she is happy and maybe I can solve my transcient response issues by using the cushions to isolate my macbook from vibrations.

 

Yours cushioned from jitter, tog

 

 

 

[DavidJPettifor]

Tog wrote:

I have no patents to my name but I am bright enough to spot bad manners when I see them. Ash and Tim quite clearly have a breadth of experience between them coupled with a jovial approach to this forum. They are simply trying to point out the futility of chasing something that may have as much impact on the sound of your music as miasma had on the spread of disease.

 

Totally agree. Well put Tog.

 

Tog wrote:

It is a sure sign of desperation when some of you resort to rudeness rather than wit.

 

They totally lose any argument they may have had when they do. Let's put jitter to bed along with the rudeness and inflated egos.

 

--

djp

 

 

 

 

 

 

[Purite Audio]

Tog I think it is important to remember that Ashley is just a salesman nothing more who is trying to sell his product, he has no technical expertise or background whatsoever, moreover he has been banned from every UK Hi-Fi forum, 'good guy' I think not.

 

[Tog]

That was not the point at all! It seems to me we all have axes to grind but what I despise is the rudeness and arrogance of some of the recent postings. It is hardly a secret that Ash owns a hifi company and he has been very upfront about possible bias. I have heard enough baloney from hifi salesmen to be able to look after myself. I haven't heard anything but common sense so far from Ash or Tim.

 

My point was about rudeness!

 

In addition in a world where the emperor often likes his new clothes covered in pseudo scientific jibber jabber I am wary of being offered expensive solutions to problems that may not exist!

 

Yours respectfully, toy

 

[Roseval]

It’s a problem pestering many forums like these.

Manufacturers twisting any question into a promotion of their products.