HydrogenAudio

Hydrogenaudio Forum => Scientific Discussion => Topic started by: cabbagerat on 2006-12-27 13:33:06

Title: Audibility of Jitter
Post by: cabbagerat on 2006-12-27 13:33:06
The threshold of audibility of phase noise in ADC and DAC clocks is a fairly contentious issue in the HiFi and audiophile world. Some sources claim that jitter is clearly audible at low levels, and some claim that high levels of jitter are inaudible. The literature describes several tests, many with conflicting results.

One of the chief difficulties in testing the audibility of jitter is that it requires a complex hardware setup, which means that many listeners would be required to be present for an time consuming (and expensive) on site test. Over the last couple of months I have been thinking about organising a distributed listening test to look at the audibility of jitter in audio applications, based on algorithms for simulating the effects of jitter on signals. These algorithms are fairly well described in RF and telecomms engineering literature, and would be interesting for comparison purposes.

The kind of thing I have in mind is this:
Use samples which are accepted to sound good -> simulate jitter -> perform listening tests -> perform more tests at different levels of jitter depending on results

The purpose of this thread is to get ideas of the Hydrogenaudio community about performing these tests. Some of the things I would appreciate input on are:
Title: Audibility of Jitter
Post by: Gigapod on 2006-12-27 15:27:38
The threshold of audibility of phase noise in ADC and DAC clocks
...

I would suggest you begin with a working definition of "Jitter", which you could post to the HA Knowledgebase. Is jitter "phase noise" ? What does it sound like? What does it do to a sine wave that goes through ADCs and DACs ? Does it matter at all ? What is the order of magnitude of jitter in PCs? and in high-end audio equipment?
Title: Audibility of Jitter
Post by: dariju on 2006-12-27 16:07:49
There is a good article about jitter @ Digital Domain:

http://www.digido.com/modules.php?name=New...icle&sid=15 (http://www.digido.com/modules.php?name=News&file=article&sid=15)

I'm looking forward to see such test... How many of you believe in "jitter effect"?
Title: Audibility of Jitter
Post by: cabbagerat on 2006-12-27 16:28:21
I would suggest you begin with a working definition of "Jitter", which you could post to the HA Knowledgebase. Is jitter "phase noise" ?
A good point. I would go with a definition of Jitter as the high frequency component of the time interval error in the clock signal. Where the cutoff of "high frequency" is must be chosen, as must the spectrum of the modelled noise. This spectrum is one very hard problem - it's likely to consist of a variety of intermodulation products, PSU noise, other noise, harmonic distortion and vary widely between different pieces of equipment.
What does it sound like? What does it do to a sine wave that goes through ADCs and DACs ? Does it matter at all ?
Jitter spreads the spectrum of a sine wave out, effectively by convolution with the spectrum of the jitter. I have no idea what it sounds like, except in some extreme cases. And whether it matters at all is one of the questions I am hoping this test will help answer. 
What is the order of magnitude of jitter in PCs? and in high-end audio equipment?
I have no idea about PCs, but total jitter figures for midrange HiFi stuff is generally somewhere between 0.1ns and 100ns - a three order of magnitude range.
Title: Audibility of Jitter
Post by: Gigapod on 2006-12-27 16:35:23
There is a good article about jitter @ Digital Domain:

http://www.digido.com/modules.php?name=New...icle&sid=15 (http://www.digido.com/modules.php?name=News&file=article&sid=15)

I'm looking forward to see such test... How many of you believe in "jitter effect"?


Having just read the article you linked to, I found it a little too much "audiophile"-minded and not very clear about the technical issue of jitter. Quoting: " ...The sonic results of passing this signal through processors that truncate the signal at -110, -105, or -96 dB are: increased "grain" in the image, instruments losing their sharp edges and focus; reduced soundstage width; apparent loss of level causing the listener to want to turn up the monitor level, even though high level signals are reproduced at unity gain..."

I know quartz oscillators have a small amount of phase noise (jitter).
How much does the phase noise in a 12MHz quartz oscillator driving a DAC at 48kHz affects the final waveform?
Title: Audibility of Jitter
Post by: Gigapod on 2006-12-27 17:03:25
... total jitter figures for midrange HiFi stuff is generally somewhere between 0.1ns and 100ns ...

100ns sound a little bit high (for a 10MHz quartz oscillator that's one full cycle) but OK, let's work with that.
How would a 10MHz signal modulation of an audio signal ranging from 20Hz to 20kHz affect the signal?
It seems all the intermodulation products would be way above the audibility range, no? Or am I missing something?
I think a typical figure in an PC-grade audio codec would be 1 ns jitter (approx. 1% jitter in a 10MHz clock), which would represent approximately 0,005% jitter in a 44.1kHz clock (if my napkin calculations are right).
My a priori guess is the ABX test would be unconclusive...

(edit) I just checked the ALC882 codec datasheet and the 24MHz bit clock input is specified with a maximum jitter of 2ns.
The S/PDIF-OUT jitter is specified at 4ns max (6.144MHz clock). I think the S/PDIF-OUT clock is PLL generated internally in the codec.
Title: Audibility of Jitter
Post by: Pio2001 on 2006-12-27 17:39:05
  • Would such a test be useful
  • Would the results of such a test be valid?
  • What sort of test procedure would be best?


Yes it would be useful. Jitter is the main cause invoked in order to explain the alledged sonic difference between the digital output of a 50 € DVD player and a 10,000 € dedicated digital drive.

The results would be valid if several conditions are fullfilled.
On top of my head I can think about
-The kind of jitter. Some kind might be audible, some kind inaudible. We must choose the kind of jitter that should cause the biggest audible effect.
-The test must be performed on a system whose jitter is small compared to the tested amount of jitter.
-The samples used must be sensitive to jitter.

A full scale 20 kHz sine with jitter introducing a 3.5 kHz artifact should be the most sensitive combination.

The most difficult part in this test, in my opinion, will be to get comprehensive jitter analysis from consumer CD players. Especially if jitter is signal dependant.
Jitter should also be analyzed directly at the clock output in order to account for hardware induced jitter. Some claim that heavy error correction or tracking corrections burdens the power supply, which could in turn affect the clock stability.
Title: Audibility of Jitter
Post by: Gigapod on 2006-12-27 17:48:15
...
Jitter should also be analyzed directly at the clock output in order to account for hardware induced jitter.
...

I think you meant "measured" in the phrase above. However, measuring jitter is not a trivial thing, whether we are measuring 0.1ns or even 10ns.
I would dare say that none of us HA readers has the equipment to directly measure jitter with any degree of precision.
Title: Audibility of Jitter
Post by: cabbagerat on 2006-12-27 18:04:04
The results would be valid if several conditions are fullfilled.
On top of my head I can think about
-The kind of jitter. Some kind might be audible, some kind inaudible. We must choose the kind of jitter that should cause the biggest audible effect.
-The test must be performed on a system whose jitter is small compared to the tested amount of jitter.
-The samples used must be sensitive to jitter.
All three of these are difficult problems - and would need careful test design to handle. Possibly starting with something along the lines of "is 1us of jitter audible? in test samples? in music?" and moving down to the harder problems if that test has a positive result would be a good plan. I don't know what the right solution is.

I would dare say that none of us HA readers has the equipment to directly measure jitter with any degree of precision.
I do have access to such equipment when I am at University, and hopefully I can get another student to teach me to use it. Real measurements would be invaluable but, as Pio2001 says, if the jitter is data dependent (and it seems to be in many cases) it becomes  a much harder problem.

Gigapod - you sound like you know quite a lot about jitter/phase noise - thanks for your input so far. Keep it coming 
Title: Audibility of Jitter
Post by: Gigapod on 2006-12-27 18:25:43
...
I do have access to such equipment when I am at University, and hopefully I can get another student to teach me to use it. Real measurements would be invaluable but, as Pio2001 says, if the jitter is data dependent (and it seems to be in many cases) it becomes  a much harder problem.

Gigapod - you sound like you know quite a lot about jitter/phase noise - thanks for your input so far. Keep it coming 

Cabbagerat, I think it's very interesting if you can measure jitter in audio equipment in a well equipped university lab, and later determine through listening tests if it has any audible effect. I think it's always satisfying to debunk a myth with some solid experimental data, and I am highly suspicious of the whole jitter audiophile shebang.
Actually a long time ago I studied precision clocks and did some superficial documentation gathering on quartz oscillators. I think nowadays even the cheapest CD player (like the ones costing 15 euros that you can buy at the supermarket) uses a quartz clock base, because quartz is so cheap. Typically I think the jitter figures will be very low, because even with 25% jitter (highly unlikely) at 10MHz that's only 25ns, but I have no experimental data to offer to support that assertion (lack of equipment, lack of time and vague motivation).
If I may offer some guidance here, I would say, take a single piece of audio equipment (e.g. a normal CD player) to your university, find the quartz oscillator, and check the jitter at the buffered output of the quartz oscillator (not directly at the quartz leads, obviously). Of course schematics would help a lot, but if you can't find the schematics it shouldn't be too difficult to find the quartz oscillator circuitry.
BTW at 10MHz a few pF will significantly affect the waveform, so I recommend a FET active probe with 1GHz bandwidth.
If you find out the jitter is of the order of 1ns, I wouldn't bother with the listening tests...
Title: Audibility of Jitter
Post by: Zster on 2006-12-28 10:31:56

There is a good article about jitter @ Digital Domain:

http://www.digido.com/modules.php?name=New...icle&sid=15 (http://www.digido.com/modules.php?name=News&file=article&sid=15)

I'm looking forward to see such test... How many of you believe in "jitter effect"?


Having just read the article you linked to, I found it a little too much "audiophile"-minded and not very clear about the technical issue of jitter. Quoting: " ...The sonic results of passing this signal through processors that truncate the signal at -110, -105, or -96 dB are: increased "grain" in the image, instruments losing their sharp edges and focus; reduced soundstage width; apparent loss of level causing the listener to want to turn up the monitor level, even though high level signals are reproduced at unity gain..."

I know quartz oscillators have a small amount of phase noise (jitter).
How much does the phase noise in a 12MHz quartz oscillator driving a DAC at 48kHz affects the final waveform?


I agree with your assessment that the article is too audiophile. Here is a more scientific approach by someone who has designed ADC and DAC for MRI and military devices and makes studio quality ADC and DAC’s.

http://www.lavryengineering.com/white_papers/jitter.pdf (http://www.lavryengineering.com/white_papers/jitter.pdf)

unfortunately jitter is very real.
Title: Audibility of Jitter
Post by: Gigapod on 2006-12-28 11:30:50
...
Here is a more scientific approach by someone who has designed ADC and DAC for MRI and military devices and makes studio quality ADC and DAC’s.

http://www.lavryengineering.com/white_papers/jitter.pdf (http://www.lavryengineering.com/white_papers/jitter.pdf)

unfortunately jitter is very real.


Thank you, that was an interesting paper.

I agree with you that jitter is real. However, the author of the paper, Dan Lavry, used FM modulation of sampling rates to simulate jitter. This is definitely not an accurate simulation. Jitter is also called "phase noise" for a good reason: it has a noise profile.

Just a note: your inexpensive PC codec uses a 24MHz clock to drive an internal PLL that itself drives the sampling rate of the A/D and D/A converters. It's next to impossible to FM modulate that internal PLL...

The second graph on page 6 is a more accurate representation of real-life jitter effects, and you can see they are below the noise floor of the signal.

Cabbagerat suggested in a post in another thread that most manufacturers don't quote jitter figures for their audio equipment gear because the effects of jitter can ultimately be measured in noise and THD figures. I think this is correct; I would only add that providing jitter figures directly would be a) difficult (because jitter is difficult to measure and where exactly do you measure it ?) and b) meaningless, as you can only hear the side effects of jitter as added noise or distortion.

Nowadays noise and distortion figures hover well below audible levels, so I confess I am not too worried about jitter, even though it's real... in a sense.
Title: Audibility of Jitter
Post by: cabbagerat on 2006-12-28 19:56:28
Cabbagerat suggested in a post in another thread that most manufacturers don't quote jitter figures for their audio equipment gear because the effects of jitter can ultimately be measured in noise and THD figures. I think this is correct; I would only add that providing jitter figures directly would be a) difficult (because jitter is difficult to measure and where exactly do you measure it ?) and b) meaningless, as you can only hear the side effects of jitter as added noise or distortion.
The problem here is that THD does not measure how bad the distortion sounds. In general, higher THD is worse, but a tube amp fan will tell you that 2% THD doesn't always sound bad. That's well accepted, but it means that audiophiles (and others) will claim that the distortion introduced by jitter is particularly bad.

Lavry's paper is fascinating. However, I agree with Gigapod that his treatment using narrowband noise and FM modulation is too simplistic to be an accurate representation of what happens in reality - the phase noise on quartz oscillators does not resemble the test signals he used. It's the best audio-specific treatment I have read.
Title: Audibility of Jitter
Post by: Gigapod on 2006-12-29 03:37:57
...The problem here is that THD does not measure how bad the distortion sounds. In general, higher THD is worse, but a tube amp fan will tell you that 2% THD doesn't always sound bad. That's well accepted, but it means that audiophiles (and others) will claim that the distortion introduced by jitter is particularly bad.
...

Imho when THD is below 0.01% it matters little whether it has only odd harmonics or only even harmonics: you just can't hear it anyways ("audiophiles" will certainly disagree...). ABX testing of power amps (Transistor x Valve) seems to support this.
Similarly, if the added noise is below the noise floor of a recording. However, note that noise spectrum analysis is required: changing the noise spectrum profile can be audible in some circumstances.

Getting back to the paper by Lavry:
FM modulation of the sampling rate -> I.M. distortion byproducts.
Phase noise (jitter) in the sampling rate -> noise byproducts.
Nothing new, really.

Unfortunately Lavry provides very little in the way of experimental data and perhaps I missed it but I didn't find any mathematical analysis. I am sure one can easily deduce a mathematical formula that will relate the amplitude and frequency of a signal, jitter (or FM modulation) in the sampling rate and noise/IM distortion levels in dB (or %).

I wouldn't be surprised if normal jitter found in commodity-priced audio equipment would have a noise effect at the quantization noise level in 16-bit CD audio. 24-bit audio probably requires a lot more care with jitter. But then 24-bit audio requires a lot more care with everything, doesn't it? 
Title: Audibility of Jitter
Post by: wimms on 2006-12-29 15:00:21
Jitter matters only during slope of the signal. In severe case jitter is causing LSB error. By taking fastest slope, and limiting jitter induced error at 1 LSB, we can find out max acceptable jitter in terms of technical perfection.
For 16-bit system, to keep sample error below 1 LSB, 15KHz signal of 0db can tolerate no more than 324ps of jitter peak.

In other words, we can change the problem into asking how many LSBs in error can we get away with without being audible. Obviously, it does matter whether these errors are white noise-like, signal correlated or alien.

I suggest few papers from:
http://www.essex.ac.uk/ESE/research/audio_...blications.html (http://www.essex.ac.uk/ESE/research/audio_lab/malcolms_publications.html)

C41 IS THE AES/EBU/SPDIF DIGITAL AUDIO INTERFACE FLAWED?, Dunn, C. and Hawksford, M.O.J., 93rd AES Convention, San Francisco, preprint 3360, October 1992

C134  JITTER SIMULATION IN HIGH RESOLUTION DIGITAL AUDIO, Hawksford, M.O.J, 121st AES Convention, San Francisco, October 2006, paper 6864
Title: Audibility of Jitter
Post by: Gigapod on 2006-12-29 15:34:28
...
C134  JITTER SIMULATION IN HIGH RESOLUTION DIGITAL AUDIO, Hawksford, M.O.J, 121st AES Convention, San Francisco, October 2006, paper 6864

Whoaaa!    Some serious math here. That'll take some time to digest, that one!!!!
       
Thanks wimms!
Should I send you the coffee bill or are you willing to lead us through this paper?

(edit) Your sig ->  Very funny!!!
Title: Audibility of Jitter
Post by: cabbagerat on 2006-12-29 17:25:00
The second one of those paper is fascinating. It's the mathematical treatment of jitter that I have been looking for for a while. One excellent quote from the paper is:
Quote
Uncorrelated jitter, although it can result in modulation noise, is generally believed to be more benign compared to jitter that has a correlation with the audio data or has a relationship to a periodic signal such as mains hum.


Much of the complex maths in the paper is derivation of a simplification of the standard bandlimited interpolation formula, with added jitter factors. This optimisation might be necessary for real-time processing, but for offline processing simple truncation can make performance of the unsimplified algorithm acceptable on modern processors. Sections 7 and 8 (one analyzing LPCM and one sigma-delta) are interesting, as are his conclusions.
Title: Audibility of Jitter
Post by: krabapple on 2007-01-12 23:11:47
The second one of those paper is fascinating. It's the mathematical treatment of jitter that I have been looking for for a while. One excellent quote from the paper is:
Quote
Uncorrelated jitter, although it can result in modulation noise, is generally believed to be more benign compared to jitter that has a correlation with the audio data or has a relationship to a periodic signal such as mains hum.


Much of the complex maths in the paper is derivation of a simplification of the standard bandlimited interpolation formula, with added jitter factors. This optimisation might be necessary for real-time processing, but for offline processing simple truncation can make performance of the unsimplified algorithm acceptable on modern processors. Sections 7 and 8 (one analyzing LPCM and one sigma-delta) are interesting, as are his conclusions.



For all its complex maths, there's an unfortunate lack of audibility data in the paper, or even a summary of previous such data,  to indicate how 'important' this ends up being to the consumer ....  so what does it do towards addressing the subject of this thread?
Title: Audibility of Jitter
Post by: Gigapod on 2007-01-12 23:30:19
...

For all its complex maths, there's an unfortunate lack of audibility data in the paper, or even a summary of previous such data,  to indicate how 'important' this ends up being to the consumer ....  so what does it do towards addressing the subject of this thread?


Because - unlike for example harmonic distortion - you can't hear clock jitter directly, as explained in the various papers, you can only hear its effects on the data as it gets converted back from digital to analog.

And as I wrote in my rant above (sorry I am repeating myself, must be getting old), the effects (basically distortion & noise) are usually much below others (for example, noise and HD in the microphone that is used to take the recording in the first place, or quantization noise in 16-bit recordings, etc), even in commodity-grade audio equipment.
Title: Audibility of Jitter
Post by: krabapple on 2007-01-13 03:32:39
well, let me offer this 2005 paper for consideration, then


http://www.jstage.jst.go.jp/article/ast/26/1/50/_pdf (http://www.jstage.jst.go.jp/article/ast/26/1/50/_pdf)


Audibility threshold for timing jitter, for 'golden eared' listeners in a two-alternative forced-choice paradigm using their preferred listening environment and samples:  250 ns.
Title: Audibility of Jitter
Post by: Woodinville on 2007-01-13 07:30:02
well, let me offer this 2005 paper for consideration, then


http://www.jstage.jst.go.jp/article/ast/26/1/50/_pdf (http://www.jstage.jst.go.jp/article/ast/26/1/50/_pdf)


Audibility threshold for timing jitter, for 'golden eared' listeners in a two-alternative forced-choice paradigm using their preferred listening environment and samples:  250 ns.



What was the spectrum of the jitter? That's really a key question, you know.
Title: Audibility of Jitter
Post by: Kees de Visser on 2007-01-13 17:36:00
For the Dutch speaking and jitter interested readers it might be interesting to know that the Dutch AES has planned an evening about "digital clocks" (wordclock distribution in the digital studio). Non-members are welcome too in Utrecht on 25/01/2007. More info can be found here (http://www.aes-section.nl/).
Jitter is one of the topics that will be discussed and demonstrated.
Title: Audibility of Jitter
Post by: DualIP on 2007-01-14 08:14:57
For tests, why not generate a simple program that can simulate jitter?

The proposed program simulates an analog source represented by in.wav, which is AD converted using a non jitterfree convertor

How this can be done (I think):

command format:
simulatejiitter.exe input.wav jitter.wav output.wav

jitter.wav contains values, that determine how much time-offset should be created for the corresponding input.wav sample. Jitter.wav can be given any shape/spectrum.

For instance. jitter sample=0 equals no time off-set, and jitter sample = 32767 results in max positive time offset (for instance 1 complete sample period)

As an example, output samples can be calculated using formula:
out[n]:=in[n] + (in[n+1])-in[n-1])*jitter[n]/65536


For simplicity, this example uses a simple lineair interpolation for approximating signal value near sample
in[n]. A 2nd order function would be already be more accurate.

Best solution would be infinite FIR/ nyquist lowpass, that can calculate any, in between sample value, but that's not what I call simple!

If out.wav is 24bit format, rounding errors are kept out.
Title: Audibility of Jitter
Post by: knutinh on 2007-01-14 16:59:36
Correlated jitter is probably worse than random jitter

High-frequency jitter is probably worse than low-frequency jitter

Any simulation of jitter for detectability threshold will be limited by the playback jitter performance - typically unknown.



By doing detectability measurements, coupling those to models from good measurements, one should be able to decide this once and for all.

Any progress in finding jitter attenuation in "typical" DAC/surround receivers?

What about inserting a RLC network along an spdif path to provoke real spdif-jitter performance of various spdif receivers?

-k
Title: Audibility of Jitter
Post by: cabbagerat on 2007-01-16 08:18:34
For tests, why not generate a simple program that can simulate jitter?
I have a MATLAB program which does this, based on bandlimited interpolation. I would be happy to post the code when I get my home PC out of storage in a few days.
Title: Audibility of Jitter
Post by: Eric Carroll on 2007-03-03 06:07:29
well, let me offer this 2005 paper for consideration, then...

Does anyone else have any additional actual published papers on this topic of the audibility of jitter or listening tests?

What journals cover this topic, if any?

My focus is to understand:
a) given a synthetic jitter profile, is it audible using DBT?
b) given a real jitter environment, is it audible using DBT?
c) can you DBT the difference between toslink and coax?

I have been looking into the basis of the audiophile belief that toslink is broken due to too much jitter and the implicit belief of the audibility of very small amounts of jitter. Given the intensity of the belief, perhaps where there is smoke, there is fire, even though this belief makes no sense to me based on my understanding of jitter and its impacts in this application space.

As far as I understand jitter impacts two things related to audio reproduction:
a) at the level of the synchronous bit transport, it influences bit error rates
b) at the level of the DAC, it causes errors in the recostruction of the waveform. The outcome of this is essentially higher noise and distortion, i.e. you just get a bump in the noise floor related to the waveform being reconstructed - its correlated to what is being reconstructed, which is a wrinkle - and possibly spectral aliases being created. 

Detailed studies of this understanding are also appreciated with actual waveforms & spectrum views.  I also want to be clear I understand the techniques to dejitter a clock, including reclocking and most importantly, buffering. The issue is about impact, not about repair or avoidance.

I don't want any more audiophile "received wisdom" on the issue of jitter, I have received lots of that.

Two papers keep getting cited at me in this discussion. One is the 1992 Stereophile article on jitter (http://www.stereophile.com/reference...ter/index.html), which had a rebuttal in the Audio Critic (http://www.theaudiocritic.com/back_i...ritic_21_r.pdf) and is not my idea of peer reviewed journal article (but has massive audiophile traction). The other is a 1992 AES paper, Is the AESEBU Digital Audio Inteface Flawed? (http://www.scalatech.co.uk/papers/aes93.pdf), 93rd AES Convention, San Francisco. Both of these seem somewhat, well, dated, to me.

The paper previously cited in this thread (http://www.jstage.jst.go.jp/article/ast/26/1/50/_pdf) has actual audibility testing, and appears to set an audibility threshold of jitter of around 250ns.

Papers without audibility studies set the threashold far lower. For example, Dunn's 1992 AES paper (http://www.nanophon.com/audio/jitter92.pdf) claims an audibility threshold of an astonishing 20ps at 20 KHz, based on his 1991 paper "Considerations for Interfacing Digital Audio Equipment to the Standards AES3, AES5, AES11, Proceedings of the 10th International AES Conference, 1991" (paper not yet found online). As another data point, "A Digital Discourse, Dr. Malcolm Hawksford; HiFi News & Record Review Feb,April, June, Aug, 1990" claims a peak jitter threshold of 400ps (also cited by Stereophile). I have not found the actual article yet, just citations and quotes.

Is Dunn's audibility curve an analytic derivation, or an audibility study? Dunn's curve of audibility is widely quoted. Anyone have a copy of this paper?

Others cited, but not yet found (I hesitate to pay the $20 AES paper fee) include "Eric Benjamin and Benjamin Gannon, "Theoretical and Audible Effects of Jitter on Digital Audio Quality", Preprint 4826 of the 105th AES Convention, San Francisco, September 1998" and "The Effects of Sampling Clock Jitter on Nyquist Sampling Analog-to-Digital Converters, and on Oversampling Delta-Sigma ADCs, 87th Convention of the Audio Engineering Society, October, 1989" (also cited by Stereophile).

There appears to be tremendous discussion of jitter measurement, but little understanding of what it actually means. A lot of this appears to me to be very old work and at best analytic, not audability based. None of it considers modern techniques to break the end to end synchronous clocking paradigm although some of them hint at what is now common practice in the telecommunications & Internet space.

Journals that publish in this area or references to further studies would be welcome.

(Zster, thanks for the reference to the Lavry overview paper (http://www.lavryengineering.com/white_papers/jitter.pdf). It is a useful overview document to help explain the impact of jitter in a well written way, and review some of the more modern methods to dejitter signals.)
Title: Audibility of Jitter
Post by: Kees de Visser on 2007-03-03 11:34:16
A small group of experts on this subject (some from this forum) is currently investigating the audibility of jitter.
The idea is to develop a jitter simulation application to enable testing (listening) without a low-jitter DAC.
Alternatively an ultra-low jitter DAC and ADC are available for DBT.
There's no strict time schedule for the project. I'll post updates in this thread.
Title: Audibility of Jitter
Post by: jlohl on 2007-03-04 09:14:33
A small group of experts on this subject (some from this forum) is currently investigating the audibility of jitter.
The idea is to develop a jitter simulation application to enable testing (listening) without a low-jitter DAC.
Alternatively an ultra-low jitter DAC and ADC are available for DBT.
There's no strict time schedule for the project. I'll post updates in this thread.

If anybody is interested to do a listening test in France, I can provide a jitter modulator JM1 from Prism and a QSC ABX comparator for DBT. Get here the manuals (http://www.ohl.to/audio/docs/abx/).
We have to verify that the setup and ABX box doesn't add any jitter (I can get jitter measuring tools).
It's not as pratical as a software simulation but could be a complementary approach.
Title: Audibility of Jitter
Post by: sthayashi on 2007-03-04 10:32:08
Is this the wrong place to ask "What IS Jitter and How does it affect people with digital music?"  And I feel a little stupid asking this question having an EE background.  The most simplistic description of Jitter is, "A Time-based error where a digital clock is not quite accurate,"

Now I've never really known whether typically when people say that, they're referring to an instance where a clock signal was held higher (or lower) longer than it was supposed to, or if a clock is operating at 44.2kHz instead of 44.1kHz.

Most of what I read about jitter issues involves ADCs, which although important, isn't all THAT important for people who prefer to listen to their music (as opposed to folks interested in recording it).

For digital transport, jitter can be a serious killer, but it has to be pretty damn serious before it affects this layer.  Basically your data lines and clock lines have to be out of sync.  By a lot.  If people want to simulate this, I'd recommend writing a program to either randomly delete samples or randomly rewrite a sample to the sample immediately preceding it.  My guess is that this happens as often as cosmic ray memory errors.

Since most applications for people with digital music is in the playback, only the DACs are relevent, and the question must be raised, "What are typical settling times for DACs?"  But I think it also raises another question.  Is Jitter at all relevent to people who don't know what the settling time of their DAC chip is?
Title: Audibility of Jitter
Post by: DonnieW on 2007-03-04 23:52:53
well, let me offer this 2005 paper for consideration, then


http://www.jstage.jst.go.jp/article/ast/26/1/50/_pdf (http://www.jstage.jst.go.jp/article/ast/26/1/50/_pdf)


Audibility threshold for timing jitter, for 'golden eared' listeners in a two-alternative forced-choice paradigm using their preferred listening environment and samples:  250 ns.



I've read this and found it largely reflects the findings of many friends.  Wonder if this can be repeated exactly with a different group?
Title: Audibility of Jitter
Post by: Kees de Visser on 2007-03-05 06:00:57
Quote
Audibility threshold for timing jitter, for 'golden eared' listeners in a two-alternative forced-choice paradigm using their preferred listening environment and samples:  250 ns.
I've read this and found it largely reflects the findings of many friends.  Wonder if this can be repeated exactly with a different group?
Single number jitter specifications don't say much if you don't know the jitter spectrum.
If the effects of jitter are audible at all, it will be interesting to know which part(s) of the spectrum is/are responsible.
Measurements of real world equipment indicate that the jitter spectrum can't be assumed to be flat. Jitter simulation should be able to reproduce this behavior.
Title: Audibility of Jitter
Post by: mcbear on 2007-03-21 10:05:16
For digital transport, jitter can be a serious killer, but it has to be pretty damn serious before it affects this layer.  Basically your data lines and clock lines have to be out of sync.  By a lot.  If people want to simulate this, I'd recommend writing a program to either randomly delete samples or randomly rewrite a sample to the sample immediately preceding it.  My guess is that this happens as often as cosmic ray memory errors.

Since most applications for people with digital music is in the playback, only the DACs are relevent, and the question must be raised, "What are typical settling times for DACs?"  But I think it also raises another question.  Is Jitter at all relevent to people who don't know what the settling time of their DAC chip is?


If the the sampling clock for the DAC (whatever rate it has) has a high jitter (and I guess this thread
is about what is "high"), the sampling is not equidistant in time anymore, thus the sampling theorem
is violated. This has an effect on the spectrum and the reconstructed signal. Question is, when is it
audible...?
The effect of jitter being so high that actually data is lost is not part of the discussion.
Title: Audibility of Jitter
Post by: 2Bdecided on 2007-03-21 11:31:27
The paper previously cited in this thread (http://www.jstage.jst.go.jp/article/ast/26/1/50/_pdf) has actual audibility testing, and appears to set an audibility threshold of jitter of around 250ns.

Papers without audibility studies set the threashold far lower. For example, Dunn's 1992 AES paper (http://www.nanophon.com/audio/jitter92.pdf) claims an audibility threshold of an astonishing 20ps at 20 KHz, based on his 1991 paper "Considerations for Interfacing Digital Audio Equipment to the Standards AES3, AES5, AES11, Proceedings of the 10th International AES Conference, 1991" (paper not yet found online). As another data point, "A Digital Discourse, Dr. Malcolm Hawksford; HiFi News & Record Review Feb,April, June, Aug, 1990" claims a peak jitter threshold of 400ps (also cited by Stereophile). I have not found the actual article yet, just citations and quotes.

Is Dunn's audibility curve an analytic derivation, or an audibility study? Dunn's curve of audibility is widely quoted. Anyone have a copy of this paper?

Others cited, but not yet found (I hesitate to pay the $20 AES paper fee) include "Eric Benjamin and Benjamin Gannon, "Theoretical and Audible Effects of Jitter on Digital Audio Quality", Preprint 4826 of the 105th AES Convention, San Francisco, September 1998" and "The Effects of Sampling Clock Jitter on Nyquist Sampling Analog-to-Digital Converters, and on Oversampling Delta-Sigma ADCs, 87th Convention of the Audio Engineering Society, October, 1989" (also cited by Stereophile).


250ns and 400ps are not contradictory. One is saying "we've tested it subjectively - at around 250ns it starts to become audible". The other is saying "from first principles, if you keep it below 400ps, for the most sensitive possible signals, it will have a smaller impact on the signal than the limits of the system itself (i.e. the sample rate/bandwidth and bitdepth)".

These are two different approaches to audio engineering. One says "we can make it as bad as we want as long as no one can hear it". The other says "we will make it as good as we can to the point where this part can never be the limiting factor".

The real world has to sit between the two. You can't engineer something so that nothing is the limiting factor! You have to have some understanding of human ears to know when to stop improving everything (or be permanently depressed that nothing is good enough).

Conversely, you can't make everything "as bad as it can be before it causes an audible problem" because if you chain all these separate things together you can be fairly sure that you will have an audible problem at the end!


I like the idea of an experiment, but I don't see how a typical HA public test can work or be valid. We're all listening with unknown levels of jitter.

It would be like testing the audibility of -120dB of noise while we're all listening with soundcards which add noise at somewhere between -108dB and -60dB. Not hearing the -120dB of noise through these sound cards proves nothing. The same is true of jitter.

Cheers,
David.
Title: Audibility of Jitter
Post by: udauda on 2009-08-14 22:12:03
Here's another study on the threshold of jitter:
(http://www.sea-acustica.es/Sevilla02/mus05001.pdf)http://www.sea-acustica.es/Sevilla02/mus05001.pdf (http://www.sea-acustica.es/Sevilla02/mus05001.pdf)
Title: Audibility of Jitter
Post by: udauda on 2009-08-14 22:45:21
http://www.jstage.jst.go.jp/article/ast/26/1/50/_pdf (http://www.jstage.jst.go.jp/article/ast/26/1/50/_pdf)

The authors of the paper above, Dr. Ashihara & Dr. Kiryu answered questions(1 (http://www.head-fi.org/forums/f7/interesting-paper-jitter-audibility-211926/#post2551882) , 2 (http://www.head-fi.org/forums/f7/interesting-paper-jitter-audibility-211926/index4.html#post2558434)) raised by Steve N. at Empirical Audio (http://www.empiricalaudio.com/). These exceptions are from hciman77 (http://www.head-fi.org/forums/f7/interesting-paper-jitter-audibility-211926/)(Jim), who received the answers via email:


Quote from: hciman77 link=msg=2557920 date=
Dear Jim,

Thank you for the e-mail. I suppose that you read our paper titled
'Detection threshold for distortions due to jitter on digital
audio(http://www.jstage.jst.go.jp/article/ast/26/1/50/_pdf).' (http://www.jstage.jst.go.jp/article/ast/26/1/50/_pdf)). Before this
paper was published in the Acoustical Science and Technology, we had
published another paper 'The maximum permissible size and detection
threshold of time jitter on digital audio.' Unfortunately, it was written in
Japanese.

In our first experiment, which was reported in the Japanese paper, we used a
fixed listening condition and fixed materials. All of 14 participants were
university students without any special training. The audio system that we
used consisted of the following equipment.

D/A converter --- SEK'D ADDA2496S
preAmp. --- Luxman C-7
main Amp. --- Luxman M-7
loudspeakers --- DIATONE DS-205

They costed about $10,000. I don't know if they belong to high-end or not.

All participants could distinguish between sounds with and without time
jitter when the jitter size was 9216 ns. A few could when it was 1152 ns. No
one could when it was as small as 576 ns.

There was a question, however, if the result would depend on the listening
environments and the skill of the listeners. That is why we carried on the
second experiment. This second experiment is reported in the paper, the one
that you probably read.

Listeners in the second experiment were all professionals, audio engineers,
recording/mixing engineers, musicians, etc... Sound materials were selected
by the listeners so that each listener could use his (her) familiar
materials. The experimenter (we) visited the listeners' studios or listening
rooms so that we could use listeners' own DAC, amplifiers, loudspeakers and
headphones. The system configurations, therefore, varied among listeners.
They were mostly mid-end or above, I suppose.

As you can find in the paper, some listeners could distinguish the sounds
when time jitter was 500 ns. It could not be detected, however, when the
jitter was as small as 250 ns.

In both experiments, there was considerable difference in listeners'
performance. I don't know, however, if it was because of their audio
experience. We had expected much better performance in the second experiment
because the listeners were professionals and they could use their favorite
environments and materials.

Our conclusion up to now is that the normal hearing listeners' detection
threshold for time jitter in program materials is several hundred ns.

I appreciate that you are interested in our paper. Thank you for asking
questions.

Best wish

--
ASHIHARA Kaoru

.
.
.

Quote from: hciman77 link=msg=0 date=
Hello Jim,

Additional comments came from my co-worker, Dr. Kiryu.
Prior to the second experiment, we had sent the materials with time jitter
of several amounts to some of the participants. They could, therefore, train
themselves with the materials. In fact, one listener told us that he could
detect time jitter of several ns. However, in the experiment that was a
strict double-blind test, his score was much worse as written in the paper.
Recently, the materials were sent to Dr. Kiryu's friend who is an
audiophile. This man said that the sufficient training made it possible to
detect time jitter of 150 ns.

I want to add some more.
We had considered about the maximum permissible jitter in audio package
media. When random time jitter does not cause any distortions larger than
1/2 LSB (Least Significant Bit), it does not degrade the quality of sounds
because the distortions in this case are smaller than the quantization noise
level. When there is time jitter, the maximum distortion occurs where the
slope of the waveform is its maximum. The size of distortion can be obtained
by multiping the slope by jitter size. Does it make sense to you? My English
may be awkward sometimes. Please make it up for with your imagination.
Anyway, if you can find the maximum slope (inclination?) in the waveforms of
the sound materials, you can estimate the maximum permissible jitter size.
We estimated the maximum permissible jitter size by checking the maximum
slope in the music waveforms in many CDs. The values varied considerably
between 182 ps and 2567 ps. This means that in certain materials, time
jitter has to be smaller than 182 ps to guarantee a 16-bit resolution.

In our research, the material that was most susceptible to jitter was a
music played by a music box and it contained a lot of high-frequency
components. One hundred eighty-two ps correspond to the maximum permissible
jitter in a pure tone of about 13.3 kHz. If a 20 kHz pure tone has to be
reproduced with a resolution of 16 bit, the maximum permissible jitter size
is about 121 ps. By the way, I do not know any loudspeakers or headphones
that have linearity corresponds to a 16 bit resolution. After all, I won't
believe someone who says that he can detect time jitter of 100 ps or less in
CDs.

I visited the site of Headfi.org and found your heated discussion on this
topic.

Keep going!

-----------------------------------------
Title: Audibility of Jitter
Post by: Arnold B. Krueger on 2009-08-16 11:50:13
Here's another study on the threshold of jitter:
(http://www.sea-acustica.es/Sevilla02/mus05001.pdf)http://www.sea-acustica.es/Sevilla02/mus05001.pdf (http://www.sea-acustica.es/Sevilla02/mus05001.pdf)


That's a horrible paper. I'd be embarassed for the authors, let alone cite it. These guys demonstrated have zero clue about how to do proper subjective testing.
Title: Audibility of Jitter
Post by: itisljar on 2010-02-09 18:21:25
Well, bumping the old thread - has the test been conducted, and where are the results?
Title: Audibility of Jitter
Post by: John_Siau on 2010-05-24 16:25:52
We had considered about the maximum permissible jitter in audio package
media. When random time jitter does not cause any distortions larger than
1/2 LSB (Least Significant Bit), it does not degrade the quality of sounds
because the distortions in this case are smaller than the quantization noise
level.


The above statement is incorrect.  The quantization noise of a digital audio system does not fully mask all distortion that is at a lower level.  This is a common misconception that ignores masking theory.

For example, a 3 kHz tone can be heard as much as 30 dB below the level of a white noise signal if the level of the 3 kHz tone exceeds the threshold of hearing.

Our ears are very good at detecting tones at levels that are below the level of the ambient noise.

This does not mean that the results of their paper are necessarily incorrect, it just shows that they ignored masking theory when speculating what distortion levels may be audible.

Let me say it again - just to be very clear: 
1) The -96 dB FS quantization noise of a 16-bit CD system does not magically mask all distortion and music that falls below -96 dB FS.
2) 3 kHz tones can be heard in a 16-bit TPDF system down to levels as low as about -126 dB FS if system gain is turned up enough to allow the tone to be reproduced at a level that is sufficiently above the threhold of hearing.

Notes:
I intentionally selected 3 kHz - worst case (see Fletcher-Munson)
Title: Audibility of Jitter
Post by: pdq on 2010-05-24 16:57:01
John, the post that you quote refers specifically to RANDOM jitter. How can this possibly produce anything remotely resembling a 3 kHz tone?
Title: Audibility of Jitter
Post by: John_Siau on 2010-05-24 18:15:19
John, the post that you quote refers specifically to RANDOM jitter. How can this possibly produce anything remotely resembling a 3 kHz tone?

Thanks for the correction - I failed to note that their paper was confined to random jitter. The distortion caused by this random jitter should be much less audible than the distortion caused by siusoidal jitter. 

To reach audibility, the distortion caused by random jitter may need to be 20 to 30 dB higher than the distortion caused by sinusoidal jitter.

The "random jitter" used in this experiment is frequency limited by the Nyquist theorem.  Consequently, the jitter-induced distortion will have nearly the same spectral shape as the jitter.  If the spectrum of the band-limited random jitter is white, we should expect the spectrum of the jitter-induced distortion to be nearly white.  TPDF dither noise will be very effective at masking this spectrally-white jitter-induced distortion.  If the jitter-induced distortion is the same amplitude as 16-bit TPDF dither noise, the system noise level will increase by 3 dB.  If the jitter-induced distortion is 6 dB lower than the 16-bit TPDF noise, system noise will increase by 1 dB.  In this experiment, the jitter-induced distortion is simply a white noise signal that gets added to the system noise. 

Note: Use RMS noise summing equations to calculate resulting noise.

Digital audio transmission systems tend to generate jitter at very specific frequencies.  The spectrum of the code-induced jitter at the end of a S/PIF cable is much closer to sinusoidal than random.  Spectrally white random jitter is not likely to occur in the real world.  Jitter composed one or two dominant sinusoidal frequencies is much more common.  In my opinion it is more important to investigate the audibility thresholds for sinusoidal jitter. 

Obviously the investigation of random jitter is a good first step as it requires far fewer tests than an investigation of random jitter.  With random jitter we have one variable - amplitude.  An investigation of sinusoidal jitter would require two variables - amplitude and frequency.  Many tests would be required to plot the audibility curves.

We should be able to estimate the audibility of sinusoidal jitter-induced distortion using masking theory.  Has anyone published these calculations?

Are there any good papers on the audibility of sinsoidal jitter?
Title: Audibility of Jitter
Post by: John_Siau on 2010-05-24 20:48:20
We should be able to estimate the audibility of sinusoidal jitter-induced distortion using masking theory.  Has anyone published these calculations?

Are there any good papers on the audibility of sinusoidal jitter?



Julian Dunn calculated the audibility threshold of jitter-induced sidebands produced by sinusoidal jitter.  He took making effects into account when calculating audibility.  His calculations are based upon a peak playback level of 120dB SPL and he assumes that un-masked sidebands become audible at 0 dB SPL. 

Peak playback levels are usually lower than 120 dB SPL, and audibility thresholds will usually be slightly higher than 0 dB SPL (due to ambient noise), so his jitter-audibility plot is a worst-case audibility plot.  These results could be scaled for other playback levels and ambient noise levels.

http://citeseerx.ist.psu.edu/viewdoc/downl...p1&type=pdf (http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.63.7090&rep=rep1&type=pdf)

See section 3.3 for an explanation, and figure 9 for a plot of "maximum inaudible jitter amplitude" vs frequency.

In summary of Julian Dunn's calculations:
1us at jitter frequencies below 200 Hz should be inaudible
1ns at a jitter-frequency of 600 Hz should be inaudible
100 ps at a jitter-frequency of about 3 kHz should be inaudible
20 ps a jitter-frequency of 20 kHz should be inaudible


A detailed paper on the derivation of theses numbers can be found here:

http://www.aes.org/e-lib/browse.cfm?elib=6111 (http://www.aes.org/e-lib/browse.cfm?elib=6111)

Title: Audibility of Jitter
Post by: 2Bdecided on 2010-05-25 09:12:07
His calculations are based upon a peak playback level of 120dB SPL and he assumes that un-masked sidebands become audible at 0 dB SPL.
Can I explain this in suitably scientific language?

It's taking the Michael.


Quote
See section 3.3 for an explanation, and figure 9 for a plot of "maximum inaudible jitter amplitude" vs frequency.

In summary of Julian Dunn's calculations:
1us at jitter frequencies below 200 Hz should be inaudible
1ns at a jitter-frequency of 600 Hz should be inaudible
100 ps at a jitter-frequency of about 3 kHz should be inaudible
20 ps a jitter-frequency of 20 kHz should be inaudible
So you need less than 20ps jitter - otherwise, when you play back 20kHz sine wave at 120dB SPL, the jitter-induced noise might have a total power equivalent to 0dB SPL.


Quote
A detailed paper on the derivation of theses numbers can be found here:

http://www.aes.org/e-lib/browse.cfm?elib=6111 (http://www.aes.org/e-lib/browse.cfm?elib=6111)
I didn't even need to look - the derivation is obvious - it's the starting point that's silly. (IMO - YMMV!)

Cheers,
David.
Title: Audibility of Jitter
Post by: Arnold B. Krueger on 2010-05-25 14:04:40
His calculations are based upon a peak playback level of 120dB SPL and he assumes that un-masked sidebands become audible at 0 dB SPL.
Can I explain this in suitably scientific language?

It's taking the Michael.


Quote
See section 3.3 for an explanation, and figure 9 for a plot of "maximum inaudible jitter amplitude" vs frequency.

In summary of Julian Dunn's calculations:
1us at jitter frequencies below 200 Hz should be inaudible
1ns at a jitter-frequency of 600 Hz should be inaudible
100 ps at a jitter-frequency of about 3 kHz should be inaudible
20 ps a jitter-frequency of 20 kHz should be inaudible
So you need less than 20ps jitter - otherwise, when you play back 20kHz sine wave at 120dB SPL, the jitter-induced noise might have a total power equivalent to 0dB SPL.


The first fallacy here is the idea that the human threshold of hearing remains at 0 dB while a human is listening to 20 Hz at 120 dB. IOW, there is a presumption that theshold shifts never happen, even in the presence of 120 dB sounds.

The second fallacy is that there would ever be a natural sound that is a 120 dB 20 Hz pure tone with all other sounds 120 dB down.

The third fallacy is that there is anybody actually listens to reproduced sound in a context where the listening environment's residual noise is at 0 dB or below, other than as part of a lab esperiment.

I think I'll stop now! ;-)
Title: Audibility of Jitter
Post by: John_Siau on 2010-05-25 18:09:00
The first fallacy here is the idea that the human threshold of hearing remains at 0 dB while a human is listening to 20 Hz at 120 dB. IOW, there is a presumption that theshold shifts never happen, even in the presence of 120 dB sounds.

The second fallacy is that there would ever be a natural sound that is a 120 dB 20 Hz pure tone with all other sounds 120 dB down.

The third fallacy is that there is anybody actually listens to reproduced sound in a context where the listening environment's residual noise is at 0 dB or below, other than as part of a lab esperiment.


I don't disagree with any of these comments.  Julian Dunn never claimed that jitter becomes audible at these levels, just that we could guarantee that it would be inaudible if held below these levels.

In my post, I suggested that we could scale Julian Dunn's calculations to reasonable listening situations.  I was not implying that these levels were typical, or that full-scale 20 kHz tones are found in any music recordings.

With a little more work, we could also adjust Julian's calculations for an audio input spectrum that is more typical of music.

What must be understood is that random jitter is also an extrreme case that also NOT typical of real audio hardware.  For this reason, the random jitter audibility test results are as unrealistic as Julian's graph.  I suspect realistic threholds for typical "real-world" jitter spectra will fall somewhere in between these extreeme cases.

Both papers provide good resources for finding a more realistic answer.
Title: Audibility of Jitter
Post by: Arnold B. Krueger on 2010-05-26 16:09:26
What must be understood is that random jitter is also an extreme case that also NOT typical of real audio hardware.  For this reason, the random jitter audibility test results are as unrealistic as Julian's graph.


I agree that random jitter is one of the extreme cases. It should in general be the hardest to hear because it lacks tonality. Its cause is likely to be irreducable thermal noise.

IME most jitter traces back to the environment or the process.  Power supply frequencies are common. Jitter at the same rate as data blocks is not uncommon. Then there are the cases where the signal jitters itself.

Quote
I suspect realistic threholds for typical "real-world" jitter spectra will fall somewhere in between these extreme cases.


So far I see none of the cases where jitter is really hard to discern, just different cases that are extremely small.

When all possible forms of masking are consdered, the amount of jitter that can be masked can be huge. I'm not sure that I can support those cases being used to calculate an average of extreme cases.

We hear very little about people hearing jitter during LP playback, yet there is jitter in LP playback that is commonly less than 60 dB down.

Quote
Both papers provide good resources for finding a more realistic answer.


They may be a starting point.
Title: Audibility of Jitter
Post by: Woodinville on 2010-05-26 18:36:01
We hear very little about people hearing jitter during LP playback, yet there is jitter in LP playback that is commonly less than 60 dB down.


Like, WOW, man!

I keep pointing out to people that .55555_Hz jitter isn't that bad.
Title: Audibility of Jitter
Post by: Arnold B. Krueger on 2010-05-26 20:06:15
We hear very little about people hearing jitter during LP playback, yet there is jitter in LP playback that is commonly less than 60 dB down.


Like, WOW, man!


LOL!

Quote
I keep pointing out to people that .55555_Hz jitter isn't that bad.


Unfortunately  .55555 Hz  isn't the only kind of jitter that LPs have.  Anything that causes the record to be other than perfectly flat creates FM distorition. Then there is the FM distortion that is inherent in offset arms. It FM modulates everything with whatever bass is present. LPs also pick up whatever jitter the analog tape had, which includes capstan and scrape flutter that can go up into the 100s of Hz.

Needless to say, the high end reviewers seem to have cultivated selective deafness to the many audible flaws that are inherent in the LP format.
Title: Audibility of Jitter
Post by: udauda on 2010-05-27 05:58:32
BTW I found a scientific paper about this. It is called:
Golden Ears and Meter Readers: The Contest for Epistemic Authority in Audiophilia


I just obtained a copy- and it is rather a rhetorical paper that gives an overview on the fight between subjectivists & objectivists, and its long history. Thnx for the recommendation tho.

By the way, I found this powerpoint presentation (http://redirectingat.com/?id=475X763&xs=1&url=http%3A%2F%2Fmentor.ieee.org%2F802.11%2Fdcn%2F04%2F11-04-1458-00-000n-jitter-requirements.ppt&sref=http%3A%2F%2Fwww.head-fi.org%2Fforum%2Fthread%2F492989%2Fdigital-cable-make-a-difference%2F30%23post_6664151) which goes over most of the studies related the audibility of clock-jitter. Since I found the info extremely valuable, I made a excerption out of it:
(http://i78.photobucket.com/albums/j92/udauda/jit.png)

Hopefully the authors don't get angry at me because of the unauthorized excerption..
Title: Audibility of Jitter
Post by: Arnold B. Krueger on 2010-05-27 11:48:45
How many of the references that went into this page meet TOS 8?

2?
Title: Audibility of Jitter
Post by: udauda on 2010-05-27 14:00:45
How many of the references that went into this page meet TOS 8?
2?


No need for sarcasm. 

[1] yes- and it is quite well-known.
[2] yes- done by Dolby.
[3] objective measurements only- no need to meet TOS 8.
[4] yes (http://www.jstage.jst.go.jp/article/ast/26/1/50/_pdf)- and we have discussed it on this thread. Also there is an addendum from the authors (http://www.hydrogenaudio.org/forums/index.php?showtopic=51322&view=findpost&p=651585).
[5] not referenced above- no need to meet TOS 8.
[6] no- written in 1970, well before David Clark's test in 1982.
Title: Audibility of Jitter
Post by: Arnold B. Krueger on 2010-05-27 22:12:11
How many of the references that went into this page meet TOS 8?
2?




Quote
[1] yes- and it is quite well-known.

OK

Quote
[2] yes- done by Dolby.

OK

Quote
[3] objective measurements only- no need to meet TOS 8.

IOW, it sheds no light on audibility

Quote
[4] yes (http://www.jstage.jst.go.jp/article/ast/26/1/50/_pdf)- and we have discussed it on this thread. Also there is an addendum from the authors (http://www.hydrogenaudio.org/forums/index.php?showtopic=51322&view=findpost&p=651585).

The first link is broken, the second link's contents sheds no light on audibility

Quote
[5] not referenced above- no need to meet TOS 8.

IOW, it sheds no light on audibility

Quote
[6] no- written in 1970, well before David Clark's test in 1982.


IOW, it sheds no light on audibility.
Title: Audibility of Jitter
Post by: Soap on 2010-05-28 00:06:44
Quote
[4] yes (http://www.jstage.jst.go.jp/article/ast/26/1/50/_pdf)- and we have discussed it on this thread. Also there is an addendum from the authors (http://www.hydrogenaudio.org/forums/index.php?showtopic=51322&view=findpost&p=651585).

The first link is broken, the second link's contents sheds no light on audibility

Works here.
Title: Audibility of Jitter
Post by: udauda on 2010-05-28 00:11:44
If you just stop being Asianophobic.. nah I'm just kidding. I totally understand your concern. Most of the listening test-related studies from asia do not really follow the proper DBT(BS.1116-1) procedure, thus end up being null.

[3] The paper (http://www.iic.tuis.ac.jp/edoc/journal/ron/r7-2-8/index.html) merely discusses that jitter components in high-end systems are usually less than 1ns. Again, no need to shed light on audiblity.
[4] The Link (http://www.jstage.jst.go.jp/article/ast/26/1/50/_pdf) works fine- an addendum from the authors (http://www.hydrogenaudio.org/forums/index.php?showtopic=51322&view=findpost&p=651585) works too. Another Link (http://www.hydrogenaudio.org/forums/index.php?showtopic=51322&st=0&p=463828&#entry463828) FYI.
[5] Again, not even mentioned in the presentation itself. Not sure if it meets the TOS 8. (wish I could obtain a copy)
[6] Certainly outdated. As you've said, does not meet TOS 8.

Except with [6], I still see the excerption as a good outline.
Title: Audibility of Jitter
Post by: Arnold B. Krueger on 2010-05-28 00:31:49
Quote
[4] yes (http://www.jstage.jst.go.jp/article/ast/26/1/50/_pdf)- and we have discussed it on this thread. Also there is an addendum from the authors (http://www.hydrogenaudio.org/forums/index.php?showtopic=51322&view=findpost&p=651585).

The first link is broken, the second link's contents sheds no light on audibility

Works here.


Works here, now. I haven't changed anything.
Title: Audibility of Jitter
Post by: Arnold B. Krueger on 2012-10-11 17:00:16
How many of the references that went into this page meet TOS 8?
2?


No need for sarcasm. 

[1] yes- and it is quite well-known.


I have the paper right here before me, and it says quite clearly that the thresholds of audibility that it plots out were "calculated"

Quote
[2] yes- done by Dolby.
[3] objective measurements only- no need to meet TOS 8.
[4] yes (http://www.jstage.jst.go.jp/article/ast/26/1/50/_pdf)- and we have discussed it on this thread. Also there is an addendum from the authors (http://www.hydrogenaudio.org/forums/index.php?showtopic=51322&view=findpost&p=651585).
[5] not referenced above- no need to meet TOS 8.
[6] no- written in 1970, well before David Clark's test in 1982.


That makes the count of TOS8-compliant papers exactly 2, no?

(no sarcasm intended) ;-)
Title: Audibility of Jitter
Post by: punkrockdude on 2012-10-11 21:01:57
I have not read everything but here is a test I did with my Focusrite ISA828 unit with ADC, RME Multiface II and a Black Lion Audio microclock mkII.

The FLAC files are here and just line them up in a sequencer application and/or just listen.

http://www.interfearingsounds.com/blandat/..._comparison.zip (http://www.interfearingsounds.com/blandat/rme_multiface_2_microclock_mk2_comparison.zip)
http://www.interfearingsounds.com/blandat/..._comparison.zip (http://www.interfearingsounds.com/blandat/focusrite_isa828_converters_microclock_mk2_comparison.zip)

For the lazy ones, here are the files phase flipped so that you can hear what's been nulled and not:

http://www.interfearingsounds.com/blandat/...se_flipped.flac (http://www.interfearingsounds.com/blandat/rme_multiface2_microclock_phase_flipped.flac)
http://www.interfearingsounds.com/blandat/...se_flipped.flac (http://www.interfearingsounds.com/blandat/isa828_adc_microclock_phase_flipped.flac)

This has all been taken from my old topic over at TapeOP messageboard. http://messageboard.tapeop.com/viewtopic.php?t=70170 (http://messageboard.tapeop.com/viewtopic.php?t=70170)
Title: Audibility of Jitter
Post by: punkrockdude on 2012-10-14 13:38:58
What's your opinion on the difference in audible difference in my last post? What makes the behaviour? Regards.
Title: Audibility of Jitter
Post by: Arnold B. Krueger on 2012-10-15 14:48:06
What's your opinion on the difference in audible difference in my last post? What makes the behaviour? Regards.


Any possible audible differences are effectively masked by a gratuitous warbled tone.
Title: Audibility of Jitter
Post by: punkrockdude on 2012-10-22 19:49:13
What's your opinion on the difference in audible difference in my last post? What makes the behaviour? Regards.


Any possible audible differences are effectively masked by a gratuitous warbled tone.

Are you serious?
Title: Audibility of Jitter
Post by: Arnold B. Krueger on 2012-10-22 20:34:04
What's your opinion on the difference in audible difference in my last post? What makes the behaviour? Regards.


Any possible audible differences are effectively masked by a gratuitous warbled tone.

Are you serious?


If  *you* are serious, you'll post a log of your ABX tests based on these files, perhaps using FOOBAR2000 as your comparison software. ;-)
Title: Audibility of Jitter
Post by: punkrockdude on 2012-10-22 21:40:05
What's your opinion on the difference in audible difference in my last post? What makes the behaviour? Regards.


Any possible audible differences are effectively masked by a gratuitous warbled tone.

Are you serious?


If  *you* are serious, you'll post a log of your ABX tests based on these files, perhaps using FOOBAR2000 as your comparison software. ;-)

Why I wrote the last sentence is because of "A gratuitous warbled tone" and "POSSIBLE audible difference" (capitalized two words), like if you didn't really care that much.
Title: Audibility of Jitter
Post by: 2Bdecided on 2012-10-23 11:14:40
What's your opinion on the difference in audible difference in my last post? What makes the behaviour? Regards.
Your 05 file is a fraction of a sample earlier than your 04 file - i.e. they are not perfectly in sync. Hence they won't null out. It's somewhat difficult to correct for sub-sample delays - you'd have to add a sub-sample delay to file 05 to make it null properly. However, adding that sub-sample delay will typically change the audio a little, so it's tricky to rely on null tests in this case. It could be done, with the caveat that you'd have to ignore any differences that were due to the sub-sample delay, and hope no actual differences were hidden by this. It would be hard to convince people of the validity of a positive or negative result - especially in comparison with the convincing -100dB result you posted.

But there is some good news - the sync is good enough to run an ABX test, especially on any ABX test tool that doesn't try to implement (or lets you disable) perfect seamless switching.

Without trying an ABX test of the two files and posting your results, you have not demonstrated an audible difference. Expectation bias is too powerful a thing for anyone on this board to believe anyone who says "I hear a difference" without a passed ABX test to prove the fact.

Cheers,
David.
Title: Audibility of Jitter
Post by: punkrockdude on 2012-10-23 12:26:10
2Bdecided: Interesting about subsamples. So there is actually space that is smaller than the audio file's actual time limit (sample rate)?
Title: Audibility of Jitter
Post by: dhromed on 2012-10-23 13:25:31
File 2 and 3 null out almost perfectly. Amplification of that difference by 50 and then another 20 dB* produces a beautiful example of a quiet muffled song, drowning in a thick ocean of noise. Can we reject any chance of audible difference here right off the bat?

Also, I'm not sure why there are 4 files on offer, but only 2 inverted files. I didn't use the inverted ones. Was I supposed to compare 2/3 and 4/5?

4 and 5 have a tiny subsample difference, so the diff is a complete piece of music, but without the bass. I don't have the qualifications to explain why the bass region vanishes.

Quote
like if you didn't really care that much.


The warbled tone extends rather far into the sample, and I suppose it was a basis for Arnold to dismiss the samples immediately. I couldn't ABX any of it, in any case.

Quote
Your 05 file is a fraction of a sample earlier than your 04 file


How did you know 5 is earlier than 4? Better tools?


*) Audacity apparently doesn't allow more than 50dB amplification in a single step.
Title: Audibility of Jitter
Post by: Arnold B. Krueger on 2012-10-23 13:32:10
2Bdecided: Interesting about subsamples. So there is actually space that is smaller than the audio file's actual time limit (sample rate)?


There always is. Samples are always separated by a finite amount of time, and anything that is finite can be made smaller.

Adjusting files in increments of less than a sample can be done. What you do is upsample the file to a sample rate by an amount that is as high as you need to make it, or at least is as high as you can practically make it, do the adjustment on the upsampled file, and then downsample the file back to the original sample rate.

This will scare the %$#!! out of some people, but if you have good resampling software, it can have useful results.
Title: Audibility of Jitter
Post by: Arnold B. Krueger on 2012-10-23 13:41:56
What's your opinion on the difference in audible difference in my last post? What makes the behaviour? Regards.
Your 05 file is a fraction of a sample earlier than your 04 file - i.e. they are not perfectly in sync. Hence they won't null out. It's somewhat difficult to correct for sub-sample delays - you'd have to add a sub-sample delay to file 05 to make it null properly. However, adding that sub-sample delay will typically change the audio a little, so it's tricky to rely on null tests in this case. It could be done, with the caveat that you'd have to ignore any differences that were due to the sub-sample delay, and hope no actual differences were hidden by this. It would be hard to convince people of the validity of a positive or negative result - especially in comparison with the convincing -100dB result you posted.

But there is some good news - the sync is good enough to run an ABX test, especially on any ABX test tool that doesn't try to implement (or lets you disable) perfect seamless switching.


Agreed. File synch is generally good enough if its within 1/100th of a second.

Quote
Without trying an ABX test of the two files and posting your results, you have not demonstrated an audible difference.


I don't think the OP did this consciously, but by not first doing his own test, he's effectively telling us that he doesn't have time to do things right, but figures that we have nothing better to do than to spend the necessary time doing his homework for him.

Quote
Expectation bias is too powerful a thing for anyone on this board to believe anyone who says "I hear a difference" without a passed ABX test to prove the fact.


I don't think he actually claimed an audible difference, thereby escaping the TOS 8 issue.

At any rate I fell for his ruse, tried to ABX the files and found the problem that I identified, which he now seems to want to pick a fight with me about. ;-)

He is either very ignorant or very arrogant. ;-) 

I'm about ready to do the good old plonk thing. Should I spend the time to see where his IP address leads?  ;-)

Title: Audibility of Jitter
Post by: pdq on 2012-10-23 16:16:51
4 and 5 have a tiny subsample difference, so the diff is a complete piece of music, but without the bass. I don't have the qualifications to explain why the bass region vanishes.

Taking the delta between samples a fixed time interval apart acts as a low-pass filter with a 6 dB roll off, until the period approaches the time delta. It then acts as a notch filter when the period and the time delta are equal.
Title: Audibility of Jitter
Post by: 2Bdecided on 2012-10-23 17:21:01
2Bdecided: Interesting about subsamples. So there is actually space that is smaller than the audio file's actual time limit (sample rate)?
The sample rate sets a frequency limit, not a time limit. The waveform at times "between" sample points is very well defined. However, frequency components at or above half the sample rate can't exist in a baseband sampled system.

Cheers,
David.
Title: Audibility of Jitter
Post by: 2Bdecided on 2012-10-23 17:22:09
At any rate I fell for his ruse, tried to ABX the files and found the problem that I identified, which he now seems to want to pick a fight with me about. ;-)

He is either very ignorant or very arrogant. ;-)
He might be confused. I couldn't follow your argument at all.
Title: Audibility of Jitter
Post by: 2Bdecided on 2012-10-23 17:30:32
4 and 5 have a tiny subsample difference, so the diff is a complete piece of music, but without the bass. I don't have the qualifications to explain why the bass region vanishes.

Taking the delta between samples a fixed time interval apart acts as a low-pass filter with a 6 dB roll off, until the period approaches the time delta. It then acts as a notch filter when the period and the time delta are equal.
Is this right? You're describing the comb filter you get when you add suitably delayed signals together. However, when you subtract delayed signals, I think the basic response is a high pass filter - and if the delay is increased to give a comb filter effect, the lowest frequencies are always attenuated - i.e. DC is always in a notch filter.

(don't take my word for this - far too sleep deprived to be 100% sure of anything today  )

Cheers,
David.
Title: Audibility of Jitter
Post by: Kees de Visser on 2012-10-23 18:21:30
To illustrate the importance of sub-sample accuracy in null-tests I made a new version with an 80 millisample delay manually switching in and out. It makes it easy to hear how much better the null is with the delay. I still can't get it as good as the RME though. It would be interesting to do some lab measurements with the device running on internal and external clock.
null-test flac (http://www.galaxyclassics.com/public/isa828_adc_microclock_phase_flipped_80msamplesdelay.flac)
null-test wav (http://www.galaxyclassics.com/public/isa828_adc_microclock_phase_flipped_80msamplesdelay.wav)
Title: Audibility of Jitter
Post by: pdq on 2012-10-23 18:53:39
4 and 5 have a tiny subsample difference, so the diff is a complete piece of music, but without the bass. I don't have the qualifications to explain why the bass region vanishes.

Taking the delta between samples a fixed time interval apart acts as a low-pass filter with a 6 dB roll off, until the period approaches the time delta. It then acts as a notch filter when the period and the time delta are equal.
Is this right? You're describing the comb filter you get when you add suitably delayed signals together. However, when you subtract delayed signals, I think the basic response is a high pass filter - and if the delay is increased to give a comb filter effect, the lowest frequencies are always attenuated - i.e. DC is always in a notch filter.

(don't take my word for this - far too sleep deprived to be 100% sure of anything today  )

Cheers,
David.

Oops...I meant high pass. 

BTW, if the period is twice the delay then the signal is amplified 6 dB.
Title: Audibility of Jitter
Post by: Arnold B. Krueger on 2012-10-23 20:12:28
At any rate I fell for his ruse, tried to ABX the files and found the problem that I identified, which he now seems to want to pick a fight with me about. ;-)

He is either very ignorant or very arrogant. ;-)
He might be confused. I couldn't follow your argument at all.


Even after listening to his samples?
Title: Audibility of Jitter
Post by: 2Bdecided on 2012-10-24 10:02:06
At any rate I fell for his ruse, tried to ABX the files and found the problem that I identified, which he now seems to want to pick a fight with me about. ;-)

He is either very ignorant or very arrogant. ;-)
He might be confused. I couldn't follow your argument at all.


Even after listening to his samples?
Yes. They consist of a bell, then a guitar, and a bloke singing. Fine for the purposes of investigating the audibility of jitter or any other DAC error in a signal consisting of a bell, then a guitar, and a bloke singing

Of course I agree about the necessity of ABXing, and the fact it's a possible audible problem.

Cheers,
David.
Title: Audibility of Jitter
Post by: krabapple on 2013-09-24 17:52:47
FWIW, here's a good faith attempt to supply examples of  audible jitter


http://www.cranesong.com/jitter_1.html (http://www.cranesong.com/jitter_1.html)
Title: Audibility of Jitter
Post by: Arnold B. Krueger on 2013-09-26 03:31:13
FWIW, here's a good faith attempt to supply examples of  audible jitter


http://www.cranesong.com/jitter_1.html (http://www.cranesong.com/jitter_1.html)



It appears to be at least partially in error. File B is inverted. You probably want to invert it to avoid an irrelevant potentially confounding variable.

I picked two files at random and subtracted them producing a residual that averages about 70 dB down. The residual, when raised by 40 dB may show signs of FM distortion being the difference between the two files as it sounds kinda warbly.
Title: Audibility of Jitter
Post by: Arnold B. Krueger on 2013-09-26 03:48:34
The threshold of audibility of phase noise in ADC and DAC clocks
...

I would suggest you begin with a working definition of "Jitter", which you could post to the HA Knowledgebase. Is jitter "phase noise" ?


Depends what you call noise. If your definition of noise includes only random or pseudo-random signals then no. Jitter error is FM or phase modulation. The modulating signal or modulating function can be and often is deterministic.

Quote
What does it sound like?


Depends on the nature of the modulating function.

If the modulating function is random or pseudo-random then all that happens is that the noise floor gets raised.  The nature of the rise may have a broad spectrum or it may have a narrow spectrum that is bunched up around the signal, in which case you have modulation noise.

If the modulating function is deterministic and reasonably simple (it often is - e.g. a sine wave or a square wave) then in small quantities it adds a rough character. If the modulating function has high enough amplitude but still has a simple spectral content, then it sounds like a vibrato. 

If the modulating function is complex then the effects are more like what you get with a random or pseudo random function.

Quote
What does it do to a sine wave that goes through ADCs and DACs ?


Simply stated it FM or phase modulates the signal.

Quote
Does it matter at all ?


Jitter matters if it is large enough to be audible.

One curious factoid is that traditional forms of analog recording and playback are often relatively speaking engorged with audible jitter, and the golden ears never seem to hear it.  Digital transcriptions of analog recordings perpetuate their jitter unless the jitter has been forcibly removed, which can be done.

Jitter in digital gear tends to be much lower and is generally inaudible.

Quote
What is the order of magnitude of jitter in PCs? and in high-end audio equipment?


It can be all over the map. PCs with high quality audio interfaces and even just good cheap ones are generally free of audible jitter.

But as usual it is impossible to make a perfectly general statement.
Title: Audibility of Jitter
Post by: stv014 on 2013-09-26 10:56:10
Not sure if this is relevant here, but there is currently an ongoing discussion - including listening and hardware tests on the last few pages - related to jitter and its audibility on another forum here (http://www.head-fi.org/t/668878/jitter-correlation-to-audibility).
Title: Audibility of Jitter
Post by: Arnold B. Krueger on 2013-09-26 12:35:42
Not sure if this is relevant here, but there is currently an ongoing discussion - including listening and hardware tests on the last few pages - related to jitter and its audibility on another forum here (http://www.head-fi.org/t/668878/jitter-correlation-to-audibility).



Quote from that article:

"For instance, Empirical Audio uses two oscillators that are both specified at 2psec RMS jitter. The two oscillators sound radically different to me when used in a re-clocker in a resolving audio system."

Sighted evaluations - IOW golden eared techno junk.

Next! ;-)
Title: Audibility of Jitter
Post by: krabapple on 2013-09-27 02:59:00
FWIW, here's a good faith attempt to supply examples of  audible jitter


http://www.cranesong.com/jitter_1.html (http://www.cranesong.com/jitter_1.html)



It appears to be at least partially in error. File B is inverted. You probably want to invert it to avoid an irrelevant potentially confounding variable.



Did you read the explanation of the files?

"All files are phase canceled with the B file, which is the reference."

http://www.cranesong.com/jitter_2results.html (http://www.cranesong.com/jitter_2results.html)

Title: Audibility of Jitter
Post by: stv014 on 2013-09-27 10:53:33
Sighted evaluations - IOW golden eared techno junk.


That may be the case for the opening post, but not necessarily for many of the replies. For the listening tests, skip to page 8. UltimateMusicSnob again has some surprising ABX results, although for now only with rather high amounts of jitter (however, the actual levels of the jitter products are fairly low).

Quote from: krabapple link=msg=0 date=
Did you read the explanation of the files?

"All files are phase canceled with the B file, which is the reference."


There might be a real problem, however, the difference files have far higher level (I mean by orders of magnitude) than expected for the amount of jitter, even if the gain applied to the files is taken into account.
Title: Audibility of Jitter
Post by: 2Bdecided on 2013-09-27 13:04:09
I think UltimateMusicSnob is becoming the new Guruboolez, but probably to a far higher degree.
Title: Audibility of Jitter
Post by: Wombat on 2013-09-27 14:38:24
I think UltimateMusicSnob is becoming the new Guruboolez, but probably to a far higher degree.

This, or a very sophisticated troll payed by High-End business to reinforce audiophile 'facts' in popular forums 
Title: Audibility of Jitter
Post by: 2Bdecided on 2013-09-27 17:35:59
Don't joke, a similar thought had crossed my mind - though the jitter being ABXed is still a lot greater than anything normal equipment would exhibit.

I haven't followed it, but there's a report that he's ABXed pre-ringing at 22kHz or something?

FWIW I've tried and failed to ABX the files over there. It's strange - when listening critically I can hear plenty of flaws in the source content, but not the "large" amounts of jitter. I'd have to listen far less critically to enjoy such flawed sources (or for that matter any music), so it's interesting what other people hear and care about.

Even the artefacts that I can ABX (mp3 encoders etc) are often far smaller and less annoying than the faults in a typical good recording. To me, at any rate.

Cheers,
David.
Title: Audibility of Jitter
Post by: Wombat on 2013-09-27 19:46:50
i already made up my mind when he described things he listens for around the Ravel sample. Since i lately often appeared as bugger in here i decided to shut up about it.
Title: Audibility of Jitter
Post by: greynol on 2013-09-27 19:56:51
It's difficult to describe the tonality of noise, but this really isn't the place for a discussion of the Ravel sample.
Title: Audibility of Jitter
Post by: Arnold B. Krueger on 2013-09-28 14:32:30
FWIW, here's a good faith attempt to supply examples of  audible jitter


http://www.cranesong.com/jitter_1.html (http://www.cranesong.com/jitter_1.html)



It appears to be at least partially in error. File B is inverted. You probably want to invert it to avoid an irrelevant potentially confounding variable.



Did you read the explanation of the files?


I read this which justifies my comment:

http://www.cranesong.com/jitter_1.html (http://www.cranesong.com/jitter_1.html)

"At exactly the same playback level, listen to each sample all the way through - several times. Become familiar with the recording.  The files are the same level"

What is unclear about "listen to each sample"?

Quote
"All files are phase canceled with the B file, which is the reference."


Since polarity can be audible, this conditions the proposed test for 100% positives in any comparison between the unknowns and the known, for a reason that has nothing to do with jitter.

We are talking about the validity of following the instructions to the listener, and nobody not even us should encumber the listening test with details about how the samples were made.

The proposed test is still flawed by the introduction of an irrelevant but potentially confounding variable.