1

Topic: So can you hear it? The jitter test.

The jitter test is up on the website.

Please download the sample clips and tell us which sound best or worst!

ITUNES

http://hddaudio.net///wordpress/wp-cont … mple-a.m4a

http://hddaudio.net///wordpress/wp-cont … mple-b.m4a

http://hddaudio.net///wordpress/wp-cont … mple-c.m4a

http://hddaudio.net///wordpress/wp-cont … mple-d.m4a

FLACS

http://hddaudio.net///wordpress/wp-cont … ple-a.flac

http://hddaudio.net///wordpress/wp-cont … ple-b.flac

http://hddaudio.net///wordpress/wp-cont … ple-c.flac

http://hddaudio.net///wordpress/wp-cont … ple-d.flac

2

Re: So can you hear it? The jitter test.

sample-a.m4a  vs  sample-b.m4a

n    A    X    B    Choice    Score
1    2    1    1      A         0/1
2    2    2    1      A         1/2
3    2    1    1      B         2/3
4    2    1    1      B         3/4
5    2    2    1      A         4/5
6    2    1    1      B         5/6
7    2    1    1      B         6/7
8    2    2    1      A         7/8
9    2    2    1      B         7/9
10    2    2    1      B         7/10
11    2    2    1      B         7/11
12    2    2    1      B         7/12
13    2    1    1      B         8/13
14    2    1    1      A         8/14
15    2    1    1      B         9/15

Percent correct: 60%


( Note: I could not hear meaningful
differences between the two files,
all choices were guesses. )

.............

sample-a.m4a  vs  sample-c.m4a

n    A    X    B    Choice    Score
1    2    2    1      A         1/1
2    2    1    1      B         2/2
3    2    1    1      A         2/3
4    2    2    1      B         2/4
5    2    1    1      B         3/5
6    2    1    1      A         3/6
7    2    1    1      A         3/7
8    2    1    1      A         3/8
9    2    2    1      B         3/9
10    2    2    1      B         3/10
11    2    2    1      A         4/11
12    2    2    1      B         4/12
13    2    2    1      B         4/13
14    2    1    1      A         4/14
15    2    1    1      A         4/15

Percent correct: 26%


( Note: I could not hear meaningful
differences between the two files,
all choices were guesses. )

.............

sample-a.m4a  vs  sample-d.m4a

n    A    X    B    Choice    Score
1    1    2    2      A         0/1
2    1    2    2      A         0/2
3    1    2    2      A         0/3
4    1    1    2      B         0/4
5    1    1    2      B         0/5
6    1    1    2      B         0/6
7    1    1    2      A         1/7
8    1    1    2      B         1/8
9    1    1    2      A         2/9
10    1    1    2      B         2/10
11    1    1    2      B         2/11
12    1    2    2      A         2/12
13    1    2    2      A         2/13
14    1    1    2      A         3/14
15    1    2    2      B         4/15

Percent correct: 26%

( Note: I could not hear meaningful
differences between the two files,
all choices were guesses. )

3

Re: So can you hear it? The jitter test.

I would not have selected this particular track (Nora Jones) as it has a lot of jitter already in the recording, but anyway I did the listening tests at both 16/44.1 and upsampled with SRC in Foobar 0.8.3 to 24/96. The best and worst are more obvious than the ones in-between. Like I said, there is so much jitter in this already, its difficult to pick-out additional jitter. The imaging on all of these tracks sucks IMO.

Here is my determination:

Highest jitter - Track A
Next - Track B
Next - Track C
Lowest Jitter - Track D

Between B and C it was hard to make a call. This would have been much easier with correlated jitter like you get in real systems rather than this contrived random jitter.

Next time you need a good track for this, ask me.

Steve N.
Empirical Audio

Last edited by audioengr (2009-04-05 02:21:34)

4

Re: So can you hear it? The jitter test.

Okay, what would your recommended track be?

5

Re: So can you hear it? The jitter test.

Steve

I was reading about you on Hydrogen Audio, where you seem to have left a strong impression, and I fancy it will be much the same on here, because some of the members are Scientists and Engineers and not terribly tolerant of Patent Medicine Salesmen.

Ash

Manufacturer: www.avihifi.com / Contact: Email AVI / Blog: AVI HiFI

6

Re: So can you hear it? The jitter test.

audioengr wrote:

I would not have selected this particular track (Nora Jones) as it has a lot of jitter already in the recording, but anyway I did the listening tests at both 16/44.1 and upsampled with SRC in Foobar 0.8.3 to 24/96. The best and worst are more obvious than the ones in-between. Like I said, there is so much jitter in this already, its difficult to pick-out additional jitter. The imaging on all of these tracks sucks IMO.

Here is my determination:

Highest jitter - Track A
Next - Track B
Next - Track C
Lowest Jitter - Track D

Between B and C it was hard to make a call. This would have been much easier with correlated jitter like you get in real systems rather than this contrived random jitter.

Next time you need a good track for this, ask me.

Steve N.
Empirical Audio

For further ramblings of Mr. Nugent's, I'd suggest viewers point their browsers in the direction of this external forum thread. It says it all really. roll

--
djp

Intel iMac + Beresford TC-7510 + Little Dot MK III + beyerdynamics DT 231 = Computer audiophile quality on the cheap! --- Samsung Q1 + M-Audio Transit + Sennheiser PX 100 = Computer audiophile quality on the go!

7

Re: So can you hear it? The jitter test.

Thanks for trying the tracks, Steve.

I'll email you the results.

8

Re: So can you hear it? The jitter test.

My initial reaction was to give Mr. Nugent a lot of credit for actually taking the test. Then I brushed my teeth. Dressed. Went to church. I'm not sure what, in that continuum, reminded me of who Steve is...a guy who sells extension cords by telling people that after their power has traveled many miles through the aging electrical grid that somehow, without filtering (which he doesn't believe in), it will be completely transformed, as will their systems' transparency, imaging and musicality, by the last six feet of wire, provided by him for a mere $200 a foot.

He will, no doubt, cheat. Simple enough as I'm sure he has the instruments to measure the files.

Tim

GrumpyOldArts.wordpress.com

9

Re: So can you hear it? The jitter test.

At the outset of this test, I noted that the majority of CD players and DACs have jitter in the range 0.1 to 2ns, and a few are even better.  I therefore decided to apply the jitter from 4 to 100ns, so that the replay equipment wouldn't significantly skew the results.  The 100ns file was intended as an example of really horrendous jitter, so that you could easily recognise what it sounded like.

Hopefully Darren can confirm this, but I believe that only 1 or 2 people have correctly identified the 100ns file, and no one has reliably identified less than 40ns.  This was unexpected.  I don't really believe that the original recording had a lot of jitter to begin with, and such claims have little credibility in my opinion.

However, this test is a sample of just one track, and it may well be that other recordings or types of music may give a different result.  But I would be confident that the result would be quite similar, and certainly not differ by an order of magnitude.

Regarding the type of jitter, the characteristics of clock oscillator phase noise (jitter) are easily predicted mathematically, both for the oscillator itself and the in terms of the effect of the PLL (if there is one).  The jitter spectrum used in these tests followed this conventional wisdom.  It is true that sometimes there may also be periodic components to the jitter (sometimes incorrectly called correlated jitter), and not just noise-like jitter, but these will usually be small compared to the total jitter energy.

If anyone wishes to provide alternative music files, get in touch with Darren, and I will be happy to repeat the test.

Cheers,
Chris.

Last edited by cs (2009-04-05 18:20:33)

10

Re: So can you hear it? The jitter test.

cs wrote:

At the outset of this test, I noted that the majority of CD players and DACs have jitter in the range 0.1 to 2ns, and a few are even better.  I therefore decided to apply the jitter from 4 to 100ns, so that the replay equipment wouldn't significantly skew the results.  The 100ns file was intended as an example of really horrendous jitter, so that you could easily recognise what it sounded like.

Hopefully Darren can confirm this, but I believe that only 1 or 2 people have correctly identified the 100ns file, and no one has reliably identified less than 40ns.  This was unexpected.  I don't really believe that the original recording had a lot of jitter to begin with, and such claims have little credibility in my opinion.

However, this test is a sample of just one track, and it may well be that other recordings or types of music may give a different result.  But I would be confident that the result would be quite similar, and certainly not differ by an order of magnitude.

Regarding the type of jitter, the characteristics of clock oscillator phase noise (jitter) are easily predicted mathematically, both for the oscillator itself and the in terms of the effect of the PLL (if there is one).  The jitter spectrum used in these tests followed this conventional wisdom.  It is true that sometimes there may also be periodic components to the jitter (sometimes incorrectly called uncorrelated jitter), and not just noiselike jitter, but these will usually be small compared to the total jitter energy.

If anyone wishes to provide alternative music files, get in touch with Darren, and I will be happy to repeat the test.

Cheers,
Chris.

Chris, it's completely possible that, in this case, psychological bias worked in the other direction: We didn't expect to hear a difference, and so we didn't. I know I didn't hear one sample that obviously had distortion, but I also admittedly didn't listen for long or pay that much attention. Listening for jitter sounds too much like tasting for bitter to me, and after a couple of runs of the first few bars of that lovely, jitter-infested track, I found myself compelled to stop listening for what was wrong and listen to the music. I loaded up the album and sat back.

Tim

GrumpyOldArts.wordpress.com

11

Re: So can you hear it? The jitter test.

tfarney wrote:

Chris, it's completely possible that, in this case, psychological bias worked in the other direction: We didn't expect to hear a difference, and so we didn't. I know I didn't hear one sample that obviously had distortion, but I also admittedly didn't listen for long or pay that much attention. Listening for jitter sounds too much like tasting for bitter to me, and after a couple of runs of the first few bars of that lovely, jitter-infested track, I found myself compelled to stop listening for what was wrong and listen to the music. I loaded up the album and sat back.

Tim

Tim,

Yes I agree that is perfectly possible.
However, the very fact that you were able to so easily forget the test and just listen to the music confirms that jitter is a red herring.  If it was as bad as some of the subjectivists or pseudo-engineers claim, you would have found the sound horrible to listen to.

Chris.

12

Re: So can you hear it? The jitter test.

cs wrote:
tfarney wrote:

Chris, it's completely possible that, in this case, psychological bias worked in the other direction: We didn't expect to hear a difference, and so we didn't. I know I didn't hear one sample that obviously had distortion, but I also admittedly didn't listen for long or pay that much attention. Listening for jitter sounds too much like tasting for bitter to me, and after a couple of runs of the first few bars of that lovely, jitter-infested track, I found myself compelled to stop listening for what was wrong and listen to the music. I loaded up the album and sat back.

Tim

Tim,

Yes I agree that is perfectly possible.
However, the very fact that you were able to so easily forget the test and just listen to the music confirms that jitter is a red herring.  If it was as bad as some of the subjectivists or pseudo-engineers claim, you would have found the sound horrible to listen to.

Chris.

Well, maybe I can ignore it easier than most. Or maybe not. Jitter may very well be audible at some level if your equipment is resolving enough, your recording is perfect enough and you're listening for it intently enough. But true or not, it's hardly worth arguing as it only proves that some audiophiles are more interested in the tweaking than the listening, and their vendors are more than happy to take advantage of them.

Tim

GrumpyOldArts.wordpress.com

13

Re: So can you hear it? The jitter test.

Steve has kindly suggested some tracks, which I will get over to CS for processing.

Look out for test number two soon...

14

Re: So can you hear it? The jitter test.

Excellent. But the story is already emerging: If you pick just the right tracks (as opposed to one like "Come Away With Me" that is widely considered to be an excellent recording) and listen critically, not for the beauty of the music, but for what might be wrong with it, on a system more resolving than those found in the homes of most audiophile journalists, you may be able to hear amounts of jitter that are highly unlikely to be found once the signal gets past the filters of a decent DAC.

A vital concern to the music lover, to be sure. If we hear the jitter in Steve's tracks, he will be exonerated, elevated from his current status: He is merely a master of overstatement in the name of commerce. Well, until we get to that mains cable thing.

Or to put it more succinctly, I don't think any of us ever claimed that jitter does not exist. We simply said it was not an issue in competent equipment. And that point is confirmed here already.

Tim

Last edited by tfarney (2009-04-06 12:09:08)

GrumpyOldArts.wordpress.com

15

Re: So can you hear it? The jitter test.

Using Shure 530s and my MacBook Pro I can't tell the difference between them.  I guess if I had to guess, I would say sometimes I thought sample b sounded slightly better...

16

Re: So can you hear it? The jitter test.

You should have one file that is so bad that the jitter is very obvious, that way we can get a taste of what we should be listening for in the other samples.  Consider it a training sample...

17

Re: So can you hear it? The jitter test.

How do I find out which sample was which for Come Away With Me?