Head-Fi.org › Forums › Equipment Forums › Dedicated Source Components › An interesting paper on jitter audibility
New Posts  All Forums:Forum Nav:

An interesting paper on jitter audibility - Page 8

post #106 of 112
12/28/06 at 12:32pm
Quote:
Originally Posted by audioengr View Post
Sorry, I just cannot read and comprehend your words with that monkey picture staring at me. I just cannot stop giggling.....
That picture is of my nephew.
Reply
post #107 of 112
12/28/06 at 2:33pm
Quote:
Originally Posted by drarthurwells View Post
That picture is of my nephew.
And it started off as jitter causing an incredibly small error in the genes somewhere....
Reply
post #108 of 112
12/28/06 at 2:42pm
Quote:
Originally Posted by drarthurwells View Post
Originally Posted by drarthurwells:
One other factor is amplification of error effects in a source. A very small error (in reading a CD) goes through many steps of processing and amplification before it even reaches a speaker or headphone. A small reading error can then emerge as much larger an error before reaching a speaker or headphone. Related to large scale chaos eventually emerging from an insignificant small beginning.




Art: It's like making a copy from a copy, then a third copy from the second copy, then a fourth copy from the third copy, and so on.

The last copy is very different from the original - adulterated in many ways. The very small effects of jitter are magnified and distorted along the signal path, which is analogous to making copies along the way.

The basic premise of Chaos theory is that initial minor changes in a system can lead to cataclysmic or large changes down the road. I think this is analogous to the jitter situation where very small and insignificant phase errors can result in audible effects.

As far as experimental testing of such effects using brief AB test comparisons - unreliable. Give me a week with A and a week with B, alternating over six weeks, and then I could detect a small jitter difference and could reliably tell you if A or B were playing - could not do this in a few 5 minute trials of AB.

So, skeptics will ask why if it takes such experience to hear a problem that is not more readily discernible, then why is it a problem?

The answer is that we listen over extended periods and learn to hear the good things and the bad things over time, and then get dissatisfied with problems we learn to detect with much experience.

A CD player with high jitter will sound good at first, but over months will become unsatisfactory. Most AB testing is "at first" - like meet sampling - and should be over months to really be discriminating.
I don't think the chaos theory is being done justice here, but I totally agree with the point you are making.
Often it takes time and experience (a learning curve) to be able to detect small anomalies and differences in sound.
Reply
post #109 of 112
12/28/06 at 3:20pm
Quote:
Originally Posted by drarthurwells View Post
The basic premise of Chaos theory is that initial minor changes in a system can lead to cataclysmic or large changes down the road. I think this is analogous to the jitter situation where very small and insignificant phase errors can result in audible effects.
Sure, you can state a theory that it could cause audible differences. But in the context of chaos theory applied to jitter in digital audio signals, you're still failing miserably at demonstrating that it does cause an audible difference.

(Argh, gotta get to work or I'd type more.)
Reply
post #110 of 112
12/30/06 at 4:45pm
Originally Posted by drarthurwells: That avatar picture is of my nephew.

Quote:
Originally Posted by Kees View Post
And it started off as jitter causing an incredibly small error in the genes somewhere....
Good point.

Millions of years ago some very tiny event, seemingly insignificant, linked carbon atoms with non-carbon atoms to produce an organic molecule.

Now we have life in millions of diverse forms, past and present, and more to come.
Reply
post #111 of 112
12/30/06 at 4:48pm
Quote:
Originally Posted by Kees View Post
And it started off as jitter causing an incredibly small error in the genes somewhere....
Or, a jitter causing a disturbance in the jeans.
Reply
post #112 of 112
12/30/06 at 5:04pm
Thread Starter 

Norah Jones - Jittered and Jitter-Free

First off many thanks for Steve N. for sending me the Jitter-free copy of Norah Jones.

First off I made my own copy of the CD and did a quick comparison between the three CDs and there were some interesting results.

1) Steve's copy is 45:35 the original and my copy are both 45:09 , Steve's copy has 2 seconds padding on all but the last track, Nero which I used makes a bit-copy.

2) I tested all CDs using

EAC test DA extract
Nero CD Speed
Nero DAE test utility

Both copies (mine and Steve's) read noticeably slower than the Commercial CD sometimes my copy was marginally faster than Steve's , sometimes it was slower but there was precious little in it

EAC extracted tracks from all CDs with no reported errors and the peak levels on all versions were the same for all tracks. Steve's copy showed 99.8% on 4 tracks and 100% on the rest, the original and my copy both showed 99.8% on 3 tracks (the same tracks) but in my experience this is EAC being paranoid and 99.8% does not mean audible errors. There is nothing here to suggest that the CDs are really any different.

Nero DAE shows all CDS will extract with zero errors.

So as far as the Computer is concerned the CDs are functionally identical nothwithstanding some minor differences.

I have been doing listening tests on the original and Steve's copy.

I have done sighted two transport tests with switching but that does introduce another variable and although I could not discern any difference it would not be fair to use that example as the differences between the transports could mask a difference. In any case synching is a problem as the copy has 2 second padding and so it has to be resynched after each track

I have also set up a random play program with the same track (Track 10 - Painter Song) on the two CDS playing in a changer without me knowing which is which. This does introduce a slight time lag between tracks about 10 seconds. I had an indepdendent person (the missus) load the CDs without telling me which was in which slot and the player does not do CD text so there are no clues as to which is which. There is no difference in mechanical noise, actually in play the player is dead silent. I have kept the volume the same throughout the exercise.

I have been listening to the two versions continuously and between toast breaks for a couple of hours. Sometimes the whole tracks sometimes sections of the tracks.

I cannot discern any difference between the commercial CD and the jitter-reduced copy, in my system with my ears and so forth, hardly conclusive I know but that is my data point.

It was fun trying though and I hope I did approach the challenge with an open mind and and open ears.

One unfortunate side effect is that I quite like Norah Jones now - it is like I have been bludgeoned into submission.

Thanks for the Classical sampler - will give it a play now.

Edit: Tannhauser overture is up first - thanks - one of my favourite pieces - a somewhat languid offering - who is doing it ?

Coincidentally I was just thinking I must put some Wagner on to be an antidote to Norah Jones...
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Dedicated Source Components
Head-Fi.org › Forums › Equipment Forums › Dedicated Source Components › An interesting paper on jitter audibility