world of jitter

[kajak12]

http://www.goodsoundclub.com/Forums/ShowPost.aspx?PostID=2171



I do not know. People say a lot of words about jitter and speculate a lot about the topic. I never heard from anyone any real definition how a high level of jitter sounds in realty. I heard 4 (!!!!) complete contradictory descriptions how high level of jitter manifest itself subjectively – interesting that they all derived from the engineers whose credibility is very high (not juts the audio-tweakers who fart sound out of their audio components by changing capacitors). No one says that jitter is good but there is probably a lot of more then juts perusing a lower jitter. At least it is what I feel as when I experienced objectively-less jitter digital then it was not necessary that this digital performer better. For instance the Museatex original transport coupled with Bidat using Meitner’s C-lock synchronization mechanism that has NO MEASURABLE JITTER at all. However, this combo does sound very-very unfortunately. The top of the line Wadia, another instance, have very low jitter but those Wadia perform awfully. Certainly there are a lot of other things in digital that describe Sound but unfortunately the digital people talk only about jitter. Try to talk with them about something else, not to mention Music.  Another problem with the people who “do digital” are mostly just unable to objectively asses subjectively what the heal they do. So they build the sonic microwaves and push to the audio pubic a new marketing BS telling then of how they fought the jitter for instance…. The explanation from Lavry Forum that you linked does sound very reasonable. What is unreasonable to me that with objectively better clock in my TL0 I got that that horrible sound…

[Drizt]

Some of that is hard to read as the English used is a bit broken.

In the end the only thing that matters is what we end up hearing.  If one wants to objectively try to correlate what they hear with changes they have made then blind testing is the ultimate tool to do this.  Subjectively assessing these things is too hit and miss.  Humans are terribly unreliable and prone to being influenced by external forces.  If you remove all of the external influences and just use your ears (with proper level matching) then you are left with blind testing.

The amount of jitter is irrelevant if you can not reliably hear the differences it causes.  Blind testing can help you assess things like this. 

[kajak12]

That's the kind of response I would expect from someone like you

take your blind testing and stick it up your arse
your posting on the wrong forum

you push my buttons and you get what you deserve no more nice guy start collecting postage stamps leave audio alone
i have no time for people like you its a waste of data on the www
we learn nothing from you at all (we do laugh)

[zenelectro]

Here's an interesting take on jitter which seems to correlate more closely with what you guys (mario / steven) hear when comparing various clocks.

http://www.youtube.com/watch?v=-65gN44G9hU

Ivor talks about 'focused jitter' which I am presuming is adding small amounts of jitter that is random in nature, (like noise) and focused WRT bandwidth.

Terry

 

[Drizt]

Here a PDF that might be of interest.

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

Table 1 Size (r.m.s.) of random jitter added to materials
and the number of listeners who could discriminate
between sounds with and without jitter.
Size of random jitter Number of listeners who
(r.m.s.) discriminated sounds
2 ms 23
1 ms 11
500 ns 6
250 ns none

[zenelectro]

Here a PDF that might be of interest.

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

Table 1 Size (r.m.s.) of random jitter added to materials
and the number of listeners who could discriminate
between sounds with and without jitter.
Size of random jitter Number of listeners who
(r.m.s.) discriminated sounds
2 ms 23
1 ms 11
500 ns 6
250 ns none

Drizt,

What is your take on this paper?

IMO it is poorly done and has not a great deal of information to offer.

More later - busy.

Terry

[Drizt]

My take is that its a step in the right direction on how to test small things like jitter properly.  The fact that they digitally introduced the jitter was interesting.  It meant that they had a base line for the 'minimum' amount of jitter that made it all the way through the chain.  The regular amounts of jitter in that system would also be added along the way.  It could be done better, sure.  But the blind testing element is essential.  Sighted testing of small things like jitter is pretty much useless in my opinion.  People heavily underestimate how influenced people are by external influences under sighted tests.  People like to say they trust their ears, but those same people are afraid of blind tests which exclusively use your ears.  Has me bewildered.  But each to their own.  Hopefully nobody take offence to my expressed opinions.

Edit: Here's a video that helps to demonstrate how easily people are influenced by external influences
<a href="http://www.youtube.com/watch?v=BYTlN6wjcvQ" target="_blank">http://www.youtube.com/watch?v=BYTlN6wjcvQ</a>

[stevenvalve]

My take is that its a step in the right direction on how to test small things like jitter properly.  The fact that they digitally introduced the jitter was interesting.  It meant that they had a base line for the 'minimum' amount of jitter that made it all the way through the chain.  The regular amounts of jitter in that system would also be added along the way.  It could be done better, sure.  But the blind testing element is essential.  Sighted testing of small things like jitter is pretty much useless in my opinion.  People heavily underestimate how influenced people are by external influences under sighted tests.  People like to say they trust their ears, but those same people are afraid of blind tests which exclusively use your ears.  Has me bewildered.  But each to their own.  Hopefully nobody take offence to my expressed opinions.

Edit: Here's a video that helps to demonstrate how easily people are influenced by external influences
<a href="http://www.youtube.com/watch?v=BYTlN6wjcvQ" target="_blank">http://www.youtube.com/watch?v=BYTlN6wjcvQ</a>

I agree with some things, but one of the things that worries me in this video, is when he talks about power cords, and said how can power cables have any effect on the sound when there is miles of cable in the wall and beyond. I know they do change the sound, he said they don't,  he is wrong about that.

[Drizt]

I know they do change the sound, he said they don't,  he is wrong about that.

How do you know?  Ordinarily people would get upset when one asks if you have done any blind testing but in this case you have said he is wrong and you are right.  What testing have you done? 

[Drizt]

I see lots of people with ears mate, usually they come in pairs.  Having ears doesn't make you gifted

steve is gifted with ears

P.S.  Steve can answer for himself.

[kajak12]

I see lots of people with ears mate, usually they come in pairs.  Having ears doesn't make you gifted

steve is gifted with ears

P.S.  Steve can answer for himself.

what a stupid question to ask steve you know he uses his ears and not blind testing (only drizt on this site uses blind testing)

[Drizt]

what a stupid question to ask steve you know he uses his ears and not blind testing (only drizt on this site uses blind testing)

Steve can't answer for himself?

What is stupid is saying you use your ears as an argument against blind testing.

Blind testing is using your ears and only your ears.  Sighted testing is flawed from the start as you can see or know what is under test.

Steve has said he is right and the other guy is wrong.  That's fine, but I'd like to hear how he came to that conclusion and what proof he has.   If he doesn't have proof it just becomes a he said she said type of thing.  In that case it would have been better if Steve said ..... I have found otherwise.... or something to that effect. 

[zenelectro]

My take is that its a step in the right direction on how to test small things like jitter properly.  The fact that they digitally introduced the jitter was interesting.  It meant that they had a base line for the 'minimum' amount of jitter that made it all the way through the chain.  The regular amounts of jitter in that system would also be added along the way.  It could be done better, sure.  

I think the paper was a set up. They stacked the cards right from the outset in favour of a no score outcome and preyed on a technically
limited audience of which you appear to be.

It's unfortunate and contrary to the work of people who have spent a lot of energy and time researching this quite deep field.

The following articles offer basic grounding:

http://www.tnt-audio.com/clinica/jitter1_e.html
http://www.tnt-audio.com/clinica/jitter2_e.html

As you can see from the second link adding purely random jitter has the effect of raising the noise floor and it is the most innocuous form. 

Enough said.

T

[Drizt]

A bit of a dig?  Understanding the technical aspects of jitter and finding the threshold of audibility are slightly different things.   I'm more interested in reading the results of blind tests that were conducted to find the threshold of audibility.  Do you have any links to blind tests that were conducted for this purpose? 

I have a basic understanding of jitter, and understand what you have said.  I don't need to read much more on the topic (theory), I'm more interested in real world results.

One can read and read about theory, but in the end its the real world applications that matter.  Do we disagree on anything in particular hear?

I think the paper was a set up. They stacked the cards right from the outset in favour of a no score outcome and preyed on a technically
limited audience of which you appear to be.

It's unfortunate and contrary to the work of people who have spent a lot of energy and time researching this quite deep field.

The following articles offer basic grounding:

http://www.tnt-audio.com/clinica/jitter1_e.html
http://www.tnt-audio.com/clinica/jitter2_e.html

As you can see from the second link adding purely random jitter has the effect of raising the noise floor and it is the most innocuous form. 

Enough said.

T

[treblid]

There is an related interesting blog post on this, music-brain... May be interesting to some, or may not be. :p

Some day must really try to understand what jitter really is.. Until then, I'd just lurk... :p

[Drizt]

There is an related interesting blog post on this, music-brain... May be interesting to some, or may not be. :p

Some day must really try to understand what jitter really is.. Until then, I'd just lurk... :p

Have a read of the links that z posted above.  They are pretty easy to understand and follow, even if you have to re-read it a couple of times.  Its pretty easy to get a grasp of the basics.  I find it more interesting to find at what threshold (under what conditions) that the effects of jitter become audible.  After all, if its not really audible its not really worth worrying about.  

Edit: Your link doesn't seem to work ?

[treblid]

There is an related interesting blog post on this, music-brain... May be interesting to some, or may not be. :p

Some day must really try to understand what jitter really is.. Until then, I'd just lurk... :p

Have a read of the links that z posted above.  They are pretty easy to understand and follow, even if you have to re-read it a couple of times.  Its pretty easy to get a grasp of the basics.  I find it more interesting to find at what threshold (under what conditions) that the effects of jitter become audible.  After all, if its not really audible its not really worth worrying about. 

I know the basics.. But the problem is I only know the basics... As a developer working in an oceanography company measuring anything we can on the ocean and translating that into meaningful data. But our sensors only take 1-2 samples a second typically (save storage space and power and save on sat charges) so the little I know about jitter can't even be applied in audio (44100 samples a second)... with higher sampling frequency, one generally gets more data points, but the downside is more more noise as well, it's that relationship (resampling, oversampling, buffering to counter this and that) I don't understand. And how these techniques apply in the field of audio...

And on an intereting sidenode, say with ocean sensors, a squid lay eggs on 'em and all the collected data just go down the drain. In a sense that's true with audio as well as squid eggs can be thought of as bad CDs.. :lol:
 

Edit: Your link doesn't seem to work ?

It should work now. Took away the "..

[stevenvalve]

I know they do change the sound, he said they don't,  he is wrong about that.

How do you know?  Ordinarily people would get upset when one asks if you have done any blind testing but in this case you have said he is wrong and you are right.  What testing have you done?  

When i listen to my system, and after a while close my eyes, it does not sound any different. I usually do no blind testing. I trust my ears and interaction with my brain. I can hear what is happening to the sound, eyes open or closed. It sounds  the same either way. If my observations are wrong because i leave my eyes open, well why have i gone down the right path. Are you saying it is all wrong, because i did not do it blind. Do power cables sound different.  Yes, if you can not hear that. you need to ask yourself why.

[Drizt]

When i listen to my system, and after a while close my eyes, it does not sound any different. I usually do not blind testing. I trust my ears and interaction with my brain. I can hear what is happening to the sound, eyes open or closed. It sounds  the same either way. If my observations are wrong because i leave my eyes open, well why have i gone down the right path. Are you saying it is all wrong, because i did not do it blind. Do power cables sound different.  Yes, if you can not hear that. you need to ask yourself why.

I'm not saying it 'is' wrong.  I'm saying it 'could' be wrong (in the sense of actual differences vs. perceived differences)

Ultimately what I believe is that if the difference really does exist then you will be able to pass a blind test.  If you can only hear it during sighted testing then that's when I think you need to ask yourself some hard hitting questions.

In the end I think a lot of audiophiles are like fox mulder, 'I want to believe'.  A lot of audiophiles believe in absolutely everything, including the tooth fairy. 

[kajak12]

terry have you got anyway of measuring jitter on steves wadia?