It’s All a Preference

[Phelonious Ponk]

My point is any distortion becomes part of the signal as it is now part of the voltage swings. It therefor becomes part of the whole and affects the whole. You guys rarely see me post an absolute but in this case I'm doing just that. The question is how much they affect the whole not if they do or don't. They do.

Correct. The question is not "does the noise exist?" If we can see it on a scope, it exists and it is a part of the signal. I don't think anyone is arguing with that. The question is "is it audible?" I'm waiting for someone to explain to me how, and show me some evidence of inaudible noise having an audible impact on that signal. I'm guessing I'll keep waiting.

Tim

 

[Phelonious Ponk]

Well, that's the point; it's not really inaudible. Its volume is low enough that music will usually mask it, but it's not inaudible if the system volume is loud enough.

OK. Play your system louder than you will ever play it. Now, without touching the amp, turn off the source. put your ear to the speaker. If you can't hear "the noise," and as a backup, no noise measures in the human audible range, the noise is inaudible. Is that simple enough? We're not discussing theoretical system abuse, we're talking about noise affecting music reproduction.

Tim

 

[jkeny]

If, with no music playing, I have to turn my system up way past any normal playing level to hear this noise, again, with nothing to mask it, it is practially inaudible. And by practically, just to be clear, I mean it can't be heard by good human ears at any playback level that good human ears could tolerate. Inaudible in its intended use. So the question remains, how does this inaudible noise affect the quality of my playback and what is its effect?

Tim

Tim, a couple of points:
- you fail to take account of the dynamic nature of music & the non-linear systems that we use to playback this audio stream & the non-linear hearing that we use. By isolating the noise & saying that it can't be heard is a changing the context of what is being said - it may well be the noise modulation that's at play here & is audible?
- asking someone to prove what is a theory, at the moment, is rushing it a bit - this needs careful design of experiments to prove or disprove but that doesn't negate the possibility of it's existence as a mechanism to explain some of the incongruities that face some of us (I know you don't seem to experience any of these incongruities & that's fine)

 

[Phelonious Ponk]

Tim, a couple of points:
- you fail to take account of the dynamic nature of music & the non-linear systems that we use to playback this audio stream & the non-linear hearing that we use. By isolating the noise & saying that it can't be heard is a changing the context of what is being said - it may well be the noise modulation that's at play here & is audible?
- asking someone to prove what is a theory, at the moment, is rushing it a bit - this needs careful design of experiments to prove or disprove but that doesn't negate the possibility of it's existence as a mechanism to explain some of the incongruities that face some of us (I know you don't seem to experience any of these incongruities & that's fine)

When I hear "noise" I think linear. Noise floor. Are we talking about non-linear distortions that vary with the music signal? If we are, I'm very confused by recommendations that I shut off the music, turn my system all the way up and put my ear to the speaker to see if it's audible. Maybe we're having two separate conversations here. Though the same priciples would apply to inaudible non-linear distortion.

And I'm not looking for proof, John, just evidence. An explanation of this possibility you're referring to. I've seen none of it. Neither here or elsewhere. Just speculation, theory. There are all kinds of things that can be measured that can't be heard. And if they can't be heard, they have no effect on what we hear. It really is that simple. If, somehow, they are a "mechanism" contributing to an effect that can be heard, then we're right back in the world of the audible. Which means it is measurable with both instruments and ears. And yet, while we can speculate, theorize, and doubt that all may not be well for pages, we haven't been able to come up with a single bit of evidence of any kind pointing to an audible affect as a result of further reducing the inaudible mechanisms. If there are high-end audio businesses that's business models are based on the theory that further reduction of inaudible artifacts somehow audibly improves the signal - and there are quite a few of them - how is it that none of them have come up with any publishable, viable, credible evidence? Why is it, that after all these years they're still presenting it all as "theory?" Still presenting the further reduction of the inaudible as their best case for an audible result?

Tim

 

[jkeny]

When I hear "noise" I think linear. Noise floor. Are we talking about non-linear distortions that vary with the music signal? If we are, I'm very confused by recommendations that I shut off the music, turn my system all the way up and put my ear to the speaker to see if it's audible. Maybe we're having two separate conversations here. Though the same priciples would apply to inaudible non-linear distortion.

But is the result of noise interacting with a non-linear system, linear or non-linear? Please show me how you know, have tested & can prove that it is non-linear INAUDIBLE distortion?

And I'm not looking for proof, John, just evidence. An explanation of this possibility you're referring to. I've seen none of it. Neither here or elsewhere. Just speculation, theory. There are all kinds of things that can be measured that can't be heard. And if they can't be heard, they have no effect on what we hear. It really is that simple. If, somehow, they are a "mechanism" contributing to an effect that can be heard, then we're right back in the world of the audible. Which means it is measurable with both instruments and ears. And yet, while we can speculate, theorize, and doubt that all may not be well for pages, we haven't been able to come up with a single bit of evidence of any kind pointing to an audible affect as a result of further reducing the inaudible mechanisms. If there are high-end audio businesses that's business models are based on the theory that further reduction of inaudible artifacts somehow audibly improves the signal - and there are quite a few of them - how is it that none of them have come up with any publishable, viable, credible evidence? Why is it, that after all these years they're still presenting it all as "theory?" Still presenting the further reduction of the inaudible as their best case for an audible result?

Tim

That is the nature of what is being discussed now - stuff that hasn't yet been proven. Again using the reasoning that it hasn't been proven yet & therefore it mustn't exist, is a circular argument, don't you think?
You also seem to mix up & get caught up in the commercial nature of high-end, as if this has some relevance? Some of the high-end is pushing the boundaries, some is not - it's just pushing commercial interests. But again this in no way defines what we are talking about or brings any light to the situation.

 

[jkeny]

There are all kinds of things that can be measured that can't be heard. And if they can't be heard, they have no effect on what we hear. It really is that simple.
Tim

It isn't that simple, unfortunately. If your assumption was based on the fact that you have everything measured in situ, dynamically & with some real applicability to what we listen to - music, I would be inclined to believe you. But it's not! So when you say "can't be heard" you need to be specific about this statement.

Using Sine Waves is NOT a proof of audibility - we are talking about interactions, not static, non-interacting systems that are designed to be somewhat easier to measure & make sense of the measurements.

So it is a gross simplification to say that with all music turned off you can hear no noise from your speakers therefore there is no noise in your system!!

 

[Robh3606]

This seems to be at the heart of a lot of people's jaundiced view of vinyl - it's SNR, it's snap,crackle,pop artifacts, etc. In digital we may be trading these artefacts & limitations for other distortions which are less obvious at the conscious level but more intrusive!

What might they be??

So it is a gross simplification to say that with all music turned off you can hear no noise from your speakers therefore there is no noise in your system!!

First of all there will always be noise and everyone knows that. The question is if it is at a level that is objectionable. I never particulary liked grove noise on LP's but once the music started it was for the most part masked except for quietest of pasages. You could listen right through the rather audible surface noise and enjoy the music. Are you proposing that there are some distortions in digital that we don't obviuosly hear like surface noise that unconsciously detracts from our enjoyment??


Rob

 

[Phelonious Ponk]

But is the result of noise interacting with a non-linear system, linear or non-linear? Please show me how you know, have tested & can prove that it is non-linear INAUDIBLE distortion?

You're speculating, rather wildly I might add, about the possibility that linear noise, inaudible when no music is playing, can somehow attach itself to the signal and become audible, non-linear distortion when music is playing. I've never seen any evidence of this. I can't recall a single instance of an EE without a financial stake even referring to such an unlikey possibility,* and I spend WAY too much time on audio sites; ask my wife. This is an unsupported, unfounded "what-if." You've shown absolutely no evidence beyond supposing it could happen, nothing indicating that it has happened, yet you think the burden of proof is mine? And you're asking me to prove a negative, no less? Sorry, it just doesn't work that way, John. This is the point at which you show me. You're proposing a theory. Back it up.

That is the nature of what is being discussed now - stuff that hasn't yet been proven. Again using the reasoning that it hasn't been proven yet & therefore it mustn't exist, is a circular argument, don't you think?

No, because if this metamorphasis of inaudible linear noise into audible linear distortion is, indeed, audible, it is measurable, by both instruments and ears. And therefore, it is demonstrable. That no one, in an industry full of solutions to such problems, has ever bothered to demonstrate the audibility of that for which they continue to sell the silencing pretty much tells the tale. Close the circle. Put some argument in your argument.

Tim

*Guys -- EEs or otherwise -- with a financial stake in the reduction of the inaudible promote these kinds of theories all the time in high-end audio. They have created a whole subculture of supposed what-ifs.

 

[jkeny]

What might they be??
Rob

I'm saying that issues like RFI, EMI, noise modulation, Jitter, pre-echoes to name but a few can result in audio that is less than satisfying but not as obvious as groove noise or snap, crackle, etc.
These digital distortions can give rise to sibilance or to a general unease with the sound - an inability to full relax into it!

 

[Phelonious Ponk]

I'm saying that issues like RFI, EMI, noise modulation, Jitter, pre-echoes to name but a few can result in audio that is less than satisfying but not as obvious as groove noise or snap, crackle, etc.
These digital distortions can give rise to sibilance or to a general unease with the sound - an inability to full relax into it!

Finally we agree, John. I not only agree, I'll expand and simplify: High frequency distortions are more annoying than low frequency distortions. And that includes the snap, crackle pop of record wear. Reasonable amounts of normal surface noise which is deep? Very easy to ignore. Same for moderately boomy speakers and even a low hum. Once you hear it, it may begin to drive you to distraction but it's pretty easy not to hear it. The glaring treble of bad, early digital? Almost impossible to ignore. Can't remember the last time I heard it, though.

Tim

 

[jkeny]

You're speculating, rather wildly I might add,

in your opinion!

about the possibility that linear noise, inaudible when no music is playing, can somehow attach itself to the signal and become audible, non-linear distortion when music is playing. I've never seen any evidence of this.

RFI noise can be inaudible without music playing. Do you deny RFI noise can have an audible effect on audio?

I can't recall a single instance of an EE without a financial stake

Are you getting personal now & entering into the realm of motivation?

even referring to such an unlikey possibility,* and I spend WAY too much time on audio sites; ask my wife. This is an unsupported, unfounded "what-if." You've shown absolutely no evidence beyond supposing it could happen, nothing indicating that it has happened, yet you think the burden of proof is mine? And you're asking me to prove a negative, no less? Sorry, it just doesn't work that way, John. This is the point at which you show me. You're proposing a theory. Back it up.

I'm afraid we are getting to the usual type of adversarial statements now - you prove it, no you prove it - unhelpful to say the least. You made statements about noise which I called you on, simple as that!

No, because if this metamorphasis of inaudible linear noise into audible linear distortion is, indeed, audible, it is measurable, by both instruments and ears. And therefore, it is demonstrable. That no one, in an industry full of solutions to such problems, has ever bothered to demonstrate the audibility of that for which they continue to sell the silencing pretty much tells the tale. Close the circle. Put some argument in your argument.

Tim

*Guys -- EEs or otherwise -- with a financial stake in the reduction of the inaudible promote these kinds of theories all the time in high-end audio. They have created a whole subculture of supposed what-ifs.

Sorry, Tim your constant reference to this "high-end" & "financial stake" leads me to believe your concern seems to be with the industry & fear of being duped, not with the substance of what is being discussed here?

 

[Robh3606]

I'm saying that issues like RFI, EMI, noise modulation, Jitter, pre-echoes to name but a few can result in audio that is less than satisfying but not as obvious as groove noise or snap, crackle, etc.
These digital distortions can give rise to sibilance or to a general unease with the sound - an inability to full relax into it!

Well which one of the above would not be present in an all analog system?? Jitter possibly, can we substitute wow flutter and rumble?? I used to hear pre-echo on vinyl so it seems we have almost the same list without surface noise. So is that what's missing?? A more obvious but endering nasty that we just accepted as part of the experinence that lulled us into a false sense of security because we accepted the obvious??

Rob

 

[jkeny]

Finally we agree, John. I not only agree, I'll expand and simplify: High frequency distortions are more annoying than low frequency distortions. And that includes the snap, crackle pop of record wear. Reasonable amounts of normal surface noise which is deep? Very easy to ignore. Same for moderately boomy speakers and even a low hum. Once you hear it, it may begin to drive you to distraction but it's pretty easy not to hear it. The glaring treble of bad, early digital? Almost impossible to ignore. Can't remember the last time I heard it, though.

Tim

You have absolutely no sibilance in your digital now, congratulations!!

 

[Phelonious Ponk]

You have absolutely no sibilance in your digital now, congratulations!!

I've recorded several hundred voice-overs, John. I can assure you that if you don't have any sibilance in your digital, or analog, you have lost some high frequencies somewhere. Sibilance is in the mouth and the microphone long before it gets to tape or digits. Do you know how to tell where natural sibilance ends and "digital" sibilance begins? Like I said, I've recorded several hundred voice overs, I know what it sounds like. Do I hear any excess sibilance in my digital? On good recordings, no, I don't. Sorry you do.

Tim

 

[jkeny]

Well which one of the above would not be present in an all analog system??

So are you trying to tell me that these noise sources have exactly the same effect in digital systems as they do in analogue playback?

Jitter possibly, can we substitute wow flutter and rumble??

No we can't. I would refer you to reading Amir & Donh postings in this forum on jitter to understand what it's audible effects are

I used to hear pre-echo on vinyl so it seems we have almost the same list without surface noise.

Is this pre-echo the same as what is created by certain interpolation filters?

So is that what's missing?? A more obvious but endering nasty that we just accepted as part of the experinence that lulled us into a false sense of security because we accepted the obvious??

Rob

 

[jkeny]

I've recorded several hundred voice-overs, John. I can assure you that if you don't have any sibilance in your digital, or analog, you have lost some high frequencies somewhere. Sibilance is in the mouth and the microphone long before it gets to tape or digits. Do you know how to tell where natural sibilance ends and "digital" sibilance begins? Like I said, I've recorded several hundred voice overs, I know what it sounds like. Do I hear any excess sibilance in my digital? On good recordings, no, I don't. Sorry you do.

Tim

I'll let some of the other recording engineers answer this! I can answer from the playback side - with a lot of digital playback you have signs of the issues I spoke of & given the right recording, you will hear the distortions that these underlying issues give rise to - the most notable & easily identified symptom being sibilance. I'm not talking about the natural & correct rendition of fricative consonants but an exaggeration of this natural sound leading to what is called sibilance!!

I know what fricative sounds like when I'm listening to someone talking & if you don't strive for this naturalness, then it does not auger well for your playback system's sound, I'm afraid!!

 

[jkeny]

You're speculating, rather wildly I might add, about the possibility that linear noise, inaudible when no music is playing, can somehow attach itself to the signal and become audible, non-linear distortion when music is playing. I've never seen any evidence of this.

Ever hear of IMD - intermodulation distortion?

 

[Phelonious Ponk]

Ever hear of IMD - intermodulation distortion?

Yes. Can you help me with how it is related to RFI, EMI and the other stuff we're talking about here -- which yes, can have an audible effect on audio when they are audible. And if they are, that audibility should be demonstrable. So why are we thrashing about in the realm of theory. Oh, and perhaps "wild" was my opinion. that this is speculation is pretty obvious until those theorizing it come up with something more substantial.

And John, my apologies. I didn't mean to imply anything about your personal professionalism. I know nothing about it; I' completely unfamiliar with your products and what you say the do. But I also know there are more than a few high-end companies out there selling expensive solutions to non-problems. I will assume you are not one of them.

Tim

 

[jkeny]

You're speculating, rather wildly I might add, about the possibility that linear noise, inaudible when no music is playing, can somehow attach itself to the signal and become audible, non-linear distortion when music is playing. I've never seen any evidence of this.

So we have noise at a frequency of 21KHz. Is this audible? When music is playing it can fold-down into the audible range through IMD, right? Is this wild speculation?

Yes. Can you help me with how it is related to RFI, EMI and the other stuff we're talking about here -- which yes, can have an audible effect on audio when they are audible. And if they are, that audibility should be demonstrable. So why are we thrashing about in the realm of theory. Oh, and perhaps "wild" was my opinion. that this is speculation is pretty obvious until those theorizing it come up with something more substantial.

Answered above. I'm not sure who is trashing about? IMD is pretty obvious & basic to audio!

And John, my apologies. I didn't mean to imply anything about your personal professionalism. I know nothing about it; I' completely unfamiliar with your products and what you say the do. But I also know there are more than a few high-end companies out there selling expensive solutions to non-problems. I will assume you are not one of them.

Tim

Apology accepted!

 

[Phelonious Ponk]

So we have noise at a frequency of 21KHz. Is this audible? When music is playing it can fold-down into the audible range through IMD, right? Is this wild speculation?

I don't know. First you have to show, with measurements, that there is a noise, at 21Khz, then that is folding down into the audible range, then that the amplitude of the noise is high enough relative to the music to be audible. Get all that, and you've got evidence. If it's strong enough, it's acceptable on its own. If it is on the edge of audible, you'll need to confirm it with blind listening. Short of that? Speculation. Or to use your words, theory. I withdraw the adjective "wild."

Tim