"The more I learn, the less I know..."

[Ciunas Audio]

That might be true for members of this forum,

what might?

but the field of auditory research is large and continues to be investigated the world over. There are, therefore, many experts in this particular field, of whom Dr Crum is perhaps the best known through her lectures with AES.

I don't disagree

That is different from the mechanics of hearing, however, which have been known for quite some time.

Oh, AFAIK (but what do I know anyway), the mechanics of hearing are still being worked out! There are active processes involving feedback at the physiological level which adapt to & change the ears response! I don't think you can just isolate the ear from the processing engine that receives, analysis & ultimately makes sense of the vibrations we call sound. But I could be wrong

 

[Ciunas Audio]

It is more than a frequency analysis. No-one claims it is only a frequency analysis.

But this does not distract from the fact that with considering purely the frequency domain already an awful lot of insight can be gained.

After all the cochlea is a mechanical spectrum analyser, be it non-linear, time-variant, and with some feedback from higher neural levels.

OK, but how much do those extras change it from a "mechanical spectrum analyzser" into a much more complex device? Do you have an equivalent piece of measuring equipment that might act as a good analogy to understanding what these attributes (non-linear, time-variant, with feedback) add to this mechanical analyser? For instance non-linear seems like it might just complicate matters significantly, no?

Seems like you are speculating as to the knowledge of others.

Only speculating about the posters on this thread based on the evidence presented, not to the whole world, you understand?

How much do tinnitus doctors know, you think? Or the developers of perceptual codecs?

See above

Maybe this helps:

http://www.aes.org/sections/pnw/ppt/jj/hearingtutorialv1.ppt

Or read the whole lot ...

http://www.aes.org/sections/pnw/ppt.htm

Yes, if I get the time, thanks!

 

[adamdea]

We are all speculating as nobody knows the full, accurate model of our sense of hearing. I have a view that it is more than a frequency analysis but I could be wrong. Not being an expert in the field, I don't know enough about the details to set forth a convincing enough argument (anyway, there will always be someone who takes the contrary view). I realise that this how we test our viewpoints & possibly adjust or change them completely (yea, like that's going to happen ).

I will take away from this the adjustment to my thinking that maybe a reduction in the ear's sensitivity to hearing the top octave does not cause us much degradation in our abilities to listen fully to music. In fact it causes so little that we still hear music as we did when our hearing was not degraded. Or that we think we do because our auditory memory is not reliable over a long window.

I was hoping that others might also be open to the notion that there is more going on with our hearing than just a frequency analysis & therefore a reduction in the top octave sensitivity might not be of such importance.

I was also hoping that in the spirit of the thread's title that we might have a reasonable discussion much like the one I started reading on ABX testing (which I expected to be incendiary but wasn't).

As titled "The more I learn, the less I know" - I already freely admitted that this is the case with me. I also think, as Plinius says, I proved it in this thread .Others have done an exemplary job in also demonstrating this.

So let's all step back from the plate for a bit & realise that none of us are experts in the field of hearing & it is a very complicated area of investigation. Let's all take a moment to realise that we don't know it all!

This is a very gracious post, and makes me feel a little foolish.

I think the tone of this thread was however (and I stress I am not aiming this at you jkeny) set by the OP, in which the apparent humility of the title made a striking contrast with the implicit insults and claims to higher knowledge contained in the actual post.

I agree that we should be humble about the limits of our own knowledge, but I consider that it is not humble but extraordinarily presumptuous to claim, under the guise of humility about "our" knowledge, that the entire body of scientific thinking on a subject is flawed (whoops I'm off again).

I hope to find a way of engaging in these discussions without leaving people feeling annoyed or insulted. I promise however that I do not get on my high horse because i think I have nothing to learn, but because I really do want to learn and perhaps take everything a bit too seriously.

 

[Werner]

Do you have an equivalent piece of measuring equipment that might act as a good analogy to understanding what these attributes (non-linear, time-variant, with feedback) add to this mechanical analyser? For instance non-linear seems like it might just complicate matters significantly, no?

Yes, it complicates matters. This means that quantifying this takes a lot of time, effort, and test subjects. It also means that one ends of with a gigantic amount of data, impossible to capture in a few figures of merrit.

So what? Trust it that the people who do this for a living actually are doing this.

Pertaining to our little hobby here there really is not much left that is shrouded in mystery.

 

[sergeauckland]

It is more than a frequency analysis. No-one claims it is only a frequency analysis.

But this does not distract from the fact that with considering purely the frequency domain already an awful lot of insight can be gained.

After all the cochlea is a mechanical spectrum analyser, be it non-linear, time-variant, and with some feedback from higher neural levels.




Seems like you are speculating as to the knowledge of others. How much do tinnitus doctors know, you think? Or the developers of perceptual codecs?

Maybe this helps:

http://www.aes.org/sections/pnw/ppt/jj/hearingtutorialv1.ppt

Or read the whole lot ...

http://www.aes.org/sections/pnw/ppt.htm

Werner,

Thank you very much indeed for posting that. One of the best explanations of hearing I've read.

I am particularly taken with Slide 22 and the summary, 64. Rather puts in perspective what people claim to be able to hear.

S.

 

[adamdea]

Yes, it complicates matters. This means that quantifying this takes a lot of time, effort, and test subjects. It also means that one ends of with a gigantic amount of data, impossible to capture in a few figures of merrit.

So what? Trust it that the people who do this for a living actually are doing this.

Pertaining to our little hobby here there really is not much left that is shrouded in mystery.

Very deftly put. I think it unwise,especially under the heading of, "knowing how much we don't know" to make rather bold assumptions about what the-people-who-do-this-for-a-living don't know.

I am very grateful for your link from the James Johntson AES slides which are very interesting. It is also interesting that JJ posts on hydrogen audio and if one wants to find out stuff it is possible to ask him. [NB obviously I know you know this Werner, but other people here might not.] .

I have a long term work in progress trying to formulate the correct question(s) to ask him so that i can understand whether there really is a problem with linear phase filters or not. I can;t ever get it right and donl;t want to waste his time asking the wrong question.

Connoisseurs of the horrors of internet communication may wish to read the 50 page or so road crash still I hope to be found on the Stereophile website during which 2, stereophile readers chose to pick a fight with JJ over amongst other things the concept of time resolution, and just got more and more rude and obtuse even when he revealed himself as one of the worlds leading authorities on the subject they were fighting over. The bit where they refuse to believe him because he doesn't have a PhD is actually quite upsetting though.

 

[Werner]

It is also interesting that JJ posts on hydrogen audio

JJ used to post at many fora, way back to the usenet days of the late 80s. I got to know him when still at university (me, that is).

understand whether there really is a problem with linear phase filters or not. I can;t ever get it right and donl;t want to waste his time asking the wrong question.

If you PM me I am happy to review/assist. There are a slide or two in the second ppt about hearing on that list I linked to that you may find intrigueing.

 

[cooky1257]

It is more than a frequency analysis. No-one claims it is only a frequency analysis.

But this does not distract from the fact that with considering purely the frequency domain already an awful lot of insight can be gained.

After all the cochlea is a mechanical spectrum analyser, be it non-linear, time-variant, and with some feedback from higher neural levels.




Seems like you are speculating as to the knowledge of others. How much do tinnitus doctors know, you think? Or the developers of perceptual codecs?

Maybe this helps:

http://www.aes.org/sections/pnw/ppt/jj/hearingtutorialv1.ppt

Or read the whole lot ...

http://www.aes.org/sections/pnw/ppt.htm

Very interesting, thanks for the links.

This stuff is truly fascinating and confirms my sense of wonder at the mechanisms and complex processes involved.

I particularly liked the 'why do we hear what we hear?" under Floobydust.

He's quite categorical in stating what happens beyond the basilar membrane is in most cases unknown. He also has the humility to acknowledge that what he posits is not inviolate-some would do well to learn from this .

The talk I’m about to give presents ideas gathered from a variety of papers and experiments, done by many people, over a long period of time.
– It is not inviolate.
– It is a discussion of phenomena
– The mechanism is, in most cases, unknown, once one gets beyond the basilar membrane
– There will be revisions as time goes on.
• There will ALWAYS be revisions

 

[sergeauckland]

Very deftly put. I think it unwise,especially under the heading of, "knowing how much we don't know" to make rather bold assumptions about what the-people-who-do-this-for-a-living don't know.


Connoisseurs of the horrors of internet communication may wish to read the 50 page or so road crash still I hope to be found on the Stereophile website during which 2, stereophile readers chose to pick a fight with JJ over amongst other things the concept of time resolution, and just got more and more rude and obtuse even when he revealed himself as one of the worlds leading authorities on the subject they were fighting over. The bit where they refuse to believe him because he doesn't have a PhD is actually quite upsetting though.

I find it interesting thinking about the credibility of posters on here and elsewhere, and whether "Those That Do It For A Living" are actually any more credible than those who don't.

Some TTDIFALs are commercially motivated, and want the arguement to go the way that benefits their commercial activity. Others may be academics, or otherwise engaged in Research and although they don't have a direct financial interest in the arguement, may have an emotional interest in a particular result. Others may also be academics/researchers and be genuinely interested in imparting knowledge without any obvious agenda. Others indeed may be commercially involved in Audio, but are nevertheless willing to impart knowledge regardless of whether it was in their commercial interest or not. I'm sure there are other categories of TTDIFALs I can't think of right now.

What I'm getting at is that on a Forum, where most posters are anonymous, or even when they're not, it's very hard to tell what a poster's motivation is, and therefore whether they're credible.

One benefit of the AES and other Learned Societies, is that papers are at least superficially reviewed by their peers before presentation, so there's a chance of the information being either correct, or if it's debatable, at least debated in a sensible manner.

S.

 

[ItemAudio]

Very deftly put. I think it unwise,especially under the heading of, "knowing how much we don't know" to make rather bold assumptions about what the-people-who-do-this-for-a-living don't know.

I am very grateful for your link from the James Johntson AES slides which are very interesting. It is also interesting that JJ posts on hydrogen audio and if one wants to find out stuff it is possible to ask him. [NB obviously I know you know this Werner, but other people here might not.] .

I have a long term work in progress trying to formulate the correct question(s) to ask him so that i can understand whether there really is a problem with linear phase filters or not. I can;t ever get it right and donl;t want to waste his time asking the wrong question.

Connoisseurs of the horrors of internet communication may wish to read the 50 page or so road crash still I hope to be found on the Stereophile website during which 2, stereophile readers chose to pick a fight with JJ over amongst other things the concept of time resolution, and just got more and more rude and obtuse even when he revealed himself as one of the worlds leading authorities on the subject they were fighting over. The bit where they refuse to believe him because he doesn't have a PhD is actually quite upsetting though.

He's particularly good on the analog nature of digital, too!

 

[Ciunas Audio]

Just returned from being out & looking at the JJ Johnston slides, thanks Werner. I looked into his work before but good to see the updated 2011 slides - there's a lot more information so I guess it's not all fully agreed or accepted yet i.e it is an active & on-going area of research.

One thing that strikes me immediately, as it does Cooky, are the statements which show that this is far from completely understood. The research is on-going & yes a lot seems to be known. Yes, I agree our little hobby is insignificant in the scheme of all this, Werner, & maybe my reading of it is different than yours but I wouldn't conclude "Pertaining to our little hobby here there really is not much left that is shrouded in mystery." I think there is still a fascinating areas of mystery & I note his added elements since my last reading of him so obviously he still is finding enough fascination & mystery to continue his research!

Let's just take a couple of examples in his slides:
- Disagreement over how the outer hair cells actually work & what their role is!
- a guess at what the inner hair cells do "They see the difference between the two high pass filters, IF THE FIRST IDEA ON THE PREVIOUS PAGE IS RIGHT"
- followed by a model that seems to work but as he says himself "the model does not represent the physics. That is not settled yet"
- the exact magnitude & shape of the curves he presents "are under a great deal of discussion" i.e not fully agreed or accepted

Something else we all seem to have overlooked but it was what I was getting at - we have two ears & hence we have further information as a result based on timing of signal arrival ILD & ITDs as they are called. This is not about frequencies, this is about time - it may vary by frequency but it is about time!

The emphasise on how bad pre-echo is (highlighted in red) has obviously not filtered through to audio designers as most still favour phase-linear digital.

A fascinating read & thanks for the links again. Again, this shows how little I know, I don't know how others feel? (Don't say yes it shows how little you know, please )

 

[Ciunas Audio]

Very deftly put. I think it unwise,especially under the heading of, "knowing how much we don't know" to make rather bold assumptions about what the-people-who-do-this-for-a-living don't know.

Just a correction, as this statement seems to be assigned to me, I quickly corrected Werner in his wrong assumption that I was speculating about everybody - I specifically said that I was referring to people posting here when I said "nobody is an expert in this". Apologies if it was written incorrectly but I hope this corrects the misunderstanding?

 

[adamdea]

Just a correction, as this statement seems to be assigned to me, I quickly corrected Werner in his wrong assumption that I was speculating about everybody - I specifically said that I was referring to people posting here when I said "nobody is an expert in this". Apologies if it was written incorrectly but I hope this corrects the misunderstanding?

Sure.

 

[adamdea]

I find it interesting thinking about the credibility of posters on here and elsewhere, and whether "Those That Do It For A Living" are actually any more credible than those who don't.

Some TTDIFALs are commercially motivated, and want the arguement to go the way that benefits their commercial activity. Others may be academics, or otherwise engaged in Research and although they don't have a direct financial interest in the arguement, may have an emotional interest in a particular result. Others may also be academics/researchers and be genuinely interested in imparting knowledge without any obvious agenda. Others indeed may be commercially involved in Audio, but are nevertheless willing to impart knowledge regardless of whether it was in their commercial interest or not. I'm sure there are other categories of TTDIFALs I can't think of right now.

What I'm getting at is that on a Forum, where most posters are anonymous, or even when they're not, it's very hard to tell what a poster's motivation is, and therefore whether they're credible.

One benefit of the AES and other Learned Societies, is that papers are at least superficially reviewed by their peers before presentation, so there's a chance of the information being either correct, or if it's debatable, at least debated in a sensible manner.

S.

Yes I agree. I think though that a sensible and realistic view has to be taken of people's claim to authority. People who are involved in the industry range from really brilliant designers operating at the cutting edge of audio engineering to, well, bodgers with a soldering iron. I'm pretty sure you could make and sell a dac from off the shelf chips without really understanding much of what was going on.

Even people who doubtless know their onions make commercial sales pitches which have to be treated with a pinch of salt. I felt that way about the ESS chief engineers' talk on Youtube explaining why their chips sound different from any other delta sigma dac

 

[Werner]

- we have two ears & hence we have further information as a result based on timing of signal arrival ILD & ITDs as they are called. This is not about frequencies, this is about time - it may vary by frequency but it is about time!

The emphasise on how bad pre-echo is (highlighted in red) has obviously not filtered through to audio designers as most still favour phase-linear digital.

A fascinating read & thanks for the links again. Again, this shows how little I know, I don't know how others feel? (Don't say yes it shows how little you know, please )

On myPad thing, sorry for bad editing.

First this is not JJ research, but a summary of present state of understanding, filtered through/for audio people.

As to why now updates? Afaik JJ retired and has time fir this stuff.


The two ear timing stuff has been knwon for decades. No mysteries here.

As for the pre-eco, please read again and put in proper context. Linear phase pre echo at 3kHz is indeed baaaad, try it yourself. But at 22kHz, above the nominal limit of adult hearing????
Lets say very subtle, at best. Try it.

 

[Ciunas Audio]

-snip-
As for the pre-eco, please read again and put in proper context. Linear phase pre echo at 3kHz is indeed baaaad, try it yourself. But at 22kHz, above the nominal limit of adult hearing????
Lets say very subtle, at best. Try it.

I presume you are making the common mistake of mixing up pre-echo with pre-ringing - the Gibbs effect which operates somewhere around the frequencies that you state?

It's quiet clear to me what he is saying particularly when he makes reference to linear-phase filters that MUST have pre-echo!!

Edit: Let's see what he says about pre-echo it's hardly "very subtle"

• Remember that bit about the filter time resolution and
coding?

– Pre-echo is really bad. Before an attack, you can hear injected
noise nearly down to the noise floor of the listening setup.
– Codec filterbanks, in order to be constant-delay (i.e. linear-
phase) must have pre-echo.
– Pre-echo can start the compression on the basilar membrane
before the signal arrives. This reduces the loudness of the
transient. Not good.​

• Not only that, pre-echo has some nasty consequences in
imaging. More on that later.

Hardly a description of a subtle effects, I think you'll agree, Werner?

So my point is, if this is so well known & well established information, as you make out Werner, why are so many digital designs done with linear-phase filters?

 

[Ciunas Audio]

-snip-
Even people who doubtless know their onions make commercial sales pitches which have to be treated with a pinch of salt. I felt that way about the ESS chief engineers' talk on Youtube explaining why their chips sound different from any other delta sigma dac

Which part do you disagree on:
- S-D dacs modulate the noise? or
- ESS DACs have addressed this modulation to some extent?

 

[Ciunas Audio]

-snip-

What I'm getting at is that on a Forum, where most posters are anonymous, or even when they're not, it's very hard to tell what a poster's motivation is, and therefore whether they're credible.

-snip-
S.

So let me see what you are saying - you don't debate the points made, you want to check the credentials of who is making the point? I see! If someone was talking about experimental results & the conclusions that they had arrived at, I could understand your position but that's not what's being discussed here. Frankly this is just a variation on the argumentum ad verecundiam

 

[Ciunas Audio]

-snip-

The two ear timing stuff has been knwon for decades. No mysteries here.

-snip-

Let's see what JJ has to say about this?

Interaural level differences (ILD’s)
There is much more data on this subject

• Much of the data is contradictory, or strongly individualized, or generalized to the point where it works “ok” for “most people”.
– Unfortunately, that’s the nature of the problem.

• Much data is proprietary, etc., as well.

• Just remember, ILD, ITD vary with frequency. That is the point.

So again, Werner, I wouldn't consider this a settled, agreed & "known about for decades, no mysteries here"

 

[Ciunas Audio]

On myPad thing, sorry for bad editing.

First this is not JJ research, but a summary of present state of understanding, filtered through/for audio people.

-snip-

So would you like to address the various areas I highlighted from his slides which show that in these areas, agreement still has to be reached!
I believe this supports the statement I made "Oh, AFAIK (but what do I know anyway), the mechanics of hearing are still being worked out!"