Audio Science: Does it explain everything about how something sounds?

[NorthStar]

No sweat; it's educative for me to research. And I find it fascinating that height information can be recorded and reproduced and distinguished in music listening and in films too. ...With the right mixing console techniques and mic positioning.

With today's new 3D elevated sound encoding and decoding (Dolby Atmos, DTS:X and Auro-3D) they can make up-firing speakers with a special crossover network with a notch in the 10kHz area for great height perception when directed and reflected from our ceiling. ...In movie sound.

? Music listening is an art, and art is a science that can be further developed and improved. ...Acoustics are mathematically scientific, and music listening is directly related to room's acoustics. ...Science @ the service of the art, the art of listening.

Painting is an art, and we can improve our skill in painting by developing our painting techniques. ...Photography, cameras in filmmaking, cinematography, composition, and same for music composition techniques in recordings and in reproduction from and to the locales, the venues where the recordings are made/executed, and the venues where those music recordings are reproduced, our rooms. ...The execution, positioning of our ears in relation to our entire environment.

 

[Rodney Gold]

DBT's and ABX tests have sweet fanny all to do with preferences , all they can indicate is differences. and in audio it is almost impossible for any homeowner to do any of these tests anyway ..

I dont need toole or anyone else to come do measurements in my room , there are excellent tools out there that will and I understand what to do to correct the sound TO MY LIKING!!!

Ultimately - ALL room treatments or the sound at listening position have to point to a target curve , whether they are physical or electronic , and the target curve is purely subjective ..

If you show me a FR taken at listening position with a waterfall graph , one can to some extent predict how that system will sound.. but not with any certainty.

At any rate - I am a relative newbie to this forum , but have been active on many other for many years.. and one thing I can certainly tell you is that there will NEVER be a resolution to the objective/subjective debate. These discussions are pointless and only act as a way to polarise the forum..

 

[Speedskater]

What measurement is used to tell us that one speaker, or one phono cable, can produce a better image, or a more believable soundstage than another speaker or phono cable can? ..................................

For the phono cable the soundstage difference measurement is easy (for the speaker not so easy). If you demonstrate two phono cables that have different soundstage's. Then measurement of a third phono cable can tell which of the first two cables, it's soundstage will be more similar.

 

[amirm]

At any rate - I am a relative newbie to this forum , but have been active on many other for many years.. and one thing I can certainly tell you is that there will NEVER be a resolution to the objective/subjective debate. These discussions are pointless and only act as a way to polarise the forum..

The debate on acoustics spans both camp. Very little of the research in acoustics has made it into forums in a broad way. As a result, both sides are incredulous at how much is known both about our perception and preferences. For example, if I asked you the threshold of hearing a reflection from the floor versus ceiling, would you know the answer? Would you know our preference for a front versus floor reflection? How about what makes speech and vocals intelligible? All of these have been extensively researched and not just for audiophile purposes but for education (e.g. classroom speaker intelligibility) and medical reasons.

The research into every example I gave is based 100% on listening tests. And those listening tests by the way, are not even blind in many cases! But they are controlled studies seeking understanding of one factor at a time. By combining all of this research together, a picture emerges that has applicability to our hobbies.

Because of lack of much exposure to these topics, objectivists are similarly situated to these topics. I know I was, sitting there thinking someone is taking me for an idiot when I first heard that the angle of a reflection from a loudspeaker determines our preference. Oh really? The same identical reflection coming out of a loudspeaker creates different preference in a room? It can't be, right? Well, it is. After a lot of protests in my mind, I started on long extensive journey to determine how correct his was. After reading paper after paper in both ASA and AES, well, well beyond Harman/NRC research, I realized that even though I knew everything people talk about on forums, I was full of misconceptions and really didn't know a darn thing.

I then went on to explain these topics on AVS Forum where objectivists went after like there was no tomorrow. I can point you to thread after thread where people refused to want to hear and instead wanted to follow folklore they read on forums. The level of anger and frustration they showed is well beyond anything that has happened here.

Admittedly, I am pleasantly surprised by our objectivists here adopting these views after being exposed to them. This, I did not experience before so I say this is the exception than the rule. This is research that people have not taken the time to learn and cuts through both camps. It is a mistake and sign of misunderstanding to continue to position it as a) "harman" data and b) a fight between objectivists and subjectivists and c) no one will change their views. All of these are incorrect.

 

[PeterA]

For the phono cable the soundstage difference measurement is easy (for the speaker not so easy). If you demonstrate two phono cables that have different soundstage's. Then measurement of a third phono cable can tell which of the first two cables, it's soundstage will be more similar.

That's interesting. By "demonstrate two phono cables that have different soundstage's" do you mean through measurements or through listening or through both? How is the demonstration conducted?

I realize that I may be coming across as either a skeptical subjectivist, argumentative, combative and even possibly dogmatic in this thread, but I am trying to understand whether or not our ears can tell us anything at all that measurements can not yet explain or if they are only used to confirm what we already know from the audio sciences. This is the central point to my inquiry in this thread and it is still not clear to me. Posts in this thread don't seem to address this question directly, or I am just missing what is obvious to everyone else. Or perhaps there is no consensus on the matter.

I've read interviews of Nelson Pass and Alon Wolf and others. Measurements and audio science are clearly critical to their design work and the development of their products. Pass writes, however, that they also do a lot of listening before a particular design is ready for release. Magico seems to be pushing technology. They rely heavily on measurements such as distortion figures and material sciences and computer modeling. So science is central to their work also. But Wolf wrote that only after he heard the new tweeter and mid range drivers with the new crossover that were developed for the latest M Project speaker, did he realize that this was the new direction that they needed to go. It seems that science got them there but listening tests confirmed it for them.

So my thinking is that both objective audio science and subjective listening are essential elements needed to develop the best audio components. It can't be done exclusively with one and not the other. This does not seem controversial to me or a polarizing idea. Perhaps I am just missing what is obvious to everyone else, and if so, someone should just tell me that I am dense.

 

[Al M.]

I've read interviews of Nelson Pass and Alon Wolf and others. Measurements and audio science are clearly critical to their design work and the development of their products. Pass writes, however, that they also do a lot of listening before a particular design is ready for release. Magico seems to be pushing technology. They rely heavily on measurements such as distortion figures and material sciences and computer modeling. So science is central to their work also. But Wolf wrote that only after he heard the new tweeter and mid range drivers with the new crossover that were developed for the latest M Project speaker, did he realize that this was the new direction that they needed to go. It seems that science got them there but listening tests confirmed it for them.

So my thinking is that both objective audio science and subjective listening are essential elements needed to develop the best audio components. It can't be done exclusively with one and not the other. This does not seem controversial to me or a polarizing idea. Perhaps I am just missing what is obvious to everyone else, and if so, someone should just tell me that I am dense.

I completely agree with you, and no, you're not missing anything that should be obvious to anyone else.

 

[FrantzM]

FrantzM, how do you measure whether or not a listener will hear a sense of presence or have an emotional response to a pair of speakers or to a system? Can measurements tell us this for a particular component in a particular room? What measurements tell us if a particular speaker will sound more dynamic than another speaker? What measurement tells us which system, or cartridge has more resolution?

Is everything measurable and can we know everything from audio science about how a system will sound in a particular room if we don't use our ears? Clearly audio science tells us very much, but can we know and measure everything without using our ears?

Isn't some of audio science based on listening tests?

In other words, can our ears tells us things that audio science can not yet explain?

Peter



I wouldn't be interested in measuring the emotional response but rather what provokes this perception of presence. It is reasonable to assume that there is a mixture of things, call these sound waves if you will that provoke the emotion/perception, reaction... If it exist on the physical plane then it can be measured. We can correlate these measurements to your emotion/reaction/perception. I am by no means saying we know how to measure everything but sounds can be measured and rather well.. This is what a microphone does after all...

Back to the listening tests. Perceptions have to be reliable. I have been fooled myself into thinking that things were present that were not ... so assuming a reliable perception, by no measure a small feat, then we can (try to?) correlate what was measured to what was perceived. That is done in all fields and it is after all the way an audio designer arrives at his/her signature sound. They have to measure something else how would they replicate it ? IT is true that some given measurements may not correlate with a conistent perception then they go on a hunch ... That still does not mean that what provoked the perception cannot be measured.

So

 

[Al M.]

Peter



I wouldn't be interested in measuring the emotional response but rather what provokes this perception of presence. It is reasonable to assume that there is a mixture of things, call these sound waves if you will that provoke the emotion/perception, reaction... If it exist on the physical plane then it can be measured. We can correlate these measurements to your emotion/reaction/perception. I am by no means saying we know how to measure everything but sounds can be measured and rather well.. This is what a microphone does after all...

Back to the listening tests. Perceptions have to be reliable. I have been fooled myself into thinking that things were present that were not ... so assuming a reliable perception, by no measure a small feat, then we can (try to?) correlate what was measured to what was perceived. That is done in all fields and it is after all the way an audio designer arrives at his/her signature sound. They have to measure something else how would they replicate it ? IT is true that some given measurements may not correlate with a conistent perception then they go on a hunch ... That still does not mean that what provoked the perception cannot be measured.

So

Good points.

So perhaps instead of saying that audio science cannot yet measure everything, it would be more accurate to say that audio science in many cases cannot yet establish a quantifiable relation between measurement and human aural perception.

 

[Tony Lauck]

Good points.

So perhaps instead of saying that audio science cannot yet measure everything, it would be more accurate to say that audio science in many cases cannot yet establish a quantifiable relation between measurement and human aural perception.

How can anyone confidently state that audio science can measure everything, while simultaneously stating that it cannot yet establish a quantifiable relation between measurement and human aural perception?

This strikes me as utter nonsense. If there is a basis for this statement I would certainly like to know it. (Hint: "measure everything" in the context of audio science necessarily means "everything relative to human aural perception." We are talking audio, not physics.) Is there some hidden philosophical sub-text here?

 

[DaveC]

How can anyone confidently state that audio science can measure everything, while simultaneously stating that it cannot yet establish a quantifiable relation between measurement and human aural perception?

This strikes me as utter nonsense. If there is a basis for this statement I would certainly like to know it. (Hint: "measure everything" in the context of audio science necessarily means "everything relative to human aural perception." We are talking audio, not physics.) Is there some hidden philosophical sub-text here?

I agree.

If you can't find a correlation than you are assuming you can measure everything without the evidence to back it up, which is a lot like having religious faith in science.

And again, part of the issue is that most are not aware of the extent of audio science (myself included). For example, most have not read all the papers Amir has read and are not as educated on the subject. Most don't have decades of experience interpreting the data. Also, a lot of research is done by for-profit companies who won't share everything they have learned. I think we leap to judgement too quickly when sometimes it's best to just admit we (personally) don't know everything and we don't need to form such absolute beliefs based on incomplete knowledge and understanding, instead it seems more reasonable to admit we don't know enough and should therefore keep an open mind, which means considering all information whether or not it fits in with our current beliefs or not.

 

[Atmasphere]

It is hard for me to put my head around the notion of "cannot be measured". A phenomenon very existence is proof of its measure-ability. There is no way around it , to me. IT is truism that we don't know it all but we are constantly moving ahead with our knowledge and our ways and means to measure. The measurement may exist but we may not yet be able to correlate it to our perception (I would like to say reliable perceptions but allow me to drop the epithet). I would go as far as saying that the interpretation is key to apprehend reality.

Someone else mentioned the microdynamics and macrodynamics ... What is a dynamic? .. a ratio.. How can it not be measurable?

I am an audiophile and for the most part i listen and like and base my preference on what I deem to be good sounding, an unreliable metrics. ... I fully know that my prejudices are always there and however much I fight them they color my perceptions. yet one can be trained to listen better and I believe that audiophiles do that (to a certain extent) along the way of their audio experiences...
That a phenomenon is not measured is not proof on it not being measurable. I'll stop there...

There is nothing, nothing, that a measurement can not discern in the audible range. However, measurements do not reveal preferences. There is nothing from the mic wiggle to the speaker wiggle that can not be measured.

Folks, while Harman showed some specific correlation in speaker measurements, don't confuse the ability to measure a simple audio signal with what we like in sound and preferences.

No objectivists says we can measure what Johnny likes in his stereo system. Johnny doesn't even know from year to year what he likes, or maybe from day to day. Johnny hearing system degrades nearly every year of his life.

The problem here is not our understanding of how to measure, but instead our understanding of how our human hearing/perceptual rules work. The more we understand about the latter, the better we will be able to figure out what is a 'good' example of the former.

The Audio Industry in general has ignored much of the physiological research that has gone on in the last 45-50 years so our test and measurement regime is based on our understanding of hearing from what we knew in the 1950s (even research from the mid 1960s has been ignored). We fix that- then we make progress.

In the meantime, we will be putting up with reviews wherein the equipment sounds great but 'measures' like crap.

Hey, just wait a minute everyone here! Some of you gentle folks have awesome hi-fi stereo systems. ...And some even have them acoustically calibrated...the room and all.

When was the last time you invited Dr. Floyd E. Toole, Dr. Sean Olive, Dr. Ethan Winer, and doctor Amir in your room to take some serious measurements?
All these experts have been in many rooms, they have heard many great sounding hi-fi music stereo systems from many high well calibrated audiophiles with all the necessary applications, speakers positioning, MLP, room's interaction and integration with the systems, ...they listened extensively, they measured extensively, they listened some more and made corrections when appropriate and when allowed by their owners.

You lost me at 'Dr' Ethan Winer. He is nowhere near the same league as the others you mentioned and it is not right and meet to see his name in the same sentence. He has been debunked a number of times.

 

[ddk]

Peter



I wouldn't be interested in measuring the emotional response but rather what provokes this perception of presence. It is reasonable to assume that there is a mixture of things, call these sound waves if you will that provoke the emotion/perception, reaction... If it exist on the physical plane then it can be measured. We can correlate these measurements to your emotion/reaction/perception. I am by no means saying we know how to measure everything but sounds can be measured and rather well.. This is what a microphone does after all...

Back to the listening tests. Perceptions have to be reliable. I have been fooled myself into thinking that things were present that were not ... so assuming a reliable perception, by no measure a small feat, then we can (try to?) correlate what was measured to what was perceived. That is done in all fields and it is after all the way an audio designer arrives at his/her signature sound. They have to measure something else how would they replicate it ? IT is true that some given measurements may not correlate with a conistent perception then they go on a hunch ... That still does not mean that what provoked the perception cannot be measured.

So

Emotional responses are subjective and depends on the individual. For your entertainment;

david

 

[jkeny]

The problem here is not our understanding of how to measure, but instead our understanding of how our human hearing/perceptual rules work. The more we understand about the latter, the better we will be able to figure out what is a 'good' example of the former.

The Audio Industry in general has ignored much of the physiological research that has gone on in the last 45-50 years so our test and measurement regime is based on our understanding of hearing from what we knew in the 1950s (even research from the mid 1960s has been ignored). We fix that- then we make progress.

In the meantime, we will be putting up with reviews wherein the equipment sounds great but 'measures' like crap.

Absolutely correct. I wouldn't be surprised if the existing accepted thresholds of hearing are overturned/significantly revised when more sophisticated tests become available - tests that incorporate some of the "rules" of auditory processing that we discover this research progresses. When/if that begins to happen along with more sophisticated test regimes (signals & measurements) we may get a better handle on the correlation between measurement & our perception of audio devices.

However, we are very much in the stone age as far as this is concerned & hence the various battles/flame wars that spontaneously combust

You lost me at 'Dr' Ethan Winer. He is nowhere near the same league as the others you mentioned and it is not right and meet to see his name in the same sentence. He has been debunked a number of times.

Ha! Winer who would have us back to 50 years ago when all audio problems were, he claims, last solved !

 

[Al M.]

How can anyone confidently state that audio science can measure everything, while simultaneously stating that it cannot yet establish a quantifiable relation between measurement and human aural perception?

This strikes me as utter nonsense.

What is so controversial with stating that we can measure everything about electrical/sound waves, but we cannot yet establish what all that means for, how it correlates with, human aural perception?

Below statements very much touch upon my point:

The problem here is not our understanding of how to measure, but instead our understanding of how our human hearing/perceptual rules work. The more we understand about the latter, the better we will be able to figure out what is a 'good' example of the former.

The Audio Industry in general has ignored much of the physiological research that has gone on in the last 45-50 years so our test and measurement regime is based on our understanding of hearing from what we knew in the 1950s (even research from the mid 1960s has been ignored). We fix that- then we make progress.

Absolutely correct. I wouldn't be surprised if the existing accepted thresholds of hearing are overturned/significantly revised when more sophisticated tests become available - tests that incorporate some of the "rules" of auditory processing that we discover this research progresses. When/if that begins to happen along with more sophisticated test regimes (signals & measurements) we may get a better handle on the correlation between measurement & our perception of audio devices.

However, we are very much in the stone age as far as this is concerned & hence the various battles/flame wars that spontaneously combust

 

[Atmasphere]

However, we are very much in the stone age as far as this is concerned & hence the various battles/flame wars that spontaneously combust

Got that right in spades!

IMO the audio industry ignores human hearing/perceptual rules as an inconvenient truth. We've spent so much time convincing people what good specs are supposed to look like on paper that we'll collectively have egg on our face if the specs prove out to be the Emperor’s New Clothes (which in fact they are). Human nature being what it is the most powerful motivator is to look good, the 2nd most powerful in the world is to not look bad. Until we can overcome the latter these discussions will continue... in all likelihood our grandchildren will be having the same conversations as they were going on in the 1960s not a lot different from how they are today...

 

[jkeny]

What is so controversial with stating that we can measure everything about electrical/sound waves, but we cannot yet establish what all that means for, how it correlates with, human aural perception?

Below statements very much touch upon my point:

No, my text does not mean "that we can measure everything" - let me be clear - I'm pretty sure that the "new" measurements (which incorporate psychoacoustic models) will involve very different test signals to what are currently used & most likely very heavy processing than is currently used. This is very far from the statement "that we can measure everything".

 

[Al M.]

However, we are very much in the stone age as far as this is concerned & hence the various battles/flame wars that spontaneously combust

Got that right in spades!

IMO the audio industry ignores human hearing/perceptual rules as an inconvenient truth. We've spent so much time convincing people what good specs are supposed to look like on paper that we'll collectively have egg on our face if the specs prove out to be the Emperor’s New Clothes (which in fact they are).

Yes, as long 'measurement objectivists' have religious faith in their 'science' of a limited spec list and keep ignoring the importance of human hearing/perceptual rules we will have those battles.

 

[ddk]

Got that right in spades!

IMO the audio industry ignores human hearing/perceptual rules as an inconvenient truth. We've spent so much time convincing people what good specs are supposed to look like on paper that we'll collectively have egg on our face if the specs prove out to be the Emperor’s New Clothes (which in fact they are). Human nature being what it is the most powerful motivator is to look good, the 2nd most powerful in the world is to not look bad. Until we can overcome the latter these discussions will continue... in all likelihood our grandchildren will be having the same conversations as they were going on in the 1960s not a lot different from how they are today...

What's interesting is that manufacturers and designers seem to be clear on the values and roles of both subjective and objective sciences, yet the audiophile camps remain obstinate.

david

 

[Al M.]

No, my text does not mean "that we can measure everything" - let me be clear - I'm pretty sure that the "new" measurements (which incorporate psychoacoustic models) will involve very different test signals to what are currently used & most likely very heavy processing than is currently used. This is very far from the statement "that we can measure everything".

Fair point.

However, different test signals to what are currently used, which take into consideration psychoacoustic measurements, still don't mean that we cannot measure everything about electrical/sound waves -- which is what the 'measurement objectivists' are all about. It's just the manner we go about those measurements, and the question what they all mean for human perception, that are at stake here.

 

[jkeny]

Got that right in spades!

IMO the audio industry ignores human hearing/perceptual rules as an inconvenient truth. We've spent so much time convincing people what good specs are supposed to look like on paper that we'll collectively have egg on our face if the specs prove out to be the Emperor’s New Clothes (which in fact they are). Human nature being what it is the most powerful motivator is to look good, the 2nd most powerful in the world is to not look bad. Until we can overcome the latter these discussions will continue... in all likelihood our grandchildren will be having the same conversations as they were going on in the 1960s not a lot different from how they are today...

Yes, agreed & unfortunately the audio industry is ignored by the research efforts in auditory processing.
But, as regards the audio industry itself not doing any of it's own research, I might be naive but I believe that there are some new audio initiatives that might challenge current lazy thinking & if successful in the market may also prod others audio manufacturers to take note. Th etechnologies - MQA from Meridian with their time-smearing reduction techniques; Dave, a new DAC from Chord with 1,000,00 taps ( & Rob Watt state sthat injected noise at -200dB is audible)