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

Status
Not open for further replies.
(...) It seems obvious that you consider room acoustics to be the most significant components of a playback system and if so, you certainly would not be alone. But I would attest that the more a system’s greatly raised noise floor is reduced, thereby allowing more of the previously inaudible music to now be audible, IOW, the closer a playback system gets to live music aka the absolute sound, the less important the room acoustics, first reflections, etc.
(...)


Stehno,

You will have a hard time defending this argument in this forum, but there is a lot of truth in your statement. I would be more prudent with the part "the closer a playback system gets to live music aka the absolute sound", but we have here many experiences that prove this concept.

I have seen too many systems installed in casual acoustics, never measured or treated, that after careful tweaking sound great. And yes the subjective noise floor is a critical concept, but most people minds stop on the signal to noise ratio.
 
MRI and CT machines are "burned-in" before clinical use, and it's well known that even after that the first few days of use don't give quite the image quality that is later obtained. It's probably not the wiring, more likely the X-ray tubes and/or superconductors, but who knows.

Fair enough. If they move the machine to a new wing, do they also give it time to settle in to its new environment?

Tim
 
I am not so pessimist as you - but unfortunately perhaps you are right, specially considering mainly sound reproduction in high quality stereo, the main area of interest of debate of more than 99% of this forum members.
It has nothing to do with pessimism. It has to do with data. Data in the form of reading hundreds of research papers and managing a large team of audio/signal processing experts while I was at Microsoft. Forum discussions are quite myopic in that regard because members don't have access to much body of research or involvement in research/engineering professionals.

Anyway there is still hope - e.g. at the last AES Convention there was a paper on microdynamics and measurements that could correlate it with the way it was perceived in listening experiences. Audiophiles have been talking about it since many decades ...
Well, there should be no rejoicing by audiophiles on this work as what they tested, is the true definition of the word, microdynamics. That is, short term variations in the dynamic range of music. Not some audiophile accolade used in all manner of audio fidelity assessments. In this report, they objectively filtered the sound to create these variations in the content and not any characteristics of audio gear.

They also committed such crimes: "The stimuli consisted of 42 sound clips, with duration between 11 and 21 s. All sounds were down-mixed to mono, to suppress any spatial effects."

We had riots in streets over mono testing of loudspeakers.

But yes, this is example of investigation in audio science. Listening tests to see if objective changes in content results in corresponding changes in listener preference. This is why I said organizations like AES define what audio research/science is.
 
it's not a simple answer to your question and want to give a complete answer. I'm traveling today to Monterey for business and will give an appropriate answer as soon as I am able.
 
It has nothing to do with pessimism. It has to do with data. Data in the form of reading hundreds of research papers and managing a large team of audio/signal processing experts while I was at Microsoft. Forum discussions are quite myopic in that regard because members don't have access to much body of research or involvement in research/engineering professionals.


Well, there should be no rejoicing by audiophiles on this work as what they tested, is the true definition of the word, microdynamics. That is, short term variations in the dynamic range of music. Not some audiophile accolade used in all manner of audio fidelity assessments. In this report, they objectively filtered the sound to create these variations in the content and not any characteristics of audio gear.

They also committed such crimes: "The stimuli consisted of 42 sound clips, with duration between 11 and 21 s. All sounds were down-mixed to mono, to suppress any spatial effects."

We had riots in streets over mono testing of loudspeakers.

But yes, this is example of investigation in audio science. Listening tests to see if objective changes in content results in corresponding changes in listener preference. This is why I said organizations like AES define what audio research/science is.

IMHO you are the one with wrong glasses in this forum. The majority of us do not care about the subjects you researched at Microsoft and you insist moving the debate to your favorite area and calling other people myopic. Most of us debate high quality stereo.

Thanks for the clarification of the paper on microdynamics - again it seems I was too optimistic. Interesting for multichannel.

And again, I do no think that just naming organizations defines what are the science areas - they have shown due to their intrinsic interest and affinities and naturally organizations form around these subjects and then define precisely what they are in their regulations.
 
But yes, this is example of investigation in audio science... This is why I said organizations like AES define what audio research/science is.

Or, T.C. Electronics does.


Screen Shot 2015-07-26 at 8.20.17 pm.png


Which is convenient, considering LRA is used in about a dozen of their broadcast and production products.


Screen Shot 2015-07-26 at 8.21.27 pm.jpg


Don't get me wrong, I'm grateful someone's putting the hard yards into research, but it's hardly investigation for investigation's sake - but then the line between science and commerce is always a thin one.
 
Last edited:
A definition that needs to use the word "etc. " is not clear. Unless you quoted or summarized the precise intentions and objective of each of these publications (they have them in their regulations and rules) you are not defining anything. Besides, some of them, e.g. IEEE Signal Processing, also publish other types of science, non related to audio.
Also claiming that science is the outcome of research seems to imply you are addressing scientific research - and then you are using the concept to be defined in the definition.

IMHO the key to clarify our debates is to identify the specificity of the science exclusive to audio reproduction, that includes perceptual sciences. We all know that audio uses electrical and mechanical science, as well as lots of physics and maths.

+1.
 
Amir has given us his definition of audio science. Perhaps we could also read Tim's or any one else's. I started this thread assuming, or taking for granted, that this was a clearly defined term about which everyone agrees. Now I am not so sure.

If there is a general consensus on the meaning of the term, then we can use it and move forward to discuss what it tells us about what we hear. If there is no such general consensus, or if the term is vague at best, as someone up thread wrote, then where do we go from here, if we can not even agree on the topic of the thread?

How about this: Audio Science is the pursuit to better understand how audio equipment corresponds to what we hear when attempting to reproduce a musical event. It consists of collecting data, taking measurements and conducting listening tests.

I'm sure this definition is incomplete, or even hopelessly wrong. But it is an attempt to get the conversation started so that we know what it is that we are trying to discuss. Could those who have a better understanding of what the term means explain it to me and to those others who may also be a bit confused?
 
I have seen too many systems installed in casual acoustics, never measured or treated, that after careful tweaking sound great.

There are untreated rooms that are o.k. enough to allow for great sound. Yet one thing is 'great sound', another is the performance of that same system under optimized room conditions. That performance may lie on a completely different, far superior level that you would never dream of in light of what you experience as great sound under the circumstances that you describe.

I for one could never have dreamt of my system being capable of the performance that I hear now in an optimized room. I have good reasons to believe that it now significantly outperforms other systems at a much higher price point when they play in a non-optimized room.

And yes, on the other end of what you describe, I have heard too many systems that sounded considerably less good than expected because the room was problematic.
 
rrbert apparently meant 33.27 and not 32.27 which must have been a typo. 33.27 is as much removed from 33.33 as 33.39, which was the other number he mentioned.

The reduced difference is about half what I can hear. (I can distinguish 440 vs. 441.5). This makes me think that 33.27 not close enough, as this represents an inadequate safety factor, IMO.

I tested 440 vs 441.5 side by side with ABX. I did not try recognizing them individually after some passage of time, as I do not claim to have perfect pitch. These were level controlled sine waves that lasted 1 second each, with a 0.1 second fade in and fade out to avoid clicks.
 
The reduced difference is about half what I can hear. (I can distinguish 440 vs. 441.5). This makes me think that 33.27 not close enough, as this represents an inadequate safety factor, IMO.

I tested 440 vs 441.5 side by side with ABX. I did not try recognizing them individually after some passage of time, as I do not claim to have perfect pitch. These were level controlled sine waves that lasted 1 second each, with a 0.1 second fade in and fade out to avoid clicks.

That is interesting, Tony. I have never tried to do this kind of a test. So you can hear a difference between the two sine waves in isolation when heard close to each other. Have you tried listening to a recording of solo piano, or say a string quartet or cello concerto and heard differences of this magnitude in pitch? I have read that a turntable that is slightly off perfect speed sounds less bad than one that slightly fluctuates in speed. These are different kinds of inaccuracies, ie. speed consistency versus speed accuracy.

I have not experimented with any of this but think it would be very interesting to learn what I can hear with slight variations in turntable speed. I'm sure this is a major area of turntable design and various designers have experience and opinions about this topic.
 
That is interesting, Tony. I have never tried to do this kind of a test. So you can hear a difference between the two sine waves in isolation when heard close to each other. Have you tried listening to a recording of solo piano, or say a string quartet or cello concerto and heard differences of this magnitude in pitch? I have read that a turntable that is slightly off perfect speed sounds less bad than one that slightly fluctuates in speed. These are different kinds of inaccuracies, ie. speed consistency versus speed accuracy.

I have not experimented with any of this but think it would be very interesting to learn what I can hear with slight variations in turntable speed. I'm sure this is a major area of turntable design and various designers have experience and opinions about this topic.

With music the pitch is constantly changing from note to note making identification of small changes much more difficult, but I would assume that there would be an effect. although it might be perceived in other terms. Some orchestras tune quite sharp as part of their image.

My late wife said she had always assumed that I had perfect pitch, because when I sang or whistled Mahler songs I always did them in the right key. She did have perfect pitch. Once as a teenager in a summer music camp she was performing a Mozart sonata. After a few bars she stopped the performance and apologized, saying that the piano was a half tone off pitch (about 6 percent). She said she would start over and transpose the sonata so the audience would hear it in the intended key, which she did with no problem. (The really amazing thing about this story was that I heard it when I was in high school from a fellow student at summer camp, years before I met her. We were married for several months before I realized I had married the girl in the story. At that point, she told me that transposing the Mozart sonata on the fly was easy, but the hard part was playing it in one key and hearing it in another.)
 
According to Wikipedia, most people (assumed by me to be untrained listeners) will hear a difference of 0.3% on typical music. I have a fair amount of experience pitch correcting old analog tape transfers, and with a little practice 0.1% difference is readily audible, which makes a Rega table (for example) with its typical pitch error of 0.5 - 1.0% completely unlistenable.

There's no doubt that speed stability is more important than pitch accuracy. Still, most tables Fremer has tested do relatively poorly here to, and most tables under about $5k can easily vary by 0.1% or more.
 
my personal experience with DTD pressings verses tape is that DTD pressing, done right, had a more delicate degree of detail and more refined top end, the tapes seemed to be meatier. soundstage I would give to the DTD in terms of space, the tape in terms of solidity of image.

Since tape and LP have nearly the same specs for speed stability, I have to assume that the difference (one that I often hear too, except on the most speed stable turntables with the best arms) that the lack of soundstage solidarity with the LP has much to do with the tone arm in playback and really isn't the fault of the medium but that of the playback apparatus.
 
I have never seen wire measurements change in an interconnect in a normal use. I have done those tests, like many others, years ago. It is not new to me. I have been at electronics for a long time. I thought I read the post, whether you call it warm up or not, after two hours at a meal it sounded so much better. If I got it wrong then sorry.

No problem, tomelex. Too many posts to read, too little time.

Now with all due respect, should anybody really care that you’ve never seen wire measurements change in an IC? I sure don’t. That’s kinda’ the topic of this thread, isn’t it?

The question is, have you heard a difference occur in an IC over the course of use? I have perhaps 25 – 30 times or essentially any time I’ve acquired a new set of IC’s to audition. And if I wanted to guarantee I hear an improvement at burn-in, I only need to send my IC’s out for cryo-treatment which essentially returns the materials / properties back to a state of unused or like as new. After install and round-the-clock playtime at a certain time allotment (almost like clockwork), they will achieve a fully burned in state that is evident by a sonic improvement almost like a light switch being turned on.

That’s just with IC’s but the same is true with essentially every other electrical part or component. Not only is the phenomena entirely repeatable, but I can also predict the approximate time required to reach full burn-in status. The simpler and smaller the part, the more accurate I can predict the time required to achieve full burn-in status. For example, wall plugs, outlets, IEC connectors and inlets, fuses, etc, the required time is roughly 53 hours plus or minus 30 minutes. IC’s of 1M length generally 5 ½ days of continuous burn-in.

Now many claim to have experienced the benefits of this “burn-in” phenomena but at the same time many have not. And depending on the product and if proper or superior cryo-treating is involved, the potential sonic improvements are most always anything but subtle.

But I’ve yet to hear of a single story where somebody attempted to measure these sonic differences. Perhaps it’s because those who possess sensitive measuring instruments don’t believe in “burn-in” so they won’t trouble themselves any further?

Of course there can be a time period for devices to stabilize, some more than others. Question is audability not some panzy wanny hyper technical theoretical super minute measuremets that confirm a change took place, but an audible change, ' interconnect Wires, no. fuses, no they get worse with heavy use, not better. speakers need to be broken in, they are mechanical devices after all. tubes change all the time.

Can be a time period for devices to stabilize? My passion is extreme forms of vibration control. For 15 years my fabulous performing Foundation Research line conditioners (LC’s) always sat on the carpet behind my system. About a year ago, I decided to see what I might be able to do to improve their sonics by applying a similar methodology that I apply to my other components, to these smallish passive LC’s that have very few internal parts.

What I fabricated was just a make-shift apparatus out of spare parts and though this apparatus somewhat follows my methodologies, the materials and construction was certainly inferior. Nevertheless, over the next few months I received no less than 35 – 40 distinct audible improvements from those line conditioners.

Considering these LC’s have no power supplies of their own (no internally-generated vibrations) with few internal parts and the execution was certainly inferior, I must admit I was rather surprised how receptive they were to my method of vibration control. Such audible improvements were derived solely from my attempts to control the mechanical vibrations at the LC’s. Considering these LC’s are 7 years old and well burned in, there should have been no opportunity remaining for electrical burn-in to have this effect.

Ears/brain interface change or adapt all the time. yes, and they are hardly calibrated and easily fooled by inputs from the eyes etc.

Sure. But again please don't say the eye/brain and brain/brain combo’s are more reliable. I’m unsure why people keep bringing this up as though it has any real significance for anything. It should go without saying. But it never does. I can only surmise that the eye is more easily trained than the ear.

Burn in is different than mechanical changes. Burn in is mostly electronics, not interconnect wires. Speaker wires can change when driven hard, and so can speaker coils, but that is not anything new.

Actually, you are incorrect. I can demonstrate that not only does electrical energy behave exactly like mechanical energy, at least when pertaining to audio, I can also demonstrate that electrical burn-in times are near identical to mechanical settling in times. And really that should be no surprise to anyone since electrical energy is just variation of mechanical energy. As is perhaps all energy and all matter.

In fact, at this juncture I’m pretty much convinced there is no such thing as an electrical “burn-in” process. It’s really just a variation of another mechanical settling in process. The fact that wire vibrates when current passes through it may substantiate my hypothesis. The only difference I can tell is that electrical “burn-in” has a memory whereas mechanical settling in has no such thing.
 
IMHO you are the one with wrong glasses in this forum. The majority of us do not care about the subjects you researched at Microsoft and you insist moving the debate to your favorite area and calling other people myopic. Most of us debate high quality stereo.
The bitterness in your posts is tiring Micro. It gets in the way of properly reading what I am posting to you. The myopic reference was factual as I explained. That is, because vast majority of research papers are locked behind membership to research organizations with yearly fees, almost no one hanging around forums has ready access to them. And even if they did, the requisite knowledge to understand the papers is not there. Good example was the paper you referenced where the content was something entirely different than what you summarized to be. As a result of this, little of vast amount of audio research makes it into forum discussions. Me explaining them doesn't make anyone stupid, or me smarter. It just happens that it is my business to pay the dues, and read and understand such material. Just as it is for a doctor to read and understand medical journals that I would not have access to or understanding to comprehend. I don't get insulted if they told me this.

Thanks for the clarification of the paper on microdynamics - again it seems I was too optimistic. Interesting for multichannel.
Again, it has nothing to do with optimism or pessimism. You can't run off with the headline with of a paper. You need to read the article and then comment on it as appropriate.

And again, I do no think that just naming organizations defines what are the science areas - they have shown due to their intrinsic interest and affinities and naturally organizations form around these subjects and then define precisely what they are in their regulations.
I gave the most actionable definition of the term. No one has come and said how in their respected field of science, they think differently. Yet somehow here, there is dire resistance to the same organizations in audio.

Tell me this. Do you think researchers in audio are any less smart, educated, or lacking common sense than other science fields?
 
IMHO you are the one with wrong glasses in this forum. The majority of us do not care about the subjects you researched at Microsoft and you insist moving the debate to your favorite area and calling other people myopic. Most of us debate high quality stereo.
Addressing the Microsoft part, some of the senior researchers on my team at Microsoft are/were on the peer review board of the audio journals I mentioned. And of course wrote and published many papers themselves. Here is one example: http://www.aes.org/member/profile.cfm?ID=1800973364

"James Johnston

Member since: 1978
AES Position: Governor
AES Committees: Board of Governors, Associate Technical Editors, Journal Reviewers
Technical Committees: Audio for Games, Coding of Audio Signals, High Resolution Audio, Semantic Audio Analysis, Signal Processing, Spatial Audio, Technical Council, Transmission and Broadcasting
Standards Committee: SC-02-01 (Digital Audio Measurement Techniques)

Audio Field: Basic Research - Acoustics, Psychoacoustics and DSP"


So in the context of what is accepted norm for audio science in those organizations is very much related to my experience at Microsoft, having had the good fortune of managing some of these luminaries.
 
The bitterness in your posts is tiring Micro. It gets in the way of properly reading what I am posting to you. The myopic reference was factual as I explained. That is, because vast majority of research papers are locked behind membership to research organizations with yearly fees, almost no one hanging around forums has ready access to them. And even if they did, the requisite knowledge to understand the papers is not there. Good example was the paper you referenced where the content was something entirely different than what you summarized to be. As a result of this, little of vast amount of audio research makes it into forum discussions. Me explaining them doesn't make anyone stupid, or me smarter. It just happens that it is my business to pay the dues, and read and understand such material. Just as it is for a doctor to read and understand medical journals that I would not have access to or understanding to comprehend. I don't get insulted if they told me this.


Again, it has nothing to do with optimism or pessimism. You can't run off with the headline with of a paper. You need to read the article and then comment on it as appropriate.


I gave the most actionable definition of the term. No one has come and said how in their respected field of science, they think differently. Yet somehow here, there is dire resistance to the same organizations in audio.

Tell me this. Do you think researchers in audio are any less smart, educated, or lacking common sense than other science fields?

As usually you end up focusing on secondary details of style and omit the important facts. You repeat your naive medical experience as if it was a messianic event and then end asking unrelated questions. We were not debating human behavior, just a definition of audio science that can be debated. Fortunately other people have now asked the same questions I did, I can move out.
 
Status
Not open for further replies.

About us

  • What’s Best Forum is THE forum for high end audio, product reviews, advice and sharing experiences on the best of everything else. This is THE place where audiophiles and audio companies discuss vintage, contemporary and new audio products, music servers, music streamers, computer audio, digital-to-analog converters, turntables, phono stages, cartridges, reel-to-reel tape machines, speakers, headphones and tube and solid-state amplification. Founded in 2010 What’s Best Forum invites intelligent and courteous people of all interests and backgrounds to describe and discuss the best of everything. From beginners to life-long hobbyists to industry professionals, we enjoy learning about new things and meeting new people, and participating in spirited debates.

Quick Navigation

User Menu