Why is audio objectivism so frequently focused on all the wrong things?

[MediumRare]

@Mikey We come to drink the wine but end up obsessing about the glass.

+100

 

[GrimSurfer]

At least you agree that not all the units coming off production line meet the specs, and we are talking about the spec numbers. You can only test the time to failure on a batch but not actual unit going to consumers and QC is mostly not time to failure

Stop being manipulative with other people's words. This is what I said:

I was talking about the blind testing for speakers, not the production tolerances of electronic equipment (the latter which are pretty damned tight today).

And then you misrepresent Toole:

Here lies our fundamental disagreement. To me, as stated by Toole in the youtube video posted above, the goal of Hi-Fi is to reproduce the experience of a musical performance. By looking at a small set of electronic measurements as the gold standard, the big picture is lost.

Which Amir corrected you on:

That is a goal that cannot be achieved with recorded music. All you can do is to be faithful to the recording. Dr. Toole says that right at the start:

View attachment 37137

See, it says reconstructing a captured performance. It says nothing about what came before that, i.e the musical performance as you say.

You're either being a troll or are such a strong subjectivist that what people write actually gets mistranslated between your eyes and brain.

 

[MediumRare]

What!? Don't you understand yet that the phenomenological subject takes precedence over veracity? Back to Plato's Cave with you!

You've actually touched on a fundamental issue here. Some people (Plato) insist that what is right before their eyes is not the truth. That there must be something more important that's hidden, that only an expert, a philosopher, for example, can detect. Some unobtainable truth that one must strive for but never can quite reach (but maybe get farther than YOU). If one is fundamentally in this camp then someone else creating clarity with a fact-pattern and the concept of "transparency" short of "perfection" is unacceptable because it invalidates one's entire worldview.

 

[Pluto]

Serious question: Aren't microphones at the professional level "solved"?

This is largely so. However, there is no one answer to the question of the ‘perfect’ microphone because they vary greatly in their characteristics – not in the rather random way that typical audiopiles perceive differences between pieces of equipment but for perfectly rational reasons, well-understood by practitioners.

By way of example: for quite simple reasons, many of the cleanest, flattest (i.e. “best”) microphones are omnidirectional. But that very fact renders them unsuitable for a great many applications. Contrariwise, when mics. of particular characteristics need to be employed (such as highly directional types needed to pick up dialogue on a sound stage), the physics dictates that such a mic. cannot be as flat an an omni. And so it goes on…

 

[Julf]

You've actually touched on a fundamental issue here. Some people (Plato) insist that what is right before their eyes is not the truth. That there must be something more important that's hidden, that only an expert, a philosopher, for example, can detect. Some unobtainable truth that one must strive for but never can quite reach (but maybe get farther than YOU). If one is fundamentally in this camp then someone else creating clarity with a fact-pattern and the concept of "transparency" short of "perfection" is unacceptable because it invalidates one's entire worldview.

I didn't spend enough time reading Plato to know how he dealt with stuff like the Pythagorean Theorem. How can it be more perfect?

 

[Julf]

This is largely so. However, there is no one answer to the question of the ‘perfect’ microphone because they vary greatly in their characteristics – not in the rather random way that typical audiopiles perceive differences between pieces of equipment, but for perfectly rational reasons, well-understood by practitioners.

And again, mics are tools for creating a piece of art - a recording. That recording is then reproduced (more or less accurately) by our home gear. Just because the paintbrush Rembrandt used (and the light in his studio) wasn't perfect doesn't mean it is fine to wear tinted sunglasses when visiting the Rijksmuseum.

 

[pkane]

I didn't spend enough time reading Plato to know how he dealt with stuff like the Pythagorean Theorem. How can it be more perfect?

Pythagorean theorem is not all that perfect: it's broken for non-euclidean spaces

 

[FrantzM]

Hi

I have taken the time to read the OP. My take:

There is one stated goal of High Fidelity: Reproducing the signal in the source as faithfully as possible. We are at a technological juncture where we need to convert/transform electrical signals into acoustics ones for us humans to enjoy the reproduction. We are not yet at the level of a direct interface that would bypass our fallible senses and just present the signal to our brains ... even then, such neural interfaces shall vary in quality ...

So we are left with several stages of transformations:
1) Convert the acoustic event into electrical signal
2) Convert the resulting signal (which has losses due to conversion) to acoustical signals.

Let's focus on 2 for now.

The signal must be converted to acoustical signals. For that we have a chain of conversion/transformation.:
We have supports (Edison Wax cylinders, Shellac disks, Tape, Vinyl LB, Cassette, 8-Track cartridge, ELCaset, DAT, CD, digital files). I suppose my listing to be chronological and I may have missed some supports .. still .. There has been progression n the quality of the supports. Digital files can be as close as technologically feasible to store the electrical signal. We extract the signal from the supports and present them to the transducers. It seems common sense to keep this chain as faithful as possible. As linear as possible , just add the required level of amplification to the transducers.
Let's , now analyze our perception. It seems that there is a threshold of perception for some things.. We can't hear the differences of 0.01 dB at any frequency for exemple. Now, allow me a truism : A chain is as strong as its weakest link.. it bears to reason to make sure that all the links are as strong as possible.. Yes? Ok.... Let's analyze a basic 2-channel digital system.
DAC
Preamp
Amp
Speakers/transducers

In this sense we would want a DAC with the least amount of signal deterioration, same for the preamp and same for the amp... We leave the transducers, out, for now. Notice that the deterioration are cumulative/additive... Little bit of deterioration in the DAC is amplified by the preamp which adds its bits of deterioration and send it to the amp which will do the same to the original signal ... It does thus make sense to have a chain of links that are as strong as technologically feasible, repeating myself here... This could become an aesthetic exercise: We aim for the least amount of corruption/deterioration even if we're not sure our fallible ears will perceive it ... Just in case. We go beyond our threshold and there is satisfaction in it, for some here. For me in particular. The beauty of this , thanks to technology is that this increase in fidelity/faithfulness doesn't always come with an increase in price. We have $9 dongle that are challenging $15,000 hand-made Digital to Analog Converters (We think better not to just call this particular contraption: a "DAC" ) and amplification modules that, at circa $1000, surpass amplifers costing more than $100,000.
Thus on the electronics side, we are at a number race. We try to do as well as we can.. threshold of hearing be damned .... We do it because , we find beauty in it. It induces a sense of satisfaction. It may not be entirely necessary but ... In that sense some , arbitrary level of performance are defined... we may not be able to her past 0.1% or 80 dB SINAD but when a $9 item has a SINAD superior to that of a $15,000 model ...

We know by now what to measure to insure ourselves of the transparency or quality of our electronics ... Some will come up with lame excuses as music is not a sine then we come up with multi tone measurements and those show us what our chain does to the signal , even music and we find out that these days with really good electronics? not much .. We have determined that electronics are now at a level where they are mostly indistinguishable from each other (when competently built)... We then turn our head toward other elements of values: Looks, performance, reliability, SINAD, THD, etc. Nothing wrong with that IMO. Healthy... compared that to people comparing Ethernet cables and coming up with descriptives such as the midrange "bloom" of an Ethernet cable...

Then we come to the transducers and there it is still a delicate field to navigate...

When we can deem competently built electronics are virtually transparent... it is not so with speakers/headphones. We know more or less what to measure but a transducer performs in an environment and that is as diverse as human personalities... with headphones for exemple even the shape of one's head can have an impact on the reproduction, presentation of the signal.. When in a room// things go even more complicated/ Not that these effects are not measurable, they are but more difficult than electronics. These are nonetheless sorely needed and some metrics give us a starting point... a speaker with a dropping FR and that does 40 Hz at -20 dB down with respect to a 1 KHz reference is likely bass-shy... The FR on-axis tells us a lot but not all.. same with directivity .. Power compression is another metric we need to see more of it ... distortion in transducers vary with the level ... And that is where our perceptions starts making things more difficult: Signals of different levels mauy mask each other in ways that they are not perceived correctly .. High level of distortion in the bass may notbe perceived well... THe room modes can play a role in our perception of the music.. The reverberation time of the room in which the speakers is placed .. Yet we tend to hear past these.. We need to continue to measure to know why speakers sound the way they do. We need to know more, we need to know why such speaker is preferred to another, what make a speaker good, preferred ...
There has been many studies on the subject. They tend to remain in arcana or academia. Few groups have done what Harman has, they are quite popular here at ASR but other groups have made some interesting studies , among these surprisingly, Bang and Olufsen, I will maintain that some of their speakers are superb and I find them subjectively beautiful to behold (e.g

). I find the Beolab 5 incredibly good sounding to the extent that now they figure in my search for an endgame system ... I dare think the 50's would be superior...) . Studies on loudspeakers tend to be more complex than that of mere electronics... Yet there are things that come up regularly.. One of these is that smooth bass and sufficient bass is preferred by most people and that across geographies, gender and age... Of corse that suppose that other parameters are removed ... Same people with their baises would score things differently...

At the end blind test is the ultimate for transducers. They allow us to remove our baises. Blind tests are difficult to perform . For electronics, we have a good set of measurements that we know insure transparency or at the very least adequacy. We are looking for the same with transducers. We have some they are necessary but not yet ( IMO) sufficient.

ASR is a Science oriented forum as as such tends to favor higher measurable , repeatable performance. A scientific bias if there was ever one. A good bias to have IMO. an objective bias

Peace

 

[Ron Texas]

The objective is to have a musically satisfying experience. If something increases your satisfaction it's good, assuming it's within your budget. So real questions like does a Hypex amp with it's great measured performance sound any better than a Crown Drive Core with their significantly worse measured performance, with a given pair of speakers, or many pairs, need to be answered through blind testing with a meaningful sample size. Toole and Olive spent their careers on blind testing. I have no idea what it cost to develop the body of knowledge which is their legacy.

 

[Island_Kenny]

Stop being manipulative with other people's words. This is what I said:



And then you misrepresent Toole:



Which Amir corrected you on:



You're either being a troll or are such a strong subjectivist that what people write actually gets mistranslated between your eyes and brain.

chill ... and i don't think that you need to resort to personal attack

Toole emphasized on "listeners", and that's OP's main point. The judge should be the listeners, not the analyzers. That's his position, and mine as well. The goal of sound recording is to reproduce a performance for listeners to enjoy. No matter what you read, I believe this is what Toole said. And I also pointed that even Toole's definition about sound reproduction is too narrow, as sound production can be created without recording sound, and reproducing a sound recording can be achieved without speakers/headphones, and could result in far superior listener experience in certain settings. But his blind tests is not limited to the sound reproduction by his definition.

 

[murraycamp]

Perhaps your existential angst would be better directed at the IEEE or AES for promoting those goals, not Amir or this site.

In essence we focus on the equipment so that in use it disappears -- the best sounding gear should have no "sound" at all.

Word.

 

[GrimSurfer]

chill ...

Toole emphasized on "listeners", and that's OP's main point. The judge should be the listeners, not the analyzers. That's his position, and mine as well. The goal of sound recording is to reproduce a performance for listeners to enjoy. No matter what you read, I believe this is what Toole said. And I also pointed that even Toole's definition about sound reproduction is too narrow, as sound production can be created without recording sound, and without speakers, and could result in far superior listener experience in certain settings.

There's a difference between recording and playback that you still don't understand.

 

[Cosmik]

The objective is to have a musically satisfying experience. If something increases your satisfaction it's good

What is the difference between swapping between pieces of playback equipment that have sonic differences, and mixing your own version of the recording? None. So why not do it properly and feed it into some studio equipment that really allows you to tailor it to your own satisfaction? Distortion, EQ, delay, etc.

Audiophiles call deliberately nonlinear pieces of equipment "amplifiers" and "DACs" and "speakers" rather than "effects units". They are in denial about what they really are, preferring instead to think of them as better or worse amplifiers etc. If their bag is distortion and EQ and so on, they would have more satisfaction on tap if they used plain, simple linear equipment supplemented with their own 'mastering effects'.

 

[Island_Kenny]

There's a difference between recording and playback that you still don't understand.

would you please tell me the differences? I don't see how this can be so complicated.

 

[murraycamp]

Design by preference is a road to nowhere, and only by attempting to adhere to rigid definitions and specifications can a system be called a serious audio system that will reveal to all listeners things they never expected to hear.

Well-stated.

 

[Ron Texas]

What is the difference between swapping between pieces of playback equipment that have sonic differences, and mixing your own version of the recording? None. So why not do it properly and feed it into some studio equipment that really allows you to tailor it to your own satisfaction? Distortion, EQ, delay, etc.

Audiophiles call deliberately nonlinear pieces of equipment "amplifiers" and "DACs" and "speakers" rather than "effects units". They are in denial about what they really are, preferring instead to think of them as better or worse amplifiers etc. If their bag is distortion and EQ and so on, they would have more satisfaction on tap if they used plain, simple linear equipment supplemented with their own 'mastering effects'.

Well, I find the "fidelity to the original" concept to be like nailing jelly to a tree. The research by Toole and Olive was about measuring statistically significant preferences, usually around 65%. There really is no rational measure. If a statistically significant group of listeners prefer a Yaggi with its nonlinear performance to a $90 Topping D10 that would make a Yaggi the preferred choice. Lots of paid reviewers loved it. It might have been the emperor's new clothes for all I know. I wouldn't buy one. Some of the measurements I have seen for Nelson Pass amps aren't so great, yet they are revered.

To a point buying electronics which measure well is safe. More work has to be done on where that point is.

We are all different. Over the weekend I went out with a relative who ordered her $49 (plus another $25% for tax and tip) steak well done. I have dined with people who ordered their food without any sauces in expensive restaurants. That's equivalent to paying $80 for a $10 blue plate, but it's what she wanted. There is no exact goal, only statistically significant preferences.

 

[Shadrach]

DBT, it's not about preferences, or about better, or worse. It's about can you discern a difference between the items tested.
So far, in published tests, very few people have been able to discern a difference to any statistical significance.
That means essentially that the listener isn't sure which unit is performing. If you can't determine which unit is performing it seems impossible to me that anyone can state a preference.
How is amirm supposed to go about a DBT with every piece of equipment he tests? Even if he somehow managed it, would you believe his results?
I thin the OP should do some DBTs. Despite all the noise regarding how difficult it is to do them strictly, you will I believe be horrified with the results even in a leaky test.
The point being for most, they're debating differences that are not audible to most of us. Find the audible differences first, then sure make posts about preferences. I quite happy to accept people have different preferences. I'm not so happy when people write about preferences when I doubt they can tell the difference between the items they picked a preference from.

 

[Robin L]

However, there is no one answer to the question of the ‘perfect’ microphone because they vary greatly in their characteristics – not in the rather random way that typical audiopiles perceive differences between pieces of equipment but for perfectly rational reasons, well-understood by practitioners.

Wrong–there is no such thing as a "Perfect Microphone". That's a straw man. All microphones have their own sound signatures. They are transducers, they are transformers of a sort, transforming sound waves into electrical waves and adding resonances and distortion. Those "rational reasons" of which you speak break down to subjective evaluations by individuals. Objectively, the Neumann u-47 is a pandora's box of colorations. But it is desired by working engineers and producers because that collection of colorations produces hit records. Don't look for logic here, this is how sausages are made.

Like the man said, once they got you asking wrong questions, they don't have to worry about answers.

 

[tmtomh]

I appreciate @Mikey starting this thread. He poses questions that have been posed before, but he does so in a way that I think is easier and more useful to engage with than some past attempts.

That said, I think the question comes down to two basic, and fairly simple points:

1. Focusing on the "wrong things" aka listening vs equipment testing: This site measures equipment and does not conduct blind listening tests because it's far more feasible to mail Amir a DAC than it is to continually fly dozens of people to Amir's home from all over the world to conduct blind listening tests. But that doesn't mean this site is focusing on the wrong things. There is scientific literature on listener preference, psychoacoustics, the mechanics of human hearing, human hearing sensitivity and auditory memory, and so on - and @amirm and many others here routinely reference that literature and comment upon it or use it in discussions. For example, Amir routinely notes when a particular measurement is not ideal but nevertheless will be below the human hearing threshold (for example. power supply noise at lower frequencies can degrade a SINAD measurement but will not be as audible as harmonic distortion at the same level at higher frequencies).

It is only when all audio-related scientific measurement is lumped together as "objectivism," with the unrealistic implicit expectation that any objectivist site will cover all objective measurement, that this site can be seen as focusing on the "wrong" thing. So in my view it's @Mikey 's question that's flawed, not this site.

2. The value of equipment measurements: As @Cosmik and others have noted, equipment objectivism is looking for transparency, literally high fidelity to the source. The source, as Floyd Toole and everyone else who understands this properly knows, is not an original performance but rather the original recording. The notion - implicit or explicit - that an audiophile sound system is supposed to create a convincing illusion of a live performance regardless of whether or not the master tape/file could ever do that (or whether there ever was a live performance, as opposed to a multi-track mixdown of many separate performances and computer-generated sounds) is precisely what Toole calls the "circle of confusion": an inability or refusal to distinguish between the audio flaws/characteristics of a recording vs its playback/reproduction.

This mistaken understanding of the purpose of hi-fi reproduction is the source of much of the BS and confusion in audiophile discussion. IMHO all the claims that various colorations and distortion (vinyl playback, tubes, nonlinear "voicing" of speakers and other equipment, etc.) improve the sound come from this same root misunderstanding: the idea that that the flaws of equipment magically happen to change the sound in precisely the way needed to create an enhanced illusion of reality compared to transparent reproduction of the master tape/file. It's manifestly nonsense.

It then follows seemingly logically from this flawed notion that if worse-measuring equipment "sounds better," then there must be an objectivist reason: there must be some as-yet unknown factor that, if it could be identified and measured, would reveal the source of the audio magic. And by this logic, objectivism is then measuring the "wrong things" because it is not measuring the alleged factors that create the magic pixie-dust illusion of "reality."

But the Harman/Toole data clearly demonstrate than in blind listening tests conducted with listeners across the world over a period of many years, listeners prefer neutral aka transparent-as-possible transducers. This establishes a close correlation between equipment measurement and listening.

All that remains is one issue: Economy. If a $3k amp has better SINAD than a $300 amp, but the $300 amp's specs are still audibly transparent, then why bother with the $3k amp?

This is an important question, but the answer should take three factors into account:

• Because of the work this site does, we see that there is no linear or reliable correlation between price and measurements. So the risk is not about wasting money on excess specs, but rather about wasting money on gear - expensive or inexpensive - that measures so poorly in one respect that it might lose perceptual transparency.
• In his reviews, Amir regularly provides informed commentary about the audibility of various specs and measurements, which can help guide members' purchasing decisions so they don't needlessly break the bank chasing unnecessary dBs of whatever.
• Distortion and noise are cumulative in the signal chain, so a noisy but inaudible DAC connected to a noisy but inaudible amplifier might produce a result whose noise is in fact audible.

It does appear that there are somewhat hard limits on the noise floors of equipment: the baseline noise produced by electrical components, the limits of Amir's measurement equipment, and so on. From there it should not be too difficult to develop a reasonable consensus here as to what a good benchmark is for most specs. At that point it will be just a matter of seeking out equipment that meets such benchmarks.

 

[GrimSurfer]

would you please tell me the differences? I don't see how this can be so complicated.

It isn't complicated.

Use the most accurate and transparent gear. Then, and only then, will you be able to hear a recording as it has been laid down.

If you don't like the sound, then look for better recordings OR equalize to taste.

Doing it any other way is simply adding a variable to the audio chain that (1) reduces transparency, (2) reduces accuracy, and (3) in no way guarantees "better sound".

This mistaken understanding of the purpose of hi-fi reproduction is the source of much of the BS and confusion in audiophile discussion. IMHO all the claims that various colorations and distortion (vinyl playback, tubes, nonlinear "voicing" of speakers and other equipment, etc.) improve the sound come from this same root misunderstanding: the idea that that the flaws of equipment magically happen to change the sound in precisely the way needed to create an enhanced illusion of reality compared to transparent reproduction of the master tape/file. It's manifestly nonsense.

Indeed.