Audiophilia and its discontents

[symphara]

I've said it before and I'll probably keep saying it, I think trompe l'oreille (fools the ear) experiences from loudspeakers are mostly a white whale, something that is mostly impossible from the beginning.

The sound of an instrument in a space doesn't just come from its dynamic / harmonic content. It's also from its unique radiation pattern within the room. Loudspeaker radiation patterns are shaped differently than those of instruments. So even with infinitely perfect frequency and phase response, and whatever polar plot you may prefer, a loudspeaker generally won't sound quite like the instrument you are playing on it.

If you've ever taken a recording class, they show you where to put the mic(s) in order to capture a good sound from different instruments. Let me tell you that the radiation patterns of an acoustic guitar or cello are just not replicable with a speaker, unless you purpose-built it to have an F'd up radiation pattern like a guitar. At which point you have a guitar simulator in your room, not a normal stereo.


Like, sound comes out of the front in a couple places, plus the back and (to some extent) sides of a guitar. Capture all of those with several mics, fine. In order to (in theory reliably) make it sound really-real in a room, you need a speaker with transducers that fire more or less in the same directions as the guitar does. And you need to feed each transducer with the corresponding mic's signal.

I am not aware of this having been done EVER, even as an experiment, let alone via commercial stereo recording!

What we hear in recordings is what was picked up by microphones, which categorically sounds different than what you hear in a live space. If your speakers happen to make a mic sound like a live instrument, it's probably a synergistic coincidence, "broken clock is right twice a day, but in a good way" sort of thing.

The chances are maybe better if you record close, in a very dead space, mix it very dry, and play it back in a nice, live room. But this is generally considered to sound bad in terms of mixing. Not many recordings like this out there, guitar with no reverb at all.

When the natural radiation pattern is similar to that of a speaker, I think illusions are pretty achievable. So basically... close-mic'd voice and certain brass and woodwind instruments are plausible candidates. (if you play them only on one speaker at a time.) Electric guitar, totally doable since it comes through a speaker anyway. Piano, organ, orchestra, choir, full bands, drums in general... very iffy.

This is my ideological problem with trying to make recorded music sound "really real". It's a spurious illusion that probably involves some departure from true fidelity. It involves making some recordings better than they are, and by the same token, some worse. A truly accurate speaker will make recordings sound like themselves.

So I understand this is a compromise some people are willing to chase, but to me it feels like a gamble on a game I don't totally understand.

Very interesting post and I agree with much. I’ll try to explain my reasoning.

You have a performance, let’s call it P.

You record it with microphones, which is like applying a transform on it. You get R(P).

Then to mix it, to M(R(P)).

Then you encode it and put it on some media, and you get E(M(R(P))).

Then you take that home and you decode it, and you get D(E(M(R(P)))).

Then you pass it though your signal chain, getting C(D(E(M(R(P)))))).

Then you play that through your speakers, in your room K, and listening position Q, and you get S(C(D(E(M(R(P)))))), K, Q) delivered to your subjective ears.

None of these functions are homeomorphisms.

Then the orthodoxy tells us that it’s super, mega important that S is some kind of constant function on the frequency response curve. That‘s how you get the intended effect from all this, which to me would be the original P. And @DonR (and others) think it’s “religious” to think this is even questionable. I don’t think it’s religious at all. I think it’s frankly silly to even entertain it.

I agree with you in the sense that it’s impossible to get the original, we just get some approximation. Very realistic effects are incidental and not guaranteed. But if your goal is to reproduce the recording, that’s really different from my goal, which is to trick me - as much as possible - that I’m listening to an actual performance. That’s the enjoyable thing (to me) about liking sound, as opposed to liking music, which doesn’t need particularly good reproduction at all to be enjoyed.

 

[Kal Rubinson]

I went to audition the Reference 5 literally two weeks ago and not only they were unable to produce this (for me) in an hour of changing tracks, but I didn’t think they’re capable of it*. Perhaps the Meta, but at this stage I’m fully and utterly done with KEF and their sound.

But then again I’ve been in other situations where I’m presented speakers others think are great and I just hear two boxes making noise.

(*) Another example of ”it”: I have this old Jean-Michele Jarre disc called “Aero”, in DTS. There‘s a track that starts with a bird that flies around the room, and it stays in the corners for a while. On my speakers it just sounds like the bird is physically in the speaker, struggling to get out. It’s quite startling.

There are so many variables from the recording source to the equipment to the room/setup to the ears/expectations of the listener that failure is more likely than success. So, I will not question your experience but I will restate that both Revels and KEFs are capable of that illusion in my room.

 

[birdog1960]

I've said it before and I'll probably keep saying it, I think trompe l'oreille (fools the ear) experiences from loudspeakers are mostly a white whale, something that is mostly impossible from the beginning.

The sound of an instrument in a space doesn't just come from its dynamic / harmonic content. It's also from its unique radiation pattern within the room. Loudspeaker radiation patterns are shaped differently than those of instruments. So even with infinitely perfect frequency and phase response, and whatever polar plot you may prefer, a loudspeaker generally won't sound quite like the instrument you are playing on it.

If you've ever taken a recording class, they show you where to put the mic(s) in order to capture a good sound from different instruments. Let me tell you that the radiation patterns of an acoustic guitar or cello are just not replicable with a speaker, unless you purpose-built it to have an F'd up radiation pattern like a guitar. At which point you have a guitar simulator in your room, not a normal stereo.


Like, sound comes out of the front in a couple places, plus the back and (to some extent) sides of a guitar. Capture all of those with several mics, fine. In order to (in theory reliably) make it sound really-real in a room, you need a speaker with transducers that fire more or less in the same directions as the guitar does. And you need to feed each transducer with the corresponding mic's signal.

I am not aware of this having been done EVER, even as an experiment, let alone via commercial stereo recording!

What we hear in recordings is what was picked up by microphones, which categorically sounds different than what you hear in a live space. If your speakers happen to make a mic sound like a live instrument, it's probably a synergistic coincidence, "broken clock is right twice a day, but in a good way" sort of thing.

The chances are maybe better if you record close, in a very dead space, mix it very dry, and play it back in a nice, live room. But this is generally considered to sound bad in terms of mixing. Not many recordings like this out there, guitar with no reverb at all.

When the natural radiation pattern is similar to that of a speaker, I think illusions are pretty achievable. So basically... close-mic'd voice and certain brass and woodwind instruments are plausible candidates. (if you play them only on one speaker at a time.) Electric guitar, totally doable since it comes through a speaker anyway. Piano, organ, orchestra, choir, full bands, drums in general... very iffy.

This is my ideological problem with trying to make recorded music sound "really real". It's a spurious illusion that probably involves some departure from true fidelity. It involves making some recordings better than they are, and by the same token, some worse. A truly accurate speaker will make recordings sound like themselves.

So I understand this is a compromise some people are willing to chase, but to me it feels like a gamble on a game I don't totally understand.

Then how do you explain that expert guitarists reproducibly identified specific tone woods from different recorded guitars in an anechoic room?

 

[MattHooper]

A lot of the writing and vigorous, boarding on vehement, defence of audiophilia seems to have many cult-like and quasi-religious overtones. I suspect this is part of human nature we will have a hard time correcting it if we ever can.

Or, possibly, you have misunderstood (or not cared to understand) what has been written.

Hard to know for sure from a quick dismissive response.

 

[kemmler3D]

I agree with you in the sense that it’s impossible to get the original, we just get some approximation. Very realistic effects are incidental and not guaranteed. But if your goal is to reproduce the recording, that’s really different from my goal, which is to trick me - as much as possible - that I’m listening to an actual performance. That’s the enjoyable thing (to me) about liking sound, as opposed to liking music, which doesn’t need particularly good reproduction at all to be enjoyed.

I think that's a valid goal. And I really like your explanation. I would just add that at least E( ), D( ) and C( ) should have negligible effects. I tend to agree that S( ) should be as linear as possible to maximize the chances of getting something substantially similar to M(R(P)) out of the speaker cones. But given the lack of correspondence between a mic and an ear, and the spaces things are recorded and then played back in, I simply don't have much hope about being genuinely fooled on a consistent basis.

We don't have a particularly sophisticated theory on how to recover P from S(C(D(E(M(R(P)))))), K, Q) except to spend a shitload on speakers and room treatment and hope for the best, which still totally ignores R( ), after all... my goal is just to get M(R(P)) as accurately as I can, and if it produces an illusion, that's a nice bonus. But to me "the real thing" is ultimately the mix as delivered, not a simulation of the instruments in the mix.

Then how do you explain that expert guitarists reproducibly identified specific tone woods from different recorded guitars in an anechoic room?

I don't think you need to believe you're hearing an actual guitar, to hear subtle differences between different guitars on a recording.

I'm not saying speakers can't reproduce the sound of a guitar well and clearly, I'm just saying there is no reason to expect that a speaker would sound indistinguishable from a guitar. Even if we have an infinitely linear speaker, we also have to contend with microphones, dispersion patterns, and rooms.

 

[syn08]

None of these functions are homeomorphisms.

Appreciate the metaphor (I guess you mean the homeomorphism definition of having a continuous inverse function), but otherwise, at a closer examination, your metaphor quickly falls flat. The definition of a homeomorphism involves topological spaces, which, in turn, assume a metric for defining a "distance". Which comes in direct collision with the subjective approach, where any "metric" is outright rejected "S(C(D(E(M(R(P)))))), K, Q) delivered to your subjective ears". So the functions you described are not homeomorphisms, not because the lack of a continuous inverse functions, but because of missing spaces (with a metric) they should be defined on.

Sorry for nitpicking, but I find very annoying involving any STEM concepts (other than perhaps rigorous statistics) in subjective evaluations.

 

[Cbdb2]

Well the "audiophile BS" speaker clearly made him extremely happy. He even described his feelings as "reverent" upon hearing music through his new speakers.
How much "happier" do you need him to be?

"Good" is subjective, remember.



Dunno. Odds are I suppose that in blind testing he would have picked something that measure like the Revel speakers.

That doesn't mean that the sound he heard in his new Klipsch speakers wasn't an upgrade to what he'd been used to, and didn't provide him with the thrills he describes.

We all, likely, own all sorts of substandard choices by the lights of an educated enthusiast. Any TV geek can tell me "You bought the wrong TV" and any computer geek "there are better computers than the one you use, you know!" and any smart phone geek "Android is better than your apple phone" and on and on. That doesn't mean my purchases were crap, worthless or that I shouldn't be enjoying them as I do. Whether the author would have picked Revel speakers over the Klipsch under blind conditions doesn't mean the Klipsch aren't providing him with thrills, using speakers under his real-world listening conditions.

Theres people just as happy about there new power cord because some revue told them how great it was.. So what? If good is subjective than ASR means nothing.

 

[symphara]

Appreciate the metaphor (I guess you mean the homeomorphism definition of having a continuous inverse function), but otherwise, at a closer examination, your metaphor quickly falls flat. The definition of a homeomorphism involves topological spaces, which, in turn, assume a metric for defining a "distance". Which comes in direct collision with the subjective approach, where any "metric" is outright rejected "S(C(D(E(M(R(P)))))), K, Q) delivered to your subjective ears". So the functions you described are not homeomorphisms, not because the lack of a continuous inverse functions, but because of missing spaces (with a metric) they should be defined on.

Sorry for nitpicking, but I find very annoying involving any STEM concepts (other than perhaps rigorous statistics) in subjective evaluations.

Talk about not seeing the forrest for the trees!

I don't "reject" the metric, outright or not, at all. Of course it would make no sense to say those things without a metric. The final step (the subjectivity) just adds that extra uncertainty to the process.

Just jump for a second to TV screens. We can measure everything really, really well, all the way to it being delivered in our rooms. Then the two of us look at it, I say "it looks ok" and you say "nah it's too dim, I really want that highlight to punch me in the face, that's what HDR was supposed to deliver". This doesn't invalidate the process that took you there.

 

[kemmler3D]

If good is subjective than ASR means nothing.

Good is always going to be subjective because it's inherent in the concept - it's a statement of opinion. Inanimate objects do not have a concept of "good", including an Audio Precision box or a Klippel scanner. "good" does not come in units, it can't be measured, you can't observe it or record it.

ASR is very meaningful because objectively high ("good") performance / fidelity is something that a lot of people agree is also subjectively good.

However, there's no law that says you have to agree. There are people who think tape cassettes played on $5 headphones are "good" specifically BECAUSE they have low fidelity. To them, that IS good, and they're not wrong, they just don't value fidelity. You may find that silly, but de gustibus non est disputandum.

Before you argue, consider the fact that there are plenty of people who pay good money (more than a good DAC costs, maybe!) to be kicked in the family jewels by a dominatrix. To me, that does not sound good either. It sounds a little worse than the prospect of $5 headphones. But there are lots of opinions out there, and they're all wrong to someone.

 

[symphara]

If good is subjective than ASR means nothing.

This is a non sequitur. Good here is subjective, but it doesn't follow at all that ASR means nothing.

I look at it as a scientific, technical evaluation of wine.

Physical properties - I imagine @amirm sending about 3 lasers of different wavelengths through the Petri dish and looking up the result with his "Blippel Near-Field Spectrometer". Ditto for density, viscosity etc.

Then follows data about chemical composition, alcohol content, tannin, how much dead rat in there etc.

Then he tastes and gives his opinion.

The collection of measurements would be highly useful and would tell you particularly if it's something to avoid (e.g. too much dead rat) or if it's something that might be up your alley. For example, if you like dry wines but the one tested has 10% sugar you'll probably give it a very wide berth.

What it won't tell you is if the wine will taste good, if you'll really like it, or if you'll think it's worth the price. Just the sensation of having the alcohol integrated in the drink (or not) is very complex and hard to predict, for the same %.

 

[killdozzer]

It’s like “good wine”. We could probably find criteria for objectively good wine but I think opinions would vary just as widely.

There is no criteria for objectively good wine and we couldn't find one. These comparisons are futile. There's no "blueprint" (let's call it that for the sake of our chit-chat) for wine. Wine is not a reproduction. One would think this is obvious. There is a goal audio gear tries to achieve. There is none in wine.

I look at it as a scientific, technical evaluation of wine.

I'm sure you do and therein lies the entire problem. There's no such thing. The only evaluation of wine that science does is quantifying certain compounds and substances and alert about possible existence of dangerous types of alcohol. As far as science is concerned (we shouldn't even call it that, just lab analysis), every wine that doesn't harm people (more than alcohol does in the first place) is a complete success.

I hope you expected that saying such a silly thing like KEF fails to be believable would draw some attention to you. If you say something like that, the only safe bet is that it is completely up to you (in your mind) and has nothing to do with the actual speaker.

Back on topic; if I direct this gaze towards myself, I'd say that what I see as "audiophile" about me is my need for the visual aesthetic in audio gear. Subjective as it may be, this also asks for money while doesn't really affect listening. But I can't move past this. I can never disregard visual side of it all. I would never, for the life of me, own anything made by Vivid Audio.

Anyway, @MickeyBoy nice observations. You pin-pointed all the silliness, especially this new fad of using "undesirable" ways of describing "desirable", like; so good I could vomit, or so believable like I was drunk... The older version of this is "sick". And even in this thread someone said he wants to be startled by speakers. No problem, crank it up and drop a needle. There. Startled. Or I can unplug an interconnect and plug it again while volume is turned up! Startled.

 

[symphara]

As far as science is concerned (we shouldn't even call it that, just lab analysis), every wine that doesn't harm people (more than alcohol does in the first place) is a complete success.

Sheer nonsense.

I hope you expected that saying such a silly thing like KEF fails to be believable would draw some attention to you. If you say something like that, the only safe bet is that it is completely up to you (in your mind) and has nothing to do with the actual speaker.

Nothing to do with drawing attention. They just aren't believable. To me, of course.

I actually own two KEF speakers (Ci200QR, ceiling mounted). Major disappointment. They're really mediocre. Tubby sounding. I thought the famed dispersion would create a wide listening spot, given they're at 2.65m high, but not really. Waste of money, and weren't cheap either.

 

[Vacceo]

“Good sound” is subjective. It just means “sound I like”.

It’s like “good wine”. We could probably find criteria for objectively good wine but I think opinions would vary just as widely.

I went to buy KEFs, twice. R7/R11 first time, Reference 5 the second time. Because they measure well and are praised on ASR. Waste of time. I don’t care anymore that they measure well, I’m done with KEF. They don’t offer believable sound, as in, I would never listen to a KEF speaker and think that an instrument or person is almost there in the room. Perhaps this is what they mean by "believable corporeality”.

They do not distort too much what is in the recorded material, which is the best you can get out of reproduction.

That does not mean you have to like the result or that other interactions (the room in particular) may not alter the sound.

I was watching one of Guttenberg & Reichert's chat sessions and they started talking about realism, and they said their high end systems never sounded exactly like the real thing. Steve said that if he was walking down the street and heard the sound of a piano or guitar from a window he would know instantly if it was real or a recording, whatever speakers were involved. If he was blindfold in a room he could tell from one note if it was piano or a speaker, whatever $100 000 high end system is playing. So isn't this quest for realism a chimera?

I don't believe a word from Guttenberg. Not because he's insincere or a fraud, simply because his conclusions are only valid to himself.

 

[killdozzer]

Sheer nonsense.

Not only is it not sheer, but not nonsense at all. Been working in a wine lab for quite a few years, so please, curb your enthusiasm. If a winemaker asks for advice what to do next, we tell him according to what we measure and according to what he wants to make. If not, all we look for is "heads" and "tails" in a wine. If there are none, it's a success. I'd expect your views on wine to be along the lines of "to me" same as in speakers. Which is useless (to the rest).

"They" /.../ "To me"

Let me just focus on these two. They shouldn't be mixed. You're not talking about speakers, but your impressions. The realm of impressions is in one's brain, not inside a box he buys. Given your "brand name dropping", I'd say it's the real audiophile talking. I heard such bad things about Triangles, again, it says nothing about the speakers. (And don't worry, I didn't believe those either).

If you can get "what goes in - comes out" formula for any piece of gear, than "believability" shouldn't come into question.

But you're free to start talking about all the things we can't measure whenever you feel like and save us all some time.

Of course, the fact that you don't like them is as welcome as any other taste, but you're trying to poor it over into a claim about the technical performance of a piece of gear. That's the only part I'm discussing. My guess is that the main reason for this is the fact that you get impressed by exotic. I may be wrong, of course.

 

[fpitas]

Sasha and his friends can be just as subjective as they want. My problem with the article is the smug undertones of "those quaint outsider audiophiles. Aren't they something else?" Of course that's a common trope for that sort of article.

 

[Mal]

Both KEFs and Revels that I have heard can do this. (I have no comment or knowledge of the Triangle Celius.)

With my KEF LSXs, just recently, I was watching "Snowfall" and heard what I thought was dubious noises in my street, then, after some shots rang out, I realised, with relief, it was coming from the TV! But I wasn't concentrating that hard, am not an expert listener, imperfect hearing, etc. Would it sound real to, supposedly, expert listeners like Herb or Steve?

 

[birdog1960]

Not only is it not sheer, but not nonsense at all. Been working in a wine lab for quite a few years, so please, curb your enthusiasm. If a winemaker asks for advice what to do next, we tell him according to what we measure and according to what he wants to make. If not, all we look for is "heads" and "tails" in a wine. If there are none, it's a success. I'd expect your views on wine to be along the lines of "to me" same as in speakers. Which is useless (to the rest).


Let me just focus on these two. They shouldn't be mixed. You're not talking about speakers, but your impressions. The realm of impressions is in one's brain, not inside a box he buys. Given your "brand name dropping", I'd say it's the real audiophile talking. I heard such bad things about Triangles, again, it says nothing about the speakers. (And don't worry, I didn't believe those either).

If you can get "what goes in - comes out" formula for any piece of gear, than "believability" shouldn't come into question.

But you're free to start talking about all the things we can't measure whenever you feel like and save us all some time.

Of course, the fact that you don't like them is as welcome as any other taste, but you're trying to poor it over into a claim about the technical performance of a piece of gear. That's the only part I'm discussing. My guess is that the main reason for this is the fact that you get impressed by exotic. I may be wrong, of course.

and this brings us to the subjective part. My brother in law is a Master of Wine. One Christmas, he set up a blind tasting of different types (Cab, Burgundy, Pinot, Rioja etc) divided by white and red. He had printed a description of flavors and characteristics for each type and we were to identify each in small groups. All of us like wine to varying degrees but some felt they were experts. No group came close to identifying even a majority of the wines. But man, was it fun and delicious! A friend did a similar bring a bottle event for whiskey. Same results. Great party!

 

[symphara]

Not only is it not sheer, but not nonsense at all.

In retrospect, I was perhaps too quick to jump the gun. Reading your statement again, which was "every wine that doesn't harm people (more than alcohol does in the first place) is a complete success", I'd say it could make sense to an alcoholic. I didn't think of that angle before.

Of course, the fact that you don't like them is as welcome as any other taste, but you're trying to poor it over into a claim about the technical performance of a piece of gear.

I have no idea what you're talking about. I'm trying to "poor" it? Over what technical claim??

 

[kemmler3D]

Speakers vs. Wine - an OK comparison.

With wine, you could do a full chemical assay and with enough blind taste tests, correlate the ratios of each component of wine to general preference. Some correlation of preference and measurements seems possible. However, with wine, this is a waste of time because taste preferences are much more variable than audio preferences, and there are far more variables (hundreds of chemicals per bottle, I'd guess?).

The best you can probably do is test for compounds that are always or almost always unwanted, i.e. find the headless panthers.


With speakers, we have relatively few variables (at least not hundreds like you might with wine) and we have a small number that correlate very strongly with preference. And audio preference isn't nearly as variable as it is with wine.

As long as you can quantify subjective impressions among groups of people, you can also correlate them with objective measurements, and therefore quantify what "good" is, in a general sense. But at the individual level, "good" is always and in all things a matter of taste.

 

[symphara]

Speakers vs. Wine - an OK comparison.

With wine, you could do a full chemical assay and with enough blind taste tests, correlate the ratios of each component of wine to general preference. Some correlation of preference and measurements seems possible. However, with wine, this is a waste of time because taste preferences are much more variable than audio preferences, and there are far more variables (hundreds of chemicals per bottle, I'd guess?).

The best you can probably do is test for compounds that are always or almost always unwanted, i.e. find the headless panthers.


With speakers, we have relatively few variables (at least not hundreds like you might with wine) and we have a small number that correlate very strongly with preference. And audio preference isn't nearly as variable as it is with wine.

As long as you can quantify subjective impressions among groups of people, you can also correlate them with objective measurements, and therefore quantify what "good" is, in a general sense. But at the individual level, "good" is always and in all things a matter of taste.

Far from me to claim this is an exact comparison. I just tried to explain how I see the audio measurements thing. Useful to look for the bad stuff and especially where the limits of equipment are.

Looking at the thread topic again, and on the subject of hyper-realistic sound, I had forgotten about electrostatics. I got a few demos over the years and they’re, I think, among the best at the trick. With the major drawback (for me at least) of a very tight listening spot, with major coloration if you move, and not to mention bizarre/ugly looks. Again, my opinion, least someone jumps to bite my head off for making “technical claims”.

But every time I was subjected to electrostatics I wished regular speakers could do a bit more of “that”, the sensation of depth and, for lack of better words, floating sound.