Message to golden-eared audiophiles posting at ASR for the first time...

[Valhalla]

#8: There are scratch marks all over your floor from constantly re-positioning your heavy floorstanding speakers.

Room acoustic sure is one of the most important variables in every stereo system that affects our hearing perception fundamentally. I take speaker placement included in it. I belive a well place mediocre loudspeaker will surpass improperly placed speakers with higher price tags.

 

[David Harper]

The thing that seems to get routinely overlooked is that speakers, in general, have far more distortion (relatively speaking) than amps, DAC's, or CDP's. Especially the primitive dynamic drivers in wooden boxes. So it follows logically that to claim to hear, for instance, amplifier distortion thru wooden box speakers is absurd given the fact that any distortion created by an amp would be eclipsed (masked) completely by the distortion of the speakers. Why is this fact not spoken of more often on audio forums? Golden eared audiophiles who claim to hear differences in sound quality between, for instance, pre-amps seem to be completely unaware of this fact. That their speakers are producing, probably, at least ten times more distortion than any electronic component. Maybe even a hundred times more.

 

[sergeauckland]

The thing that seems to get routinely overlooked is that speakers, in general, have far more distortion (relatively speaking) than amps, DAC's, or CDP's. Especially the primitive dynamic drivers in wooden boxes. So it follows logically that to claim to hear, for instance, amplifier distortion thru wooden box speakers is absurd given the fact that any distortion created by an amp would be eclipsed (masked) completely by the distortion of the speakers. Why is this fact not spoken of more often on audio forums? Golden eared audiophiles who claim to hear differences in sound quality between, for instance, pre-amps seem to be completely unaware of this fact. That their speakers are producing, probably, at least ten times more distortion than any electronic component. Maybe even a hundred times more.

I think your signature gives you an answer. 80% of audiophiles believe they have better hearing than the average audiophile.

S.

 

[Inner Space]

The thing that seems to get routinely overlooked is that speakers, in general, have far more distortion (relatively speaking) than amps, DAC's, or CDP's. Especially the primitive dynamic drivers in wooden boxes. So it follows logically that to claim to hear, for instance, amplifier distortion thru wooden box speakers is absurd given the fact that any distortion created by an amp would be eclipsed (masked) completely by the distortion of the speakers. Why is this fact not spoken of more often on audio forums? Golden eared audiophiles who claim to hear differences in sound quality between, for instance, pre-amps seem to be completely unaware of this fact. That their speakers are producing, probably, at least ten times more distortion than any electronic component. Maybe even a hundred times more.

OK, but why then would audio scientists even propose double-blind testing of electronics? Are you saying absolutely nothing except speaker artifacts are audible?

 

[watchnerd]

OK, but why then would audio scientists even propose double-blind testing of electronics? Are you saying absolutely nothing except speaker artifacts are audible?

I widen it a bit to "transducers", personally (speakers, headphones, microphones, cartridges) and the room.

Yeah, there's little point in bind testing CDPs, DACs, and amps, except to prove how hard it is to hear a difference. You can measure and predict the likely outcome with a pretty high degree of confidence.

So it's a bit of a waste of everyone's time.

Phono stages might be an exception just because so many have bad overload margins and audible noise. But that can be measured, too.

 

[Robin L]

OK, but why then would audio scientists even propose double-blind testing of electronics? Are you saying absolutely nothing except speaker artifacts are audible?

I've heard plenty of amplifier interactions with headphones, floorstanders & monkey coffins. For one thing, there are different frequency responses from different amplifiers due to various design choices, the action of tone controls, loudness compensation and so on, where generally things don't really line up when comparing one amp to another. My guess is that the effects of distortions interacting isn't necessary additive, one distortion is probably modulating another. Another guess is that more modern designs have low enough distortion, though the power supply might be more stable in one than another. Obviously if there's double-blind testing where two amps can't be sorted out, that's usually a sign of good engineering. And there's plenty of amplifiers being made these days that perform badly enough or differently enough to be easily sorted out in a DBT.

 

[Robin L]

I widen it a bit to "transducers", personally (speakers, headphones, microphones, cartridges) and the room.

Yeah, there's little point in bind testing CDPs, DACs, and amps, except to prove how hard it is to hear a difference. You can measure and predict the likely outcome with a pretty high degree of confidence.

So it's a bit of a waste of everyone's time.

Phono stages might be an exception just because so many have bad overload margins and audible noise. But that can be measured, too.

And cutting heads for the lathes that make masters for vinyl, I have no idea how distortion they provide but I'll bet it's easily measured/heard.

 

[welsh]

An example: the Elac Debut 2.0 Reference (I hope I remembered that right—the newer and more expensive version with the slot port). The subjectivists praise it and Amir was impressed by how it measured. Will it work in my room with my requirements? No. I have a use case where I play along on a tuba with an orchestra recording. Tubas are loud, and so are orchestras when on stage. A requirement flowing from that use case is the ability to play a fortissimo timpani strike at 108 dB SPL. I can do that with four 10” woofers driven by a combined 250 watts, but just barely. Will my speakers spin well? I doubt it. But instruments played through them sound like themselves, based on decades of ear training resulting from playing in the same room with real musicians. That cannot be said for many speakers out there that apparently spin well.

But I have had the itch hard to panda pair of those Elacs at some point for one of my other systems. That’s the hobby, after all.

Rick “who does not have golden ears, but more like working ears—damaged by decades of hard work but still working” Denney

There is a lovely story about an advertising agency who required job candidates to successfully apply to rent a room from a landlord - while carrying a tuba...

 

[rdenney]

There is a lovely story about an advertising agency who required job candidates to successfully apply to rent a room from a landlord - while carrying a tuba...

Off-topic warning: I have always lived in a house except during a break in tuba playing of a few years while I went to grad school. In Austin, nobody every complained, though I'd been away from the instrument for some years so God knows they had reason to complain. Roommate was also a musician--he understood. Owned my own house in San Antonio--no complaints there, either, and I played in the front room of the house. Carrollton, Texas--no complaints, but the neighbor, whose dog consistently barked all night, lacked moral standing in my view anyway. Then I moved to Dallas, and this describes what I don't like about Dallas. Neighbor complained about tuba playing, and complained that I shouldn't be trying to play the Bach that elicited the complaint. Okay, so they are music lovers. And it was late at night. And I can't play that particular 'Cello Suite on a tuba without doing real Baroque damage.

When I moved to Virginia, the neighbors took it to the next level. The neighbor across the street complained. Baby trying to sleep and they want to leave the windows open, even at 8:30 PM. What? When are high-school band kids supposed to practice? So, I moved to the basement (something I never had in Texas). Nope, not good enough. Wife would send Husband across the street, who would mumble his request with a guilty expression (he knew it was an unreasonable request). I suggested closing the windows, or moving the nursery to a room at the back of the house. Those accommodations were rejected. I told them I had already moved to the basement, which was MY reasonable accommodation, so they were going to have to live with it. Fortunately, we decided to move to a rural location soon after.

Boy, did I learn a lot about the difference between soundproofing and soundshaping in trying to think through all this. Of course, low frequencies are the most difficult to address in either category.

Rick "rather sensitive to boomy bass dominated by sharp resonances, though" Denney

 

[welsh]

I unashamedly embrace this.

There is nothing wrong with buying nice things just because they're nice.

Well I think we have to remember that audio components are part of the furniture - we have to look at them most of the time. I would pay a bit more, for instance, for an amplifier I found attractive over a cheaper one I thought ugly, even though I would not expect it to sound better. I stress, pay a BIT more. I imagine the cases of ‘high end’ stuff run to about 90% of manufacturing cost...

 

[welsh]

Both. All of them.

Any cognitive bias can have an impact.

It's not an 'either / or' situation.

One of the favourite arguments of those who favour sighted evaluations is, of course, “I expected to like X better but actually preferred (cheaper) Y! Ergo, I have no biases!

 

[kevin1969]

Don't go overboard on the search for the best possible SINAD. There is one very knowledgeable member here who says 72db for the electronics chain is enough for nearly everyone, and 78db for a few. Same goes for speaker preference scores. A little EQ goes a long way.

I would love to read about that

 

[Marmus]

While a lot of the interest here is on subsystems, and often traced down to individual components (op amps etc) there was a comment earlier about overall performance in morning vs evening.

Just a thought (from real measurements of course) on why sometimes, like morning, a system sounds worse than evening, or at various times. One is audible background noise in the environment, ie cars, kids, mowers, traffic, airplanes etc. It gets very observable during the day but we do get used to it and not necessarily notice its effect anymore. You may just be seeing it as a percieved degeradation in listening.

Another, more electronic oriented, is the RF background noise. In a neighborhood, while calibrating and reducing S/N in an FM bounce meteor detection station and operating it over 3 years, that the RF noise in low frequencies increased dramatically in the morning as people got up and switched lights, started things, AC systems kick on etc. I could detect the ones in our house and there was a clear signature (which is what the conformal mapping alogorithyms dealt with) for the local induced noise. It started about 5am and went to about 9am where it reduced a bit and stayed level until about 10pm-12pm. I would usually listen critically after 12pm. That noise was distinct from magnetosphere vibrations which only changed at sunrise and sunset except for solar storms (detected those at the same location too). While some of the local magnetosphere is affected by our cars and magnetic objects it is VLF and on the order of a few 10s-100s of nT so would not really be noticed. Just a note here that this is the level of S/N I hunted down over the years.

One other one, power line fluctuations caused by industrial equipment nearby, and some just otherss on the line from induction and the magnificent RF antenna our power lines present. That is coupled with your house grounds and nearby local equipment grounds. I have chased that where it was affecting amplifiers in mV and mA systems. Good grounds all the way to your home earth are important. And then the voltage swings and multi-phase noise on the line can get intrusive depending on who else is on your mains and how yours are wired in your last junction box with proper size wiring.

In one case, a medical FDA lab, we had to install our own high current 60hz power generation system (it was a room sized system) since there were times when we did power tests, even on properly designed medical equipment (power supplies, software, ASICs, FPGAs, etc) that we had burn outs and glitches due to all the above mentioned V,A and phase issues.

Just some thoughts on things I have personally measured and tracked to real issues in electronic circuits, thier inputs, outputs and interpretations (like hearing them and seeing them).

So, yes I have applied all most of this stuff to my home system too. Not power generation, just is not need based on fighting the larger sources of noise, like my own AC system dB, room modes, and my lousy hearing, all measured too.

 

[welsh]

While a lot of the interest here is on subsystems, and often traced down to individual components (op amps etc) there was a comment earlier about overall performance in morning vs evening.

Just a thought (from real measurements of course) on why sometimes, like morning, a system sounds worse than evening, or at various times. One is audible background noise in the environment, ie cars, kids, mowers, traffic, airplanes etc. It gets very observable during the day but we do get used to it and not necessarily notice its effect anymore. You may just be seeing it as a percieved degeradation in listening.

Another, more electronic oriented, is the RF background noise. In a neighborhood, while calibrating and reducing S/N in an FM bounce meteor detection station and operating it over 3 years, that the RF noise in low frequencies increased dramatically in the morning as people got up and switched lights, started things, AC systems kick on etc. I could detect the ones in our house and there was a clear signature (which is what the conformal mapping alogorithyms dealt with) for the local induced noise. It started about 5am and went to about 9am where it reduced a bit and stayed level until about 10pm-12pm. I would usually listen critically after 12pm. That noise was distinct from magnetosphere vibrations which only changed at sunrise and sunset except for solar storms (detected those at the same location too). While some of the local magnetosphere is affected by our cars and magnetic objects it is VLF and on the order of a few 10s-100s of nT so would not really be noticed. Just a note here that this is the level of S/N I hunted down over the years.

One other one, power line fluctuations caused by industrial equipment nearby, and some just otherss on the line from induction and the magnificent RF antenna our power lines present. That is coupled with your house grounds and nearby local equipment grounds. I have chased that where it was affecting amplifiers in mV and mA systems. Good grounds all the way to your home earth are important. And then the voltage swings and multi-phase noise on the line can get intrusive depending on who else is on your mains and how yours are wired in your last junction box with proper size wiring.

In one case, a medical FDA lab, we had to install our own high current 60hz power generation system (it was a room sized system) since there were times when we did power tests, even on properly designed medical equipment (power supplies, software, ASICs, FPGAs, etc) that we had burn outs and glitches due to all the above mentioned V,A and phase issues.

Just some thoughts on things I have personally measured and tracked to real issues in electronic circuits, thier inputs, outputs and interpretations (like hearing them and seeing them).

So, yes I have applied all most of this stuff to my home system too. Not power generation, just is not need based on fighting the larger sources of noise, like my own AC system dB, room modes, and my lousy hearing, all measured too.

 

[welsh]

I saw a cartoon here that reproduced my own experience. Years ago, an audiophile managed to lure me into his den. He had a ghastly vinyl playlist ready to be cued up (probably the Krall woman, Jazz at the Pawn Shop etc) but I insisted on some Mozart. He eventually found the Concerto in E flat K. 271. He looked at the record as if he didn’t trust it (he probably hadn’t worked out his VTA settings for it) but reluctantly put it on. It was all I could do to stifle laughter as I saw him bobbing his head and tapping his foot like a high-end salesman. Once it was over, he said: ‘What do you think?’ I said that Mozart chose the most brutal possible option for the entry of the piano, in the first six bars, never again attempted by Mozart, but developed by Beethoven. He looked at me as if I was mad. He finally said: ‘I meant, how does the system sound?’ I said, ‘fine, I could hear all the instruments.’ He never invited me round again.

 

[welsh]

I'm thinking how I could have been spending that time and money on a real guitar instead of that permanently out of tune disaster I was fingering, Would have given me hours of useful opportunities to apply OCD to musical issues instead of convincing myself that those vintage 12AX7's were really worth it.

Try a mandolin. You spend half your time tuning it, and the other half playing out of tune!

 

[welsh]

Probably more of a data tweaker than an audiophool.

Anyway, please note that my list was essentially self-flagellation, noting all the foolish things I've done in the name of neurotic audio, kinda like a 12-stepper confessing to his multiple sins. I also would solder Canare interconnect to XLRs for microphones, had a ton of that. But there was the possibility of making money from recordings, so not totally delusion. Also made too many RCA terminated interconnects, various wires of various pedigrees.

My most recent [this morning] bit of audiotweakery was inserting little discs of carefully cut circles of Viva towel to insert between the earcups and earpads of the Philips X2HR headphones that arrived yesterday.

Making acoustic mods to headphones is not really audiophoolery, as it can make a big difference (not always for the better, but it’s usually reversable). I once spent a lot of time modding the bargain-priced Fostex RPs, and ended up with (I think) a great headphone. Plasticine was involved...

 

[charleski]

I saw a cartoon here that reproduced my own experience. Years ago, an audiophile managed to lure me into his den. He had a ghastly vinyl playlist ready to be cued up (probably the Krall woman, Jazz at the Pawn Shop etc) but I insisted on some Mozart. He eventually found the Concerto in E flat K. 271. He looked at the record as if he didn’t trust it (he probably hadn’t worked out his VTA settings for it) but reluctantly put it on. It was all I could do to stifle laughter as I saw him bobbing his head and tapping his foot like a high-end salesman. Once it was over, he said: ‘What do you think?’ I said that Mozart chose the most brutal possible option for the entry of the piano, in the first six bars, never again attempted by Mozart, but developed by Beethoven. He looked at me as if I was mad. He finally said: ‘I meant, how does the system sound?’ I said, ‘fine, I could hear all the instruments.’ He never invited me round again.

Maybe I'm just an '80s throwback, but the first thing this brought to mind was ...

 

[welsh]

Back to the fundamental tenets of the thread, which I just scanned.

I read many posts that said we can’t and shouldn’t trust our ears, and should ignore our ears when making judgments. I profoundly disagree. Our ears are the point of the whole exercise. The problem is precisely that we don’t trust our ears, and place hearing surrogates (price tag, brand, SINAD measurements, Stereophile reviews, forum darlings, poetry, etc.) in their way. And then we hear what is not there, because we tell our ears what to hear instead of listening. We imagine a difference between 100 dB SINAD and 115 and then we “hear” it. We persuade ourselves that a $1000 DAC must better than a $125 DAC, and order our ears to perceive it. Our brains and biases are falsifying the plain qualitative data coming from our ears.

I listened to my Tascam CD-401 CD player through its analog outputs. Then I listened to it through my MF V90 DAC. Couldn’t hear a difference. The cable was an audio cable, not a 75-ohm coax, dummy. So, I bought a 75-ohm cable. Still couldn’t hear a difference. Waaaal, you need a better DAC, Rick—the Benchmark DAC3 must sound better than the MF—ASR says it has a SINAD of 121 instead of 105, Stereophile says it provides better image depth and detail and emits a nice floral bouquet, TAS says it will put the drummer 6.4 feet further back on the stage, the price tag is ten times as much, etc. Or, my Koss Pro4S headphones, known for being revealing to a fault, aren’t revealing enough, and God help you if you are trying to judge these things through Advents. But the truth is, I trust my ears: whatever the MF DAC does that my Tascam can’t do is unhearable, at least by me, and that Benchmark isn’t likely to, either. I trust that. It doesn’t prevent me from experimenting with other equipment, of course, but then that’s the hobby.

Our biases can be known and filtered out—that’s the point of objective testing—not that we shouldn’t trust the one sense that is the whole reason we buy this stuff.

I’m reminded of the argument about bokeh with camera lenses (the quality of the rendering of out-of-focus details). For a long time, those who fancied themselves as objectivists tested resolution, contrast, color accuracy, MTF, and other features that are only relevant in the focus plane, precisely when bokeh is irrelevant. We don’t have a way to quantify bokeh, but we can damn sure see it. It’s a qualitative effect, which doesn’t make it any less of an effect because we haven’t figured out a way to quantitatively measure it. The subjective part is whether we like the effect, or even whether we care about it, not what it looks like. We can trust our eyes, because seeing as a sense is more important to photography than talking about seeing. My art teacher: “Open your eyes! What do you see?”

All measurements of effects are projections of that effect, like a shadow. It describes shape from one angle and point of view. It’s a model of what we hope to hear, and in research, all models are false, even if some are useful. So, an effect that is sensed may show up in measurements if we are measuring the right thing, or not if we don’t. If we lack a model, then we have to test empirically, which is what blind testing is all about. It’s a way to validate the importance and accuracy of the model.

But in the end, if we can’t hear the effect, or if, having heard the effect, we have a preference, then that’s what matters. The point of rigor is to avoid deluding ourselves, and to train our senses, not to replace the senses with measurements. The objective, so to speak, is accurate subjective assessments, not biased, misinformed, or mystical subjective assessments.

I want my system to sound musical. I define accuracy as instruments sounding like themselves in the recorded environment, not merely that the output waveform is like the input waveform. I want that to be the case, of course, though the ways in which it isn’t may or may not be subjectively important to me. Transparency is when I listen to those instruments being played and am not distracted by the reproduction process so that I can focus on what the musicians are doing. The measurements help me screen out those items likely to be distracting, just like qualitative reviews from listeners whose judgments have proven reliable for me.

Rick “getting the new-guy philosophizing and subsequent roasting out of the way” Denney

Back to the fundamental tenets of the thread, which I just scanned.

I read many posts that said we can’t and shouldn’t trust our ears, and should ignore our ears when making judgments. I profoundly disagree. Our ears are the point of the whole exercise. The problem is precisely that we don’t trust our ears, and place hearing surrogates (price tag, brand, SINAD measurements, Stereophile reviews, forum darlings, poetry, etc.) in their way. And then we hear what is not there, because we tell our ears what to hear instead of listening. We imagine a difference between 100 dB SINAD and 115 and then we “hear” it. We persuade ourselves that a $1000 DAC must better than a $125 DAC, and order our ears to perceive it. Our brains and biases are falsifying the plain qualitative data coming from our ears.

I listened to my Tascam CD-401 CD player through its analog outputs. Then I listened to it through my MF V90 DAC. Couldn’t hear a difference. The cable was an audio cable, not a 75-ohm coax, dummy. So, I bought a 75-ohm cable. Still couldn’t hear a difference. Waaaal, you need a better DAC, Rick—the Benchmark DAC3 must sound better than the MF—ASR says it has a SINAD of 121 instead of 105, Stereophile says it provides better image depth and detail and emits a nice floral bouquet, TAS says it will put the drummer 6.4 feet further back on the stage, the price tag is ten times as much, etc. Or, my Koss Pro4S headphones, known for being revealing to a fault, aren’t revealing enough, and God help you if you are trying to judge these things through Advents. But the truth is, I trust my ears: whatever the MF DAC does that my Tascam can’t do is unhearable, at least by me, and that Benchmark isn’t likely to, either. I trust that. It doesn’t prevent me from experimenting with other equipment, of course, but then that’s the hobby.

Our biases can be known and filtered out—that’s the point of objective testing—not that we shouldn’t trust the one sense that is the whole reason we buy this stuff.

I’m reminded of the argument about bokeh with camera lenses (the quality of the rendering of out-of-focus details). For a long time, those who fancied themselves as objectivists tested resolution, contrast, color accuracy, MTF, and other features that are only relevant in the focus plane, precisely when bokeh is irrelevant. We don’t have a way to quantify bokeh, but we can damn sure see it. It’s a qualitative effect, which doesn’t make it any less of an effect because we haven’t figured out a way to quantitatively measure it. The subjective part is whether we like the effect, or even whether we care about it, not what it looks like. We can trust our eyes, because seeing as a sense is more important to photography than talking about seeing. My art teacher: “Open your eyes! What do you see?”

All measurements of effects are projections of that effect, like a shadow. It describes shape from one angle and point of view. It’s a model of what we hope to hear, and in research, all models are false, even if some are useful. So, an effect that is sensed may show up in measurements if we are measuring the right thing, or not if we don’t. If we lack a model, then we have to test empirically, which is what blind testing is all about. It’s a way to validate the importance and accuracy of the model.

But in the end, if we can’t hear the effect, or if, having heard the effect, we have a preference, then that’s what matters. The point of rigor is to avoid deluding ourselves, and to train our senses, not to replace the senses with measurements. The objective, so to speak, is accurate subjective assessments, not biased, misinformed, or mystical subjective assessments.

I want my system to sound musical. I define accuracy as instruments sounding like themselves in the recorded environment, not merely that the output waveform is like the input waveform. I want that to be the case, of course, though the ways in which it isn’t may or may not be subjectively important to me. Transparency is when I listen to those instruments being played and am not distracted by the reproduction process so that I can focus on what the musicians are doing. The measurements help me screen out those items likely to be distracting, just like qualitative reviews from listeners whose judgments have proven reliable for me.

Rick “getting the new-guy philosophizing and subsequent roasting out of the way” Denney

Generally, photography is quite a sane hobby compared to high end audio. There are no super expensive cameras that perform badly - they simply would not sell. There is no lunatic modding industry to magically improve your camera. However, subjective woo does exist - witness ‘the Leica look’, Zeiss ‘3D pop’ and anything ‘filmic’.

 

[welsh]

Your last phrase there is really interesting (thanks for sharing) because 40-some years ago there was a seminal blind test involving John Atkinson, among others, in which predictably no difference was heard between amps, so Atkinson bought a Quad 405. He used it for a spell, before relapsing.

His professed reason for relapsing was that the joy had been sucked out of listening to music, which quickly became a subjectivist meme, which eventually housed all the "things-we-can't-measure" and "long-term-is-essential" issues we still hear today. At the time I thought Atkinson was wrong, but strangely sincere. I guess, per your experience, he was missing the fun of the upgrade merry-go-round. Buying stuff is fun. We all know that. I love the way ASR people thread the needle by putting a system in every room ... and the garage.

Atkinson and co. have form with relapsing. Remember the notorious Carver Challenge, when the Stereophile Golden Ears had to admit that Carver had made a solid state amplifier sound exactly like a (deliberately unidentified) ‘Reference’ valve-woo unit? Then they thought about, well, advertisement revenue, and savagely backtracked, getting Mr Harley (as I remember) to viciously attack Carver in a review, and claiming that in long-term listening, they could now all tell the difference.