Jump to content
IGNORED

Logical Fallacies in Audiophile Discussions


Recommended Posts

There are many different ways to group Audiophiles. One such way is to group the doggedly scientific folks separate from those who ignore all of the specs and all of the "known" scientific reasoning and just go with what their ears tell them.

 

From a purely academic viewpoint, I find the gray area in between these two groups to be utterly fascinating. In that gray zone lies the firm belief that we don't yet know as much as we claim to know about what makes good sound.

 

If anyone is game, I would enjoy using this thread to discuss that "gray zone" and poke around at some of the long-held scientific "facts" that seem to be in contradiction to careful listening.

 

I'll get the ball rolling with one of my favorite Audiophile Science myths:

"since none of us can hear much above 20khz, frequency response beyond that zone is of no consequence in making good sound."

 

I think that the logical fallacy here is the notion that our inability to resolve a 30khz sine wave as a pure tone categorically rules out any effect on the listening experience from such a "sound".

 

As an analogy, simply look at the other end of the audio frequency spectrum. I think we're all agreed that humans cannot perceive a 10hz sine wave as a "tone". At the same time, given sufficient amplitude I think we would all be running from the room from such an "inaudible sound". While 10hz is not strictly "audible" in one sense, it clearly can have an impact on a listening experience.

 

So while we can't resolve a 25khz sine wave as a "tone" (or even 16khz in my case), I think it is foolhardy to then assume that same tone, or some transient aspect of a lower frequency tone, has no impact on a listener.

 

Just curious, does anyone know of any ABX testing done to see if a human listener can differentiate a 15khz sinewave from a 15khz square wave? Would not all of the additional energy that squares off such a wave be at 30khz and above?

 

Even if the above test fails, could there not be some sort of masking or other psycho-acoustic interplay of "supersonic" energy with a more complicated "mid-sonic" musical passage with rich timbre?

 

I think right now we understand all there is to know about the "brute force" parts of audio. The "obvious" parts. The technology has brought us into the realm where we are plumbing the less obvious parts of it now. We should start that analysis with an understanding that we don't yet know what we don't yet know. This would make us true Scientists of sound rather than engaging in "scientific dogma".

 

Looking forward to a lively debate of this and other Audiophile logical fallacies.

 

New guy here - old guy elsewhere...Mac Mini - BitPerfect - USB - Schiit Bifrost DAC - shit cable - Musical Fidelity A3.5 - home-brew speakers designed to prioritize phase and time response (Accuton ceramic dome drivers and first-order crossovers) and a very cheaply but well corrected room...old head, old ears, conventionally connected to an old brain with outdated software.

 

"It’s easier to fool people than to convince them that they have been fooled." -- Mark Twain

Link to comment

I think the proper place to start is placebo effects.

 

How would you know the difference without ABX or some similar outside reference beyond just your ears?

 

Yet, so many things are so obvious we need nothing of the sort. Need ABX testing to distinguish between Chinese and English voices? No of course not. As a matter of fact use a good sound editor to generate 15 khz sine and square waves. Will be as obvious as the language and you need nothing further. I just did it to confirm before making that statement. Should you do an ABX test with the two waveforms, unless suffering high levels of hearing loss, you would easily score 100%.

 

Try comparing .5% distortion, and .01%.....well, you will be hard pressed to do it reliably I think. Ditto for jitter levels of digital audio to some disturbingly high levels of jitter. Yet some are convinced they are really important. How can you determine if it is for real or placebo? How do you shake the feeling that one piece of equipment simply satisfies, yet another somehow doesn't?

 

My rational mind tells me one thing, and my gut (heart felt) experience tells another. Yet clearly on a few points, my gut lead me astray. I belief the former is most often right, yet how do you argue with deep in your bones experience without feeling like you have betrayed yourself or practically admitted you have trouble perceiving anything without outside help?

 

Fascinating for sure as an emotional human being. I do fear much too often we are fooling ourselves, taking part in our own self-deception. E.G.(Hearing data cables on hard drives for instance).

 

And always keep in mind: Cognitive biases, like seeing optical illusions are a sign of a normally functioning brain. We all have them, it’s nothing to be ashamed about, but it is something that affects our objective evaluation of reality. 

Link to comment

Could you repeat the fallacy? I think my wireless internet must be running too loud.

 

"I think that the logical fallacy here is the notion that our inability to resolve a 30khz sine wave as a pure tone categorically rules out any effect on the listening experience from such a "sound"."

 

Sorry, how is this a logical fallacy? Either we can detect a 30kHz tone or we cannot, which is a question of fact, not of logic.

 

A question of logic might be states as follows:

 

Given that we cannot hear a tone higher in frequency than 22.05 kHz, any sound impulse digitally sampled at 44.1 kHz is described completely, and anything above that is inaudible overkill.

 

The logic is based on Fourier theory, i.e., that any periodic function can be expressed as a sum of individual sine waves, and any arbitrary function can be expressed as a continuous sum (integral) of individual sine waves. Then, using the Shannon/Nyquist proof, we claim that it is sufficient to truncate the summation at 44.1 kHz, as any further detail in the sound we are attempting to represent (sample) will be inaudible.

 

That logic is either sound (pun intended) or not.

 

I'm unclear whether you want to question that argument, or if you want to call into question the empirical observation that humans cannot hear or otherwise detect a pure tone above 22.05 kHz (which is a quite generous upper bound). If the latter, we need something in the way of empirical evidence, and it becomes a question of fact, rather than a question of logic.

 

 

 

Link to comment

since none of us can hear much above 20khz, frequency response beyond that zone is of no consequence in making good sound

 

Well, there was a study done with brain activity imaging showing that people detect ultrasonic sounds when present with related lower harmonics.

 

But one of the reasons not to heavily limit bandwidth is that placing a steep filter at 20 kHz has strong time-domain impact. These filters have impact longer than single cycle of 20 kHz sine. And also depending on type of the filter, heavy phase shifts already at much lower frequencies, like 2 kHz.

 

And with today's storage and converter technologies, there's no need to artificially band-limit the audio signals at all. Let it roll-off freely. For some instruments that requires bandwidth of over 100 kHz.

 

Hearing itself is specialized on detecting the shape of attack -phase. Thus for instrument sounds, preserving the attack intact is very important, more important than the steady state or decay.

 

 

Signalyst - Developer of HQPlayer

Pulse & Fidelity - Software Defined Amplifiers

Link to comment

conflating "can't recognize a tone of x hz" with "have no awareness of" or "completely uninfluenced by".

 

True, we can't hear a TONE above 20khz as a tone. Doesn't follow from that that a 25khz tone does not have any influence on the listener. Maybe we can hear it but not as a tone. Maybe only as a modification of some lower frequency tone or tones.

 

New guy here - old guy elsewhere...Mac Mini - BitPerfect - USB - Schiit Bifrost DAC - shit cable - Musical Fidelity A3.5 - home-brew speakers designed to prioritize phase and time response (Accuton ceramic dome drivers and first-order crossovers) and a very cheaply but well corrected room...old head, old ears, conventionally connected to an old brain with outdated software.

 

"It’s easier to fool people than to convince them that they have been fooled." -- Mark Twain

Link to comment

"True, we can't hear a TONE above 20khz as a tone. Doesn't follow from that that a 25khz tone does not have any influence on the listener. Maybe we can hear it but not as a tone. Maybe only as a modification of some lower frequency tone or tones."

 

That sounds like a logical fallacy to me (i.e., that a 25 kHz signal can be decomposed into lower-frequency components).

 

Link to comment

What I feel is most important is to reduce this gray area. Regarding the limits of human auditory perception:

 

http://informahealthcare.com/doi/abs/10.3109/00381797009075523

 

There is more to this of course, including how higher frequencies influence our perception of phase shifts and distortion artefacts, as already mentioned in an earlier post.

 

Ron

 

My answer to room acoustics? ...Headphones!

Link to comment

...was talking about one high frequency (slightly "ultrasonic") tone effecting our hearing of some other sonic tone.

 

As an example, when a bow is drawn across a violin string, it makes both the tone we can hear and identify, and also harmonics of various sorts. There are also "tones", and harmonics of same coming off of each of the (non-fundamentally tuned) hairs in the bow. All that cluster of harmonics, most of which are above the frequency at which we can identify something as a "tone" by itself...impinges upon our ears.

 

I'm just saying that some of that cocktail of somewhat "ultrasonic" sound may make the sonic part sound different.

 

Some of it (from the non-tuned bow hairs for instance) could heterodyne with things in the sonic range and create additional audible tones within the "normal" range of our hearing.

 

Some of it could also provide some type of yet-to-be-named psycho-acoustic masking of the sound we can hear in a subtle but relevant way.

 

New guy here - old guy elsewhere...Mac Mini - BitPerfect - USB - Schiit Bifrost DAC - shit cable - Musical Fidelity A3.5 - home-brew speakers designed to prioritize phase and time response (Accuton ceramic dome drivers and first-order crossovers) and a very cheaply but well corrected room...old head, old ears, conventionally connected to an old brain with outdated software.

 

"It’s easier to fool people than to convince them that they have been fooled." -- Mark Twain

Link to comment

"Some of it (from the non-tuned bow hairs for instance) could heterodyne with things in the sonic range and create additional audible tones within the "normal" range of our hearing."

 

So think of it this way: If you cut out everything BELOW 22kHz, would you be able to hear anything?

 

(You are saying "decomposed into" BTW):

 

cf: http://en.wikipedia.org/wiki/Fourier_series

cf: http://en.wikipedia.org/wiki/Fourier_transform

 

 

 

Link to comment

Parsing a few of your audiophile myths:

 

I think that the logical fallacy here is the notion that our inability to resolve a 30khz sine wave as a pure tone categorically rules out any effect on the listening experience from such a "sound".

 

This is why the majority of audiophiles have dogs.

 

I think we're all agreed that humans cannot perceive a 10hz sine wave as a "tone"

 

This is true, they perceive it as an earthquake...and like the effect. Many buy subwoofers the size of steamer trunks because of this.

 

Just curious, does anyone know of any ABX testing done to see if a human listener can differentiate a 15khz sinewave from a 15khz square wave?

 

My hunch is that most of this testing on the human effect of different styles of waves has been done by Helene Curtis.

 

Even if the above test fails, could there not be some sort of masking or other psycho-acoustic interplay of "supersonic" energy with a more complicated "mid-sonic" musical passage with rich timbre?

 

There could, especially if you wear a homemade aluminum foil helmut with a self-styled six inch spike protruding from the top. Are you a writer for the TV show "V" or perhaps a cast member? Just wondering.

 

We should start that analysis with an understanding that we don't yet know what we don't yet know.

 

Speaking solely for myself, I yet don't know what I don't know yet but I'm looking for an online course.

 

cheers,

 

Bill

 

 

 

 

Cheers,

 

Bill

 

 

Mac Mini 2011, 60 gb SSD, 8gb ram; PureMusic & BitPerfect; Wavelength Audio Cosecant V3 DAC; Wireworld Silver Starlight usb interconnect; McIntosh C2200 preamp; pair of McIntosh MC252 SS amps run as monoblocks; vintage MC240 Tube amp and 50th Anniversary MC275 tube amps; Krell LAT-2\'s on Sound Anchors; JL Audio F112 subwoofer; Nirvana SX ltd interconnects and speaker cables and power cords; PS Audio P5

Link to comment

So think of it this way: If you cut out everything BELOW 22kHz, would you be able to hear anything?

 

As the brain study have shown, no. And as it has also shown, the brain response is different if you have JUST the frequencies below 22 kHz or if you have frequencies below AND above 22 kHz.

 

Since it's about shape of the base frequency waveform, you cannot separate these two parts. Hearing is not doing spectrum analysis, it's analyzing the actual waveform shape, especially the initial attack part. Like plucking a string or hitting a cymbal, when the transient waveform begins.

 

So it's like instead of looking spectrum analyzer, it's looking at oscilloscope.

 

So if you take Boyk's results and look at the attack of claves you get the idea. This is the part that gives instrument a fundamental character.

 

 

Signalyst - Developer of HQPlayer

Pulse & Fidelity - Software Defined Amplifiers

Link to comment

There could, especially if you wear a homemade aluminum foil helmut with a self-styled six inch spike protruding from the top. Are you a writer for the TV show "V" or perhaps a cast member? Just wondering.

 

I don't care who you are, that's funny. :) (quoting Larry the Cable Guy)

 

Reading back over the my original post in the light of your comments was quite amusing to me. It is interesting how our current understanding of something shapes the language we use to discuss it...almost certainly hampering our efforts to describe a new theory or explanation.

 

"supersonic energy and psycho-acoustic interplay" -- LOL definitely does sound like it came from a trailer for that 'V' show.

 

For the record, my "spike" is considerably in excess of 6 inches.

 

New guy here - old guy elsewhere...Mac Mini - BitPerfect - USB - Schiit Bifrost DAC - shit cable - Musical Fidelity A3.5 - home-brew speakers designed to prioritize phase and time response (Accuton ceramic dome drivers and first-order crossovers) and a very cheaply but well corrected room...old head, old ears, conventionally connected to an old brain with outdated software.

 

"It’s easier to fool people than to convince them that they have been fooled." -- Mark Twain

Link to comment

So if you take Boyk's results and look at the attack of claves you get the idea. This is the part that gives instrument a fundamental character.

 

What I also wanted to say about this, is that if you put this through a linear phase brick-wall filter (antialiasing filter on ADC), it pretty much destroys the attack. Brain is still quite good at guessing what it is trying to represent. Just like you can (usually) recognize other person over a cellular phone call, even though the audio quality is really crappy.

 

Listening suddenly becomes much more relaxing and complete experience, when brain has less guessing to do.

 

 

 

Signalyst - Developer of HQPlayer

Pulse & Fidelity - Software Defined Amplifiers

Link to comment

Thanks for keeping the discussion lively.

 

"so think of it this way...would you be able to hear anything"

 

Probably not. Did the tree that fell in the forest still make a sound? Yes, I rather think it did. I think your point is precisely the other side of the same fallacy.

 

"hear anything"? What does that mean? Does it mean raise your hand when the person running the hearing test hits the button? If that is the definition, then no; I would not be able to hear it.

 

On the other hand, if there were other sounds present that I could hear, your theory is that in no case could an experiment be structured such that I could hear a change in those audible tones when the one I "can't hear" is turned on and off. This is an experiment I would like to see done scientifically.

 

And to your last point of "You are saying decomposed into", I did sit quite painfully through all of the Fourier (and Laplace) classes in my youth at EE school. Quite familiar. I was not, however, talking about a single fundamental sine-wave above 20khz being "decomposed into" some non-existent parts below 20khz. I was talking about our auditory experience of one or more sub-20khz tones being effected in some way by above-20khz harmonics of those fundamental frequencies.

 

Our language is shaped by our understanding...making it difficult to describe possibilities that fall slightly outside that understanding.

 

I'm enjoying both the scientific side of this discussion and the "game" of trying to get the semantics just right. The latter part is somewhat maddening, and also somewhat addictive. I fear it will wind up taking on a life of its own in this thread quite separate from the science of audio.

 

 

 

New guy here - old guy elsewhere...Mac Mini - BitPerfect - USB - Schiit Bifrost DAC - shit cable - Musical Fidelity A3.5 - home-brew speakers designed to prioritize phase and time response (Accuton ceramic dome drivers and first-order crossovers) and a very cheaply but well corrected room...old head, old ears, conventionally connected to an old brain with outdated software.

 

"It’s easier to fool people than to convince them that they have been fooled." -- Mark Twain

Link to comment

cf: ABX testing thread

 

In the ABX testing thread, it is asserted that ABX testing sometimes cannot resolve genuine differences simply because the testing environment, conditions and atmosphere is not condusive to relaxed musical listening, where such things become apparent.

 

So here is the logical fallacy (as far as I can see it): A genuine difference is proposed between A and B. A double-blind test leaves the experimental subjects unable to identify a difference between A and B in a statistically significant way. Therefore, hence and ergo, double-blind testing cannot be relied upon to detect genuine differences.

 

The flaw in that line of logic is that it presupposes that there is a true difference between A and B. If there is, how do we know that this is the case, if we can't measure it in a reproducible way?

 

In other words, you are assuming what you set out to prove, so you cannot logically conclude anything.

 

Link to comment

I'm going back to my mobius loop and mind my own business. I was told there would be no math required on CA.

 

cheers,

 

Bill

 

Cheers,

 

Bill

 

 

Mac Mini 2011, 60 gb SSD, 8gb ram; PureMusic & BitPerfect; Wavelength Audio Cosecant V3 DAC; Wireworld Silver Starlight usb interconnect; McIntosh C2200 preamp; pair of McIntosh MC252 SS amps run as monoblocks; vintage MC240 Tube amp and 50th Anniversary MC275 tube amps; Krell LAT-2\'s on Sound Anchors; JL Audio F112 subwoofer; Nirvana SX ltd interconnects and speaker cables and power cords; PS Audio P5

Link to comment

it would be great if someone could post links to files that the CA community could then test to confirm/dispel the fallacies/myths/etc. for example, 2 files from the same master - tweaking whatever variable is at issue - and ask if anyone hears a difference, and what it is, etc.

 

iPad2 + RemoteApp/VNC Viewer --> Headless Mac Mini --> iTunes * ALAC --> cheap USB cable WireWorld Ultraviolet USB cable --> Musical Fidelity V-LINK --> SonicWave Toslink --> Musical Fidelity V-DAC --> $.97 (RadioShack clearance) Monster THX Digital Coax --> AIWA NSX-3300 --> Polk RTi4\'s --> Cheapskate Listening Enjoyment[br]

Link to comment

"there are no gnomes...but there definitely are weapons of mass destruction and I'm looking across the table at a couple right now"

 

Cheers,

 

Bill

 

 

Mac Mini 2011, 60 gb SSD, 8gb ram; PureMusic & BitPerfect; Wavelength Audio Cosecant V3 DAC; Wireworld Silver Starlight usb interconnect; McIntosh C2200 preamp; pair of McIntosh MC252 SS amps run as monoblocks; vintage MC240 Tube amp and 50th Anniversary MC275 tube amps; Krell LAT-2\'s on Sound Anchors; JL Audio F112 subwoofer; Nirvana SX ltd interconnects and speaker cables and power cords; PS Audio P5

Link to comment

Lamp cord is exactly as good as the most expensive loudspeaker cable.

 

The truth validated by our own logic.

 

Cheers,

 

Bill

 

 

Mac Mini 2011, 60 gb SSD, 8gb ram; PureMusic & BitPerfect; Wavelength Audio Cosecant V3 DAC; Wireworld Silver Starlight usb interconnect; McIntosh C2200 preamp; pair of McIntosh MC252 SS amps run as monoblocks; vintage MC240 Tube amp and 50th Anniversary MC275 tube amps; Krell LAT-2\'s on Sound Anchors; JL Audio F112 subwoofer; Nirvana SX ltd interconnects and speaker cables and power cords; PS Audio P5

Link to comment

On the other hand, if there were other sounds present that I could hear, your theory is that in no case could an experiment be structured such that I could hear a change in those audible tones when the one I "can't hear" is turned on and off. This is an experiment I would like to see done scientifically.

 

I think this is fairly scientific version of the test:

http://jn.physiology.org/content/83/6/3548.abstract

 

I would say much more reliable than ABX.

 

But I also recognize that there's no way either camp can convince the other. No matter what.

 

 

Signalyst - Developer of HQPlayer

Pulse & Fidelity - Software Defined Amplifiers

Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now



×
×
  • Create New...