If you can hear it, it must be measurable

[cwall99]

Since the thread on Audiophile Audiophoolery got closed before I got a chance to most MY thoughts (and, as we all know, mine rule -- Who the heck is this cwall99 guy, anyway?).

I firmly believe that anything that can be heard can be measured. This goes for some of the, ahem, interesting adjectives that are applied to audio products.

Think about it. If you perceive a difference, there must be some physical difference in the audio signal reaching your ears. That's a physical phenomenon with manifestations in the real world. As such, those physical differences must be detectable.

But, there are a few problems with this:

First, what do all those interesting adjectives mean? When Audiophile A talks about the openness of the soundstage, does he (or she, let's be fair) mean the same thing as Audiophile B? So, the first thing that needs to be done is to create operational definitions for all these terms so that there can be confidence in the claims that we're all talking apples and not oranges.

Creating operational definitions like this, though, is no easy task. It's a lot like what I do for work: design training programs. One of the things I do is limit my programs to modifying observable behavior. Frankly, I don't care about what's going on in the learner's head, I just care that they can do what it is they're supposed to be able to do.

Still, I suspect that getting two audiophiles to agree on what these terms mean in some sort of measurable sense would be a challenge.

Second, even though a quality exists in the physical world, and its presence there means it should be measurable, that doesn't mean that the actual means to measure that quality exists. Speaking of the openness of a system's sound.... we'd need some sort of audio openometer to measure that, and I'm pretty sure that it doesn't exist yet.

So there are lots of qualities of audio that may be affected by audiophile equipment, but the means to measure them don't exist.

Third, we're all different. The physical structure of our ears varies from person to person. Just look at physical structure President Obama's outer ears versus that of his opponent's from last fall's election. You think those wings Obama sports don't make a difference in the way he hears things as opposed to how McCain hears them (we're not talking politics here, just the shape of the outer ear)?

Those physical differences aren't just limited to the outer ear. I'm sure there are all sorts of minute physical differences in our inner ears, too, based on the size of our heads, our age, any physical damage we may have incurred listening to music at extreme volumes, etc....

So, here's the problem, even if we can agree on what these terms mean, we have no way of ensuring that we perceive these differences similarly.

But, we've hit standards like this for video. Audio is just a lot trickier to measure, I believe, so it'll be a while before we can get those measuring devices and the definitions of things they're supposed to measure.

Just hoping that this hasn't already been done to death.

 

[marqueemoon]

So, here's the problem, even if we can agree on what these terms mean, we have no way of ensuring that we perceive these differences similarly.

I had a conversation with a friend not too long ago about a sweater she was wearing. We (and the other people in the room) had differing opinions about what color it was.

I think the terminology had gotten a little stupid/nonsensical though. Sometimes I swear I'm reading a description of a fine wine (minus the usual reference points).

 

[cwall99]

The goofy thing about color is that the reflected light can be measured as having a specific wavelength. That's a lot easier to measure than some weird wine-like (to borrow your metaphor) adjective like openness or density or any other goofy term used to describe audio qualities.

In the case you describe, it sounds more like "Well, I think it's fuschia," or "No, I think that's hot pink."

Given the fact that few of us walk around with a Pantone book that we can whip out to resolve differences of opinion. Still, if we're relying on our eyes to compare one color to a standard, we run into the same sort of problems I was talking about when I was considering the differences among our ears.

 

[dogscanskate]

How do you describe what you hear to someone? Scientific terms do not define what you have felt "emotionally", so you have to find within your vocabulary some adjectives, analogies to describe what you "feel".

Music is "art" and we use science to build products to convey this art form. We could just as well try to reverse engineer a "painting masterpiece" so that we could explain it's emotional impact through scientific terms. You know of anyone that does this? I think it becomes a moot point for many art forms. You just enjoy what you see, what you hear, what you taste...

Some people such as wine tasters use the same terms we use in describing their tasting experiences.

Adjectives and analogies bring forth a glimpse of what we "feel" about a musical performance that was played through a system. Science describes how the gear works.

Science does not describe art and art does not define science.

 

[electroking]

The goofy thing about color is that the reflected light can be measured as having a specific wavelength. That's a lot easier to measure than some weird wine-like (to borrow your metaphor) adjective like openness or density or any other goofy term used to describe audio qualities.

In the case you describe, it sounds more like "Well, I think it's fuschia," or "No, I think that's hot pink."

Given the fact that few of us walk around with a Pantone book that we can whip out to resolve differences of opinion. Still, if we're relying on our eyes to compare one color to a standard, we run into the same sort of problems I was talking about when I was considering the differences among our ears.

Well, visual signals are not that simple when you think a bit further.
For instance, I think we all agree that ground coffee beans yield
a brown substance, yet you would be hard pressed to tell what
wavelength corresponds to 'brown light'...

One of the problems with audio signal analysis, or with any time
varying signals for that matter, is that much of our perception
is associated with non recurring transient phenomena, whereas
most of the elementary signal descriptions are based on steady-state
analysis, averaging by spectrum analyzers and the like.

 

[JohnVF]

I agree that for most things we hear, there should probably be some measurement that applies.

But outside the role of DIY and professional equipment designers, it doesn't make a difference to me. I don't care. If I perceive a difference, that' enough for me. Because if I perceive that I like speaker A or over speaker B, I couldn't care less what measurements are out there to try to prove why. I just like one over the other. If it's really a difference that I'm hearing, or if it's just something I'm imagining...I don't really care! I feel no obligation to prove to anybody in the audio community why I choose the gear that I choose.

All that matters to me is that I can sit down at the end of the day and be taken somewhere else by the music. The gear provides that window, and I tweak it to be as open as possible.

 

[Mr. Lin]

I firmly believe that anything that can be heard can be measured. This goes for some of the, ahem, interesting adjectives that are applied to audio products.

But that doesn't necessarily mean anyone currently knows how to go about measuring such things, and that's why there's so much debate about cables, for example. The differences are real, they're there, but in most cases there's not much of a proven method for demonstrating why cable A should sound so different from cable B.

I tend to agree with John here. I don't really care about measurements, I care about what I hear, and I also think it's foolish of anyone who designs audio components to rely entirely on science and measurements, rather than by listening.

 

[cwall99]

But that doesn't necessarily mean anyone currently knows how to go about measuring such things, and that's why there's so much debate about cables, for example. The differences are real, they're there, but in most cases there's not much of a proven method for demonstrating why cable A should sound so different from cable B.

I had hoped that I addressed that point in my original post, that even though we can perceive something, we may be a long ways from creating a device that can actually measure the phenomenon. Heck, we're even a long way from agreeing on what the operational definition of that phenomenon is.

Nevertheless, I stick by my point: if we can perceive something, we're perceiving it because a real force acted on our ears. There's something in that auditory stimulus that we detect (putting aside, for the moment, any discussions on placebo effect). It's a real thing, and, therefore, it is subject to measurement.

This has nothing to do with how the music makes me feel. That's a different matter, but now that I think about it, and a lot of this sort of touches on your fundamental view of what makes us what we are, even those funky biochemical reactions that occur as we experience emotions are measurable. However, I do think our own reactions to stimuli vary based on our own experiences.

Gee, this is deeper than I wanted to go.

 

[Mr. Lin]

I had hoped that I addressed that point in my original post, that even though we can perceive something, we may be a long ways from creating a device that can actually measure the phenomenon. Heck, we're even a long way from agreeing on what the operational definition of that phenomenon is.

My mistake, you did. I just went through your first post again and you state that very thing. I guess after reading it and the following posts I just couldn't get the thought out of my head, and simply typed.

 

[marqueemoon]

I had hoped that I addressed that point in my original post, that even though we can perceive something, we may be a long ways from creating a device that can actually measure the phenomenon. Heck, we're even a long way from agreeing on what the operational definition of that phenomenon is.

I think at the hair splitting level this process of discovery will tell us more about the human brain than any gear, and that is not a bad thing.

 

[whoaru99]

But that doesn't necessarily mean anyone currently knows how to go about measuring such things, and that's why there's so much debate about cables, for example.

With all due respect, I disagee.

Cable arguments, IMO, generally stem when someone suggests applying scientific principle (theory, test/measure), rather than relying on anecdotal evidence.

I fully believe that any audible difference a cable or any other component might make can be measured, and that the methodology of these measurements resides in current technology.

 

[RichPA]

I fully believe that any audible difference a cable or any other component might make can be measured, and that the methodology of these measurements resides in current technology.

I'm in complete agreement with the first half of this, and in general agreement with the second half - but I'd add that the needed measurement methodology is not necessarily what is simply and routinely used.

 

[JonL]

With all due respect, I disagee.

Cable arguments, IMO, generally stem when someone suggests applying scientific principle (theory, test/measure), rather than relying on anecdotal evidence.

I fully believe that any audible difference a cable or any other component might make can be measured, and that the methodology of these measurements resides in current technology.

I wish it were true that the methodology exists to quantify all audible differences, and perhaps it is. I think human senses (probably ALL animal species' senses) are remarkable in their ability to discern minute and subtle distinctions. I don't think we understand enough about how the senses work to be able to craft sound scientific methods to quantify all of what our senses perceive. Where the debate degenerates is at the margins of what we can hear and what we can scientifically quantify. I have a great deal of skepticism about the claims made for many "tweaks." I'd like to see a plausible physics-based explanation for why a tweak does what is claimed, even if we cannot readily measure the claimed phenomena. Too often the claims are simply hand-waving, or bend the laws of physics, or require a faith-based approach to audio.

Music is a very complex waveform, and I'm not sure that deconstructing it with fourier transforms, looking at square waves, measuring THD and IMD with test signals, etc. can accurately and completely quantify what we hear (re)produced on our audio equipment, though these technique probably get us 98% of the way there.

So, I guess what I'm saying is that I am willing to accept that some things that we can hear are beyond our knowledge of measuring. I can cautiously accept claims that have a believable scientific underpinning even if the measurement technology to validate the claim is currently unknown. I am extremely skeptical of claims that do not have at least a plausible scientific foundation.

 

[JohnVF]

We simply don't understand how the universe works. There is no unifying scientific theory, nobody really understands what is going on. Even gravity is largely a mystery. So it doesn't surprise me when I hear something in my audio system that doesn't show up on a chart. Regardless, I'm too busy enjoying the music and the improvements I've, even if only I can hear them, to worry about it.

 

[whoaru99]

Understanding how a signal propagates through a wire or transistor is a slightly, IMO, different scale of understanding than the universe and all of creation.

EDIT: Perhaps another way to put it is just because we don't understand some things, doesn't mean we don't/can't understand others.

 

[shimniok]

Interesting topic!

Think about it. If you perceive a difference, there must be some physical difference in the audio signal reaching your ears.

I submit that in some cases--not all--our mind might play tricks on us, like the audio equivalent of optical illusions. (and other factors, like ears as you mentioned, come into play)

To restate the above, if you perceive a difference between A and B, it is because there is a difference either in the audio signal, or because of something in your ears/brain. I would think both are measurable---if only we knew what to measure and how.

Second, even though a quality exists in the physical world, and its presence there means it should be measurable, that doesn't mean that the actual means to measure that quality exists. Speaking of the openness of a system's sound.... we'd need some sort of audio openometer to measure that, and I'm pretty sure that it doesn't exist yet.

Or... it may also be that we don't know what measurements correspond to a given perception (openness, or clarity, or ...?)

When someone says a set of speakers are "overly bright sounding" -- does that simply correspond to peaks in the 6kHz, 8kHz, 10kHz, 15kHz ranges? Or is it distortion related too? Or related to off-axis response? Or something even more complex?

Stereophile, for what it's worth, says "Brightness relates to the energy content in the 4kHz-8kHz band." in their glossary.

Openness, according to Stereophile means "Exhibiting qualities of delicacy, air, and fine detail. Giving an impression of having no upper-frequency limit. " Maybe others use the term differently. But no matter, this is probably more complex... fine detail is what... a question of dynamics? distortion? phase shifts? slew rate? frequency response? Some or all of the above??

Or detail -- "The subtlest, most delicate parts of the original sound, which are usually the first things lost by imperfect components. See "low-level detail." Compare "haze," "smearing," "veiling."" -- how the hell do you measure smearing or haze? What constitute delicate parts of the sound, and what destroys them?

We know from Zilch's "guess which speaker sounds best based on measurements" thread that it isn't necessarily straightforward to predict complete how a set of speakers will sound based on curves. A recent thread that posted a link about Harman's audio lab and talked about audio perception -- I think they suggest that curves don't tell us much until they have granularity to 1/20th octave and that other measurements like off-axis and whatnot play in heavily into what "sounds good" to semi-educated listeners.

One thing I doubt we'll ever be able to do is measure or predict emotional response to music based on the system playing it, type of music, etc -- because that is too variable, based on personal experiences, tastes, etc.

Michael

 

[JonL]

We simply don't understand how the universe works. There is no unifying scientific theory, nobody really understands what is going on. Even gravity is largely a mystery. So it doesn't surprise me when I hear something in my audio system that doesn't show up on a chart. Regardless, I'm too busy enjoying the music and the improvements I've, even if only I can hear them, to worry about it.

I think you are overstating things. We certainly understand enough about how the universe works to accomplish many incredible things with technology. My post wasn't meant to seem defeatist -- as in "we'll never understand why these interconnects sound different, so let's not bother -- rather that we just don't yet know absolutely everything to look for.

 

[whoaru99]

We know from Zilch's "guess which speaker sounds best based on measurements" thread that it isn't necessarily straightforward to predict complete how a set of speakers will sound based on curves. A recent thread that posted a link about Harman's audio lab and talked about audio perception -- I think they suggest that curves don't tell us much until they have granularity to 1/20th octave and that other measurements like off-axis and whatnot play in heavily into what "sounds good" to semi-educated listeners.

Michael

No one curve is a predictor of superior performance. I think most of us already knew that.

A key takeaway for me was that if you gather all the measurement curves they made of different parameters for each speaker, the results come together in a meaningful and pretty consistent manner as to which will sound better/best to the listeners.

 

[fotno]

I just finished reading this thread in its entirety, and would only add the following. The thread title "If you can hear it, it must be measurable" doesn't fully address one thing that must be taken into consideration... Who's listening?

While I agree that audible differences can (and should) be measured to one degree or another as a tool for construction and instruction, whether those measurements can be used to indicate whether a given device will be pleasing to a given listener is another matter entirely. Earlier in this thread the ear was likened to the eye. While we may use a color chart to determine whether a sweater is fuchsia or hot pink, that chart can't tell you anything about the percentage of people in the room who find it an appealing color, and those who don't. The ears, like the tongue and the eyes, are an entrance way to the mind, and in the end, the mind determines if what one intakes is pleasing at the moment of entry.

At the end of the day, I don't think it's so much a question of "can it be measured?" as "what do these measurements mean?" to a given listener.

 

[cmb3366]

I would agree, except that in measuring you are introducing another variable, that of the measuring equipment and techniques themselves. I use my ears as the final transducers and my noggin as my final measurement tool when enjoying my stereo, so I see no reason not to use them when evaluating my stereo. I get enough science, engineering, testing and measurement in the lab at college and work, my stereo is for my enjoyment only. YMMV:thmbsp: