The science and magic of how a guitar produces acoustic sound

[zmf]

Originally Posted by Knives&Guitars

For those of use who do not know all the science, Magic is actually an appropriate word.

This reminds me of a cartoon showing a scientist finishing a very long equation on a blackboard, and arriving at the final line "then magic happens".

We are, at times, certainly awed by some things that do not yet have an explanation, as we should be as sentient beings, but "magic" is not (IMHO) an appropriate viewpoint. It can even be a dangerous concept under some circumstances. None of us know "all of science", but we don't have to fill in the blanks with magic.

Apologies. End of sermon.

[frankmcr]

Or you could say

"The Quantifiable and (As Yet) Unquantifiable Elements of How a Guitar Produces Acoustic Sound"

Not very catchy, I know . . .

[srick]

Originally Posted by zmf

This reminds me of a cartoon showing a scientist finishing a very long equation on a blackboard, and arriving at the final line "then magic happens".

We are, at times, certainly awed by some things that do not yet have an explanation, as we should be as sentient beings, but "magic" is not (IMHO) an appropriate viewpoint. It can even be a dangerous concept under some circumstances. None of us know "all of science", but we don't have to fill in the blanks with magic.

Apologies. End of sermon.

This one:A4A73E6A-8729-462E-911E-9102EBFA17F1.jpeg

I still cherish the memory of a builder’s coffee talk at WILS where the following luminaries shared what makes a good guitar. (Linda Manzer, Dick Boak, Roger Sadowsky, Michael Gurian, Richard Hoover and Chuck Emerson) Thread here:https://www.acousticguitarforum.com/...d.php?t=526482

There is magic, and then there isn’t, because a good luthier knows how to repeat the process. A good luthier can coax sound out of a pile of mystery wood, or even carbon fiber. My point here, there must be some type of science involved because it’s a repeatable process. It’s just that our mathematical models are not up to the task. And I’ll facetiously add, is it really worth going to the trouble of creating a mathematical model of what we already know?

Rick

[zmf]

Originally Posted by srick

This one:Attachment 51160

My point here, there must be some type of science involved because it’s a repeatable process. It’s just that our mathematical models are not up to the task.

Yeah! Thanks. That's it. My memory was a bit off.

[Italuke]

Originally Posted by SkipII

Methinks we are running out of topics for this forum, or some are suffering cabin fever from being inside during the pandemic.

Yes. I was looking for the popcorn emoji...

[alnico5]

Originally Posted by zmf

Since you emphasize the importance of precise terminology, perhaps you could start by defining "magic".

"Magic" is an event outside the laws of physics.

[zmf]

Originally Posted by alnico5

"Magic" is an event outside the laws of physics.

Could you expand on that thought a bit? Hard to tell which you're going.

[KevinH]

IMHO, I don't think there is any magic involved in the guitar itself. If we knew all of the elastic (and viscoelastic) properties of all the components, the details of their attachments, and created a sufficiently robust model, one could predict the response of the guitar to a plucked string - the frequency spectrum, the amplitude and all that.

There is something that might be described as the magic that a really good player can bring from an instrument. And, as someone mentioned earlier, there is a similar wonder in how a skilled luthier can transform a pile of lumber into a responsive colorful instrument. I know they are applying years of experience and practical rules they have learned, and there is nothing magical about it, but it can seem that way.

Coming from a background in science (physics/astronomy), it seems to me that the most non-scientific part of the whole effort to understand guitars is in the human ears, and what lies between them. Many times there seems to be no predictability in what one person finds to be a nice sounding guitar vs another. There are clearly people that prefer the "Taylor sound" vs the "Martin sound" and vice versa. There is no quantitative measure of what makes an exceptional sounding guitar. It's some mix of frequencies, amplitudes, attacks, sustains, harmonics, etc. Some like one mix, others like another. So, it's useful and certainly fun to understand the physics of guitars - because you may find ways to make a guitar more responsive, for example, which most people would think is a good thing. But beyond that, it seems a difficult struggle. Maybe that's a good thing. Because if we understood and were able to model every detail of a guitar, they'd become like appliances. What fun would there be in that?

[Fatfinger McGee]

Originally Posted by srick

There is magic, and then there isn’t, because a good luthier knows how to repeat the process. A good luthier can coax sound out of a pile of mystery wood, or even carbon fiber. My point here, there must be some type of science involved because it’s a repeatable process. It’s just that our mathematical models are not up to the task. And I’ll facetiously add, is it really worth going to the trouble of creating a mathematical model of what we already know?

Rick

You know, I have to admit I thought this thread was a profound exercise in navel gazing, but this is an excellent question, even if posed facetiously.

There’s absolutely science involved, it’s just pure applied science. In other words, it’s a craft. Discovered and applied empirically, not derived from theoretical science. One doesn’t have to know a wave form from a stadium wave to build a great guitar.

Basically, a guitar is a complex system, and a great luthier is co-optimizing a different set of constraints in the system every time, informed by the wood, desired sonic profile, etc. Much of it they can explain, but it’s ultimately a synthesis of deep experience. I suspect that the best ones are working from intuition at some level, which is really just our brains showing off.

Is there value in creating a mathematical model for an applied science? H-e-double-hockey-sticks yes. That’s the whole basis for AI, turning complex unsolvable problems into problems a computer can solve. Look at the transformations in search capability, natural language processing, traffic and weather modeling, etc. The computer STILL doesn’t necessarily understand the science, but it can find the connections that we can only intuit. If you think Chinese-made guitars are good now, just wait until someone trains a Luthier AI model to oversee design.

[Goodallboy]

Originally Posted by KevinH

IMHO, I don't think there is any magic involved in the guitar itself. If we knew all of the elastic (and viscoelastic) properties of all the components, the details of their attachments, and created a sufficiently robust model, one could predict the response of the guitar to a plucked string - the frequency spectrum, the amplitude and all that.

There is something that might be described as the magic that a really good player can bring from an instrument. And, as someone mentioned earlier, there is a similar wonder in how a skilled luthier can transform a pile of lumber into a responsive colorful instrument. I know they are applying years of experience and practical rules they have learned, and there is nothing magical about it, but it can seem that way.

Coming from a background in science (physics/astronomy), it seems to me that the most non-scientific part of the whole effort to understand guitars is in the human ears, and what lies between them. Many times there seems to be no predictability in what one person finds to be a nice sounding guitar vs another. There are clearly people that prefer the "Taylor sound" vs the "Martin sound" and vice versa. There is no quantitative measure of what makes an exceptional sounding guitar. It's some mix of frequencies, amplitudes, attacks, sustains, harmonics, etc. Some like one mix, others like another. So, it's useful and certainly fun to understand the physics of guitars - because you may find ways to make a guitar more responsive, for example, which most people would think is a good thing. But beyond that, it seems a difficult struggle. Maybe that's a good thing. Because if we understood and were able to model every detail of a guitar, they'd become like appliances. What fund would there be in that?

I enjoyed reading this take on it.

It's a really good explanation of the differences in what we hear. Since we're constantly talking about what we hear, knowing that one person hears something differently than another, to the degree of liking or disliking, should make a lot of issues regarding guitars (in our case) more clear.

[birdsong]

I agree that any magic is in our heads, as in, our ears. 10 of us can taste the same wine and have a different reaction. Similar for guitars. Our senses are complex chemical processes and while we know a lot about them there's a lot we still don't understand. And subtle chemistry differences can yield dramatically different responses.

I've lost some significant hearing at certain frequencies. But I can still hear and enjoy my guitars and live music. I'd love to be able to go back in time and get my old ears, and be able to compare what I heard then to what I hear now. What I hear now isn't necessarily 'low quality' - it's different quality. There are many times when I am glad that I cannot hear as well........ other times it's a little depressing.

So go ahead and produce all the "same" guitars you can - people will hear them differently.

As a scientist, I think it's also kind of magical that we don't know all there is to know about the human ear, or the complex sound wave interactions from a guitar.

John

[Cecil6243]

Odd I was just thinking about this subject yesterday and here it's a thread.

[ssynhorst]

So if magic is a phenomenon for which we do not have an explanation, well, that covers a lot of territory in the world of music. That is the magic of it. And the best part. - Stevo

[Llewlyn]

The physics of a guitar is not a phenomenon that transcends physical understanding. On the contrary, it relies on rather old and simple principles which were known since early 1900. Simulating acoustic wave propagating from an instrument is as hard as simulating any fluid dynamic system (eg a turbine), and can be done nowadays with any laptop. Comparing acoustic vs quantum mechanic is laughable: quantum physics is not only immensely harder, it relies on fundamental principles which we still don't fully understand (--> wave function collapse in relation to the observer).

The above does not mean that physics is able to predict which guitar you'd like to buy, or why I like my Martin certain days more than my SC. That is "the magic".

Ll.

[Cuki79]

For me "magic" is in the builder who choose to dedicate his life to his craft, to assemble pieces of wood and metal to make an instrument. The magic is when a person takes this instrument and expresses his/her deepest feeling by interacting with it. The magic is when one person in the audience connects to the music.

The rest is just matter.