Guitars "opening up" - How to defend the claim?

[KevWind]

Originally Posted by Organic Sounds Select Guitars

Hi friends,

An early guitar video that I made about three years ago, before I opened my shop, was of an all mahogany boutique brand guitar that was about 11 or 12 years old at the time. In the course of the video, I mentioned that the guitar has had "several years to open up", and went on to explain what I liked about its tone.

A couple of days ago, I received the following comment on that video:

"Another idiotic statement about guitars "opening up". THAT IS BEE ESS. Such claims are NEVER reinforced with science."

I typically ignore this kind of comment, but I was in more of a mood to respond this time (maybe the "idiotic" word triggered me a bit!).

Here's what I wrote, and I'm posting it here because I'd love to be corrected or to be advised on a better way to talk about this topic since it does come up periodically. Here was my response:

"The problem with looking for scientific proof here is this: try designing a study. We're talking about changes over years. How do you control all variables? If you're going to measure sound characteristics (not just volume and sustain, but also harmonic complexity, dynamic range, timbre, etc), then you have to control for how the notes are played - the force and duration of the string plucking or strum, the room temperature and humidity, the exact placement of the microphone, etc. Even if you could somehow control all of those variables, the guitar will have different strings on it. If you want to talk science, take a look at wood on its structural level, specifically the hemicellulose and lignins, and how they change as the wood ages and dries. Now add repeated vibration to those wood fibers over the course of years, and see what happens. Last thought: ask any experienced luthier if they think that solid wood guitars change over time - you'll have a difficult time finding any who will say no."

How did I do? I tried to keep it short, since this is a YouTube comment. Is there a better way to explain this succinctly? Is there a great article to reference? Am I totally misguided in my explanation? I'd love to hear from Alan Carruth or another of our wonderful AGF experts on this.

Thanks folks!

On the one hand you did great on the other hand not so much

Let me explain as I see it and perhaps add a bit to what others have already noted .

The statement "open up" is definitely not "idiotic" And yet neither is it factually accurate, what it is - is a subjective phrase (commonly used in the guitar world to imply ( improvement in tone over time ) and is a totally subjective conclusion at best ..
But "open up" is not accurate in terms of "the science" of what is happening. The science can only prove that wood objectively changes over time But as you correctly stated because of the insurmountable logistical issues involved, science cannot prove if that change can actually be detected by human hearing in terms of tonal change and especially cannot prove if that change creates an actual "improvement" in tone or not....

So the problem is if your going to fall back on the science,,, the use of phrase "open up" is not scientific. So while using that phrase is not "idiotic" --- it is also superficial at best ... Perhaps the phrase "Had to time to age " would be more accurate

[ship of fools]

I find what you had to say was as close as possible to explaining it. But I would also like to point out that arrogance and ignorance on his reply is bliss to say the least.
There is no way to show just how woods change over the years to make a difference to how a guitar changes over the years and how the sound could be effected and as you stated their are other variables to point to also.
As for answering well I have found that after almost 60 years all I can do is say and state what I feel and know and if someone disagrees well that's for them to figure out.
So carry on my friend and enjoy the sounds she makes ( or he ) and forget about those who may have a difference of opinion.

[Rpt50]

As a retired scientist, it always drives me crazy when people use phrases like "science proves" or "as proven by science" or any of the other endless variations on those phrases. Science does not "prove" anything. To prove a hypothesis true, you would have to examine every possible instance of it in the universe, which is of course impossible. Knowledge produced by science is always tentative and subject to change. William James likened "scientific truth" as a tentative and pragmatic knowledge of what works in the world, and that pragmatic knowledge is always changing as knowledge is gained.

As for guitars "opening up", I am 100% confident that, with enough funding and time, research could greatly expand our pragmatic knowledge of how the various materials used in the construction of guitars change over time, and how those changes impact the sound produced. But I doubt this would be a topic that taxpayers would find worthy of their tax dollars.

Private musical instrument companies might be more likely to do such research (and probably do), but I think it would likely be more profitable to spend their research dollars understanding how unsupported ideas infect the populations of buyers of their products, and use that knowledge to increase sales. Consider the "mystique" Martin guitars enjoy. How many people purchase a Martin guitar without even considering/trying another brand? Irrationality rules when it comes to discussion of guitars

[jim1960]

Originally Posted by Organic Sounds Select Guitars

Here was my response:

"The problem with looking for scientific proof here is this: try designing a study. We're talking about changes over years. How do you control all variables? If you're going to measure sound characteristics (not just volume and sustain, but also harmonic complexity, dynamic range, timbre, etc), then you have to control for how the notes are played - the force and duration of the string plucking or strum, the room temperature and humidity, the exact placement of the microphone, etc. Even if you could somehow control all of those variables, the guitar will have different strings on it. If you want to talk science, take a look at wood on its structural level, specifically the hemicellulose and lignins, and how they change as the wood ages and dries. Now add repeated vibration to those wood fibers over the course of years, and see what happens. Last thought: ask any experienced luthier if they think that solid wood guitars change over time - you'll have a difficult time finding any who will say no."

I had to read your post several times to convince myself that I had read it correctly the first time.
The reason? Your argument defeated your premise.

Since you're admitting up front that scientific prove would be nearly impossible to attain, you're relying on the fact that wood goes through some changes over time and those change account for some difference in a guitar's tone over time. However, you offer no proof that such changes alter a guitar's tone in any specific way and I'm doubtful that could be scientifically proven either. For all we know, those changes in the wood result in tonal changes so imperceptible that they're not worth mentioning.

All of the arguments which support "opening up" have, in my opinion, one thing in common: confirmation bias. People believe they hear it because they want to believe they hear it. In other words, they've bought into the premise so, when they listen to their guitars, they confirm the premise. However, they have no way of isolating and eliminating the effects of variables that we can, with certainty, know affect the sound of a guitar. Invoking "any experienced luthier" is a logical fallacy since no luthier is any more able to offer actual proof than anyone else. This appeal to authority falls flat for that reason.

Now it may be true that guitars "open up." Just as one side has no proof that they do, they other side has no proof that they don't. I'd argue, however, that it falls on those making the claim to provide the proof and you've pretty much admitted the proof is unobtainable.

In the end though, this is why the argument for "opening up" fails for me... in addition to the many variables that result in tonal change, the notion that someone can accurately recall the tone of any instrument across long stretches of time seems quite incredible and the player has no way of accounting for (and eliminating) an important variable that rarely gets mentioned: the player's progress. To anyone telling me their guitar sounds different now than it did two years ago, my first thought is going to be "Aren't you a better player than you were two years ago?"

In the end, people can believe what they want to believe. "Opening up" is, for all intensive porpoises (forgive me), unfalsifiable, which makes it a better argument than the argument for a flat Earth, so you have that going for you. I don't buy into it because I've never seen any good reason to buy into it.

[FanoFan]

I'm new to the acoustic world and always wondered about the opening up thing myself. Last week I played a new J45 (2 actually) and my friend's 1941 J45, the difference is resonance was amazing. But I don't know how much of it has to to with changes in how the J45 is made today versus the early 40's. But man they sounded like two totally different guitars.

[Alan Carruth]

How did this thread get so long without my seeing it!

Earl 49 wrote:
"A true study would take years of careful double-blind testing with top-notch analyzers and microphones in a lab and would easily qualify for a PhD dissertation in the physics of sound. It would be a $200K academic study minimum, not counting the guitars themselves."

I hope it would not take as much effort as Earl implies, but it's certainly not a slam dunk either.

Brent Hutto wrote:
"This topic is in a large category of points literally not worth arguing about because there is no objective "truth" involved."

That's a circular argument. You're starting out by assuming the point you want to prove: that there is no physical change in the guitar over time or with playing.

To Earl's point:
We need to ask who would benefit enough from doing such a study that they'd want to fund it. If you're making a living buying and selling vintage guitars on commission you've got a bit of a stake in it: you can buy this pre-war Martin for big bucks (and pay me x% of that) or you can buy a new one and hope that it starts to sound like that before you die. These guys have no real incentive to fund a study when they can rely on widely believed hearsay to make their point. A real study might prove them wrong, which is the last thing they want.

Manufacturers sell what they make, not what it's going to become.

Individual luthiers have some interest, if it will help them make something that sounds 'played in' off the bench without 'playing out' (or folding up) too quickly later. Some of us do a little research as we can, but we sure can't finance big studies and fancy equipment.

Most of the better research on guitar acoustics happens in mechanical engineering departments in universities, where it's an interesting way for students to learn how to do the measurements without treading on industrial toes. They can muster the equipment if somebody is interested enough.

I've never met a luthier who doesn't believe that guitars change with age and playing. As has been pointed out, much of the change happens in the first hours and days, an very few players get to work with such new instruments on a regular basis. We can disCUSS what might be changing, but that's another matter, and gets us back to figuring out what we need to do to measure whatever it is.

We need to distinguish the sorts of studies necessary to find any objective changes and those that focus on subjective ones. In the first case we are measuring guitars, and in the second we're looking at people's reactions to them. In principle measuring guitars is relatively straightforward. What's required is a reasonable degree of control over how you drive them, and how you record the output. 'Impulse' spectra can give a lot of information quite simply, even when the setup is not ideal. You do, of course, need to understand the limits of the measurements you're making. In theory consistent changes on a fairly small number of instruments with similar histories, even if those are short, could suffice to establish that something is happening, and constrain the parameters of the change.

Subjective changes are much harder to nail down. That's where you get into the need for large samples and statistical workups. It's something of an axiom in the business that there are things that are easy to measure and hard to hear, and things that are hard to measure and easy to hear. If the objective changes happen in the latter ways it can be a long slog.

I got a good example of that in the 'matched pair' experiment I did a few years ago. I made two OM style guitars from matched sets of Red spruce and mahogany. I used the tightest quality control I could at every stage, from cutting the wood through thicknessing, bracing, 'tuning' the parts, assembly finishing and setup. The 'impulse' spectra were identical within measurement limits up to 1000 Hz, which is about as far as you can directly control them. When they were completed we ran 'blind' listening tests with 50 listeners over several sessions, and they found it as easy to distinguish the two as to tell either of them from a similar 'ringer'. I'll note that many listeners commented on how similar the pair were: the ringers were less so. There was no overall 'quality' difference between the paired guitars: they were both perceived as good, but just different.

The differences were primarily in the range of 2-4 kHz, where hearing is quite acute, and the guitar is in a 'resonance continuum': that is, there are so many resonances coupling with each other that the output is at least somewhat chaotic. It's impossible to predict or control the exact spectrum except at the level of the number of peaks per octave (but not their pitches), and the difference in level between peaks and dips (which is a measure of overall damping).

We did try a small 'playing in' experiment with the two, wherein a very good player kept them both in the same room for three months, playing one for an hour or more every day, while the other remained in it's case. At the end of the time the impulse spectra were taken again, and there was a difference. The played instrument had somewhat more output in the range around the 'main top' resonance, near the open G string pitch. This is consistent with other measurements I've made. Given the fact that the two did not sound 'identical' to begin with we didn't do any subjective 'blind' tests afterward, since the most reliable measure is 'same/different'.

IMO it seems that the evidence we have is mostly 'preliminary' in nature, but the preponderance suggests that there may well be changes with playing. Why would the guitar be the only machine that doesn't wear out/in with use? We still don't have enough data to say what's changing, although wood/glue/finish are the obvious choices. Given the 'abundance' of funding available, and the nature of the problem, I don't expect this to be settled any time soon, and it may never be to the satisfaction of some. One of the reasons behind the 'matched pair' experiments is to try to make guitars that sound 'the same' off the bench, so that we can see if one plays in by direct comparison. It seems that's not possible. ;(

[Brent Hutto]

Originally Posted by Alan Carruth

Brent Hutto wrote:
"This topic is in a large category of points literally not worth arguing about because there is no objective "truth" involved."

That's a circular argument. You're starting out by assuming the point you want to prove: that there is no physical change in the guitar over time or with playing.

I don't assume or state anything about physical changes. My only assumption is that it's not possible for me to ever say what another does or doesn't think he hears. When a statement is made about an individual's subjective impression,
i.e. "I have heard many guitars open up with time..." there's no way to say what's going on in his ears and mind and it's useless to argue for or against his impression.

[Mycroft]

Originally Posted by Rpt50

As a retired scientist...
As for guitars "opening up", I am 100% confident that, with enough funding and time, research could greatly expand our pragmatic knowledge of how the various materials used in the construction of guitars change over time, and how those changes impact the sound produced. But I doubt this would be a topic that taxpayers would find worthy of their tax dollars.

Haven't ever follow the Ig Nobel awards, have you?

Originally Posted by Rpt50

Irrationality rules when it comes to discussion of guitars

That is not an entirely rational statement. <G>

[ssjk]

Hey bfm612 - excellent Jerry Lundegaard reference.

[Mycroft]

When angels dance on the head of a pin, are they doing a waltz or a tango?

The reason this is one of those unanswered question is because of the nature of the changes, if any. The inputs if you will.
First is the changes in the structure of the wood itself (and to a lesser degree, glue). That is actually the more measurable of the two, as we can compare newly cut old growth to wood that was logged 100 years ago or more. Even that is imperfect given that variability of an organic material. However, if the sample sizes were large enough it might counter that. So we do have some ability to measure how wood changes over time.

The other input is harder to measure, and that is the effect that vibration has on the structure of wood. I suppose that we can look at the structure of the wood in old guitars and compare them to the structure in new guitars, but then, how much is it vibration and how much it it simple aging?

Any study besides comparing currently new and old guitars would require the study be conducted over time, with the guitars kept in identical environments for the term of the study, but also to have identical playing time. Would 20 years be enough? 100?

The other question is: if we grant or prove that guitars structure changes over time, through either or both inputs, are those changes audible in the instrument's tone to the human ear. And if true, are they a perceived improvement or not?

Segovia, I believe, said that he thought that after a while, guitars "wore out." Was he perceiving changes in the guitar's tone over time, and did not like it? Either way, he thought that guitars changed over time.

Enough tilting at windmills. I have my own opinion on the subject, but it is based on personal experience, and so not transferable.

OMMY

[nostatic]

If I believe it opens up, and that affects my playing, then it is true. The whole "do a double blind study" or RCT about artistic/performance things is kinda nonsense. And I say that as a trained scientist.

You cannot remove human bias from the performance, so the "facts" become mostly irrelevant. It is about the relationship between performer and instrument, and that has so many intangibles *maybe* there is a benefit to optimizing details with "science" but maybe (probably) not.

[Mycroft]

As an additional thought, if a guitars sound does not change over time due to the aging process and vibration from playing, then the folks who developed the Toneright and Torrifaction processes sure put a lot of time and effort into nothing.

[jim1960]

Originally Posted by Mycroft

As an additional thought, if a guitars sound does not change over time due to the aging process and vibration from playing, then the folks who developed the Toneright and Torrifaction processes sure put a lot of time and effort into nothing.

Lots of people have made money selling placebos for all sorts of things.

[Mycroft]

Originally Posted by jim1960

Lots of people have made money selling placebos for all sorts of things.

Its good to treat your fretboard with snake oil.

[Russ C]

Originally Posted by Brent Hutto

Reminds me of the Paul Simon lyric, "A man hears what he wants to hear and disregards the rest".

or more accurately - some hear what they want to hear and disregard the rest,
some do not permit self deceit and some do both.