Official Luthiers Forum!

Andrew Wright wrote:
"I'm all for scientific analysis and the scientific method, but realistically it's impossible to utilize when it comes to making guitars. There are two major problems, as I see it."

As somebody who has been dilligently using science to figure out guitars for a long time now, I must beg to differ, at least to some extent. The scientific approach is difficult, for both of the reasons you adduced, but it is not impossible. It's also getting easier (well, less difficult) all the time.

"1. For conclusions to be valid and provable, you have to isolate a single variable. [snip]"

There are a couple of ways to isolate variables. One is to cheat: do your work on a computer model. Howard Wright did a first-rate study that way at the Universtity of Wales/Cardiff back in '96, iirc. His simplified model, which included the strings, the first four or five top resonances, the 'main air' resonance and the 'main back' mode, produced a surprisingly convincing classical guitar sound. It also allowed him to vary anything, such as the mass of the 'main top' mode, independantly of everything else, produce the tone, and then use it in a classic pschycoacoustics experiment, to deduce the most important parameters determining the tone of the guitar. There's a lot you can disagree with in his study, but the fact is that the results he produced do agree quite well with the 'voice of experience', helped to make sense of some things that had been puzzling , and pointed to useful methods of improvement. That's good science within it's limits, as far as I can see.

The main objection to Wright's method is that it loses a lot of the complexity of the 'real' guitar. One way around that is the method I used to investigate the function of 'sound ports' in the sides: I built a guitar with twenty holes in the right side that could be plugged with corks. It is easy and quick to open or close holes in any location and immediately hear the result. Since the only change is a few corks in the sides, more or less, there is no measureable effect from that, and all, or almost all, of the resulting change in sound has to be from the side holes or absence thereof. Both objective and subjective testing can be done easily and the tests are quite reproducible. I'm planning on using that same 'test mule' instrument to check out several other changes with suitable further modifications.

When all else fails we can fall back on statistics. We may not have an adequate model that accounts for the widespread preference for Brazilian rosewood in backs and sides, but it's hard to argue with three hundred years of market data that shows how much more poeple are willing to pay for it. Of course, this sort of data suggests more controlled studies, of things like comparitive wood properties and the way guitars produce tone, that might account for the prefference, and a lot of those studies will neccesarilly be statistical too. The results might not allow you to predict what any specific guitar will sound like, but, as has been said: "The race is not always to the swift, nor the battle to the strong, but that's where the smart money goes."

"2. You can measure forces, but you can't measure tone. [snip]"

Ah, but it's frequently possible to correlate measurements of forces or spectral response with subjective impressions of tone. Meyer, for example, found that guitars that had a tall, narrow 'main top' resonant peak in the output spectrum tend to sound 'harsh', 'thin' or 'cutting'. Knowing this you can begin to look for the structural and acoustic causes of tall, narrow spectral peaks, and figure out ways to control the response. The judgement of what's 'better' is always subjective, but when objective correlates can be found the pursuit of 'better' instrument tone becomes more systematic and rational. One gains a measure of control.   

With cheap computer power and the widespread availability of good test equipment, it's getting harder and harder not to waste a ton of time on scientific investigation of guitar tone. Make no mistake, the two most time-consuming things in the world are Art and Science. It is also certainly true that musical instruments gain a lot of their charm from complexity, and that's a further complication in rational study. Still, if you view it as 'job security' that complexity is sort of nice, or would be if I could figure out how to get _paid_ for all that work! ;) Of course, I can justify it to my wife by saying it will help me build better guitars, and that's true, sort of.... Anyway, despite the difficulties, it has been productive, and fun too.   

[QUOTE=Mario] The view once was that the world was flat AND had a hard stop that one would fall off from. Someone thought differently

Ah, good analogy! 'Ol Christopher also had to prove it before he was taken seriously. Had he not proven it, he would not have gone down in the history books as the one who proved it, right? Right![/QUOTE]


It probably wouldn't be productive to the discussion (which is great BTW) to point out that ancient civilizations KNEW the earth was (mostly) spherical by mathematics, or that Columbus was not the first European to step ashore in the new world. Would it?

Brock Poling38747.6488888889

_________________
Brock Poling
Columbus, Ohio
http://www.polingguitars.com

And is is pretty well documented that the Norse peoples and the Scots (and possibly others) pre-dated Columbus by quite a long period of time.

_________________
Brock Poling
Columbus, Ohio
http://www.polingguitars.com

Christopher Columbus probably stopped by the Magdalen islands where my dad was born! I'm pretty sure he stopped for a quick pee on one of the tiny islands but heck, just can't prove it! wasn't there!

Serge

I wondered what you were finding to talk about on bridge shapes

With the risk of being accused of sitting on the fence I would add this.

Science is great, but I often look at whats presented as sciance and think by the time you have added all the assumptions it is no better than educated guess work.

My favourite way with guitar making is to try and see of course I prejudice it with my opinion and that of my customers of what is the "right sound".

At the end of the day if Scott improves an instrument from the prespective of the person asking him to do the work then to me his theories work, and you have got to admire anyone who can do that kind of work through the soundhole

I would be glad to have Scott work on one of my instruments and if he improved it, I would be interested in what he did to achieve it, cause I do know for sure, I don't know all there is to know, far from it.

Lets face it this idea that some how through science we can find the perfect instrument that everyone will love just isn't going to happen, as they say beauty is in the eye of the beholder, and as my Dad used to say don't knock it, until you tried it.

no post Scott van Linge38749.6743634259

[QUOTE=Scott van Linge]
It's good to hear from a bona fide scientific luthier (I say this in all sincerety). But I'm wondering what Andrew Wright's work would say about bringing my hypothetical quiet B string up to full volume?
[/QUOTE]


I believe you mean Howard Wright:

[QUOTE=Alan Carruth] Howard Wright did a first-rate study that way at the Universtity of Wales/Cardiff back in '96, iirc. [/QUOTE]

My contributions to science to this point consist of a couple above average science fair projects in elementary school (how fast does your trash decompose and in what conditions does bread grow mold fastest).

( :

I take that back. I've also conducted a number of intensive experiments in which I've conclusively proved that my playing can make even the finest of guitars sound pretty lousy.

ANDREWWright
Managua, Nicaragua

no postScott van Linge38749.6754513889

At the end of the day if Scott improves an instrument from the prespective of the person asking him to do the work then to me his theories work

No, the theories don't necessarily work. The methods work, yes. Since my first post, I have said so. There's no denying that. Lighten this and that and yes, of course, things will wake up. Especially so if a lot of your work is on overbuilt factory instruments, ala Taylors. I have playede Scott's Gibson(at the ASIA gathering in Nashville, 2000), and it was a fine guitar, not a killer, but a better than avergae Gibbie. No denying that his work made it a batter guitar.

There is nobody disputing that at all; never has.

It's the theory and thinking as to why the changes take place that are too hoaky to let slide. Of course if we reach inside a guitar and sand this and carve that, it will wake up. Remove half the mass of any bridge by re-shaping it to any shape, and of course it will be much brighter.

It's the pseudo-science voodoo, energy-flows-in-circles and "I stood next to Garcia and felt that postive energy, man", and whatnot, that is too hoaky and that we would like to see quantified. Not the results; the results are real, but what caused them? Could a bridge of the same footprint and weight have soudned the same? He doesn't know, because he hasn't tried. He rounded one up, and the guitar improved. Holy cow! A revelation? Maybe, but maybe it's just the weight. Maybe it didn't need to be round after all.

Note that where we give solid historical examples that disprove a theory, he immediately drops that subject, and starts in a new direction. Don't y'all know snake oil when yuo step in it?

Arghhhhhh! <bg>

I hate pseudo-science. There's a guy who lives nearby who claims he has seen a government rain-making machine in the bush behind his place. Says he saw it with his own eyes. Yessir, up here in the swampy northern bush, where the last thing we need is more rain, the givernment has a rain making machine to ruin his summer. Yup... Every time it rains for more than 3 days in a row, it's the damned government making it rain. No amount of disproving it to him can change his mind, and he does the same play on everything that we see here; with every solid example to disprove him, he drops that line and goes into a new one. Ask him to show you the machine, and there's always an excuse at the ready. Not a second's hesitation; the excuse is right on the tip of his tongue. It's like these guys plan ahead of us, knowing we'll catch on to the BS, so they have the next line ready beforehand.

. Yessir, we still cannot predict the weather, but we can control it? Sure....

Arghhhh! <g>

Scott

If I am ever out your way I would be very interested to see your work.

Mario

I understand what you are saying but is your argument not that because he has not tried one element then all of the theory is wrong.

My view is just because it is beyond my experience doesn't make it wrong. At the end of the day if it is simply a way of visualising how the sound patterns move, which enables the production of a high quality end result, then the technique and theory remain valid in my eyes.

Scott,

I sanded the bridge down as you suggested, sure enough the treble that I mentioned had lessened earlier in my sanding reappeared in sweet fashion. All of these changes are small, but detectable, seemed to work for me. Also noticed that the bass sharpened up a bit as well, just a tad clearer and a bit more robust.

Thanks,
Greg

_________________
Gwaltney Guitars

[QUOTE=RussellR]
I understand what you are saying but is your argument not that because he has not tried one element then all of the theory is wrong.

My view is just because it is beyond my experience doesn't make it wrong. At the end of the day if it is simply a way of visualising how the sound patterns move, which enables the production of a high quality end result, then the technique and theory remain valid in my eyes.
[/QUOTE]

Erm, no.

I think what Mario's argument is is that the theory is nothing more than untested hypothesis. In terms of everyday use of the word theory, yes, it's a 'theory'. In terms of the scientific use of the word, it's nowhere near that. It's a mental model that doesn't have solid empirical (let alone experimental) evidence to back the claim that it works because of that explanation.

The technique remains valid if the results are there. No argument from either camp, methinks. The 'theory' is little more than a somewhat fanciful explanation for why it works, dreamt up, and not tested in any meaningful way, and has no intrinsic validity itself. It's an opinion. Not tested fact, and certainly nothing so well-founded as scientific theory (and yes, I'm neatly avoiding the fact that it's difficult to impossible to objectively define what a 'good' guitar sounds like, but there is broad concensus, or at least a collection of broad concensusses, concesi...whatever). And that's somewhat frustrating to the somewhat more science-oriented of us.

[QUOTE=Mario] I can't fight a religion, and this is what these claims are. If
you believe in them enough, it becomes your religion, with no need to
prove it because you believe. And no amount of others showing examples
of where the thinking is wrong will change the believer in his faith.
[/QUOTE]

"Religion is the opposite of science. Science attempts to make its doctrine
conform to the physical world, but religion attempts to make the physical
world conform to its doctrine."
Jay Hanson

I suggested to Scott in another thread that the best way to illustrate his
theories (or anybody else's for that matter) would be to do Before/After
recordings. If you have even a modest computer, extremely decent
recording gear is available for little money (under $200), we don't need
super high end stuff here, a good static mike and a decent audio interface
would do. Consistency is the key: same room, mike placement & distance
and relative levels and we're done. And that is also a useful tool to
compare different instruments. That's the best way. It's easy and I do it all
the time, I have recordings of all the guitars I've worked on, it gives me a
good tool to compare different body shapes, tonewoods etc. The ear is
much more precise that a mike, but also much more subjective and not
consistent.
Like Mario I think that mass is the primary factor, shape secondary. As for
angles inhibiting trebles I have to smile…
I am not disproving Scott's theories here, I am just extremely skeptical on
the approach and the jargon utilized without any clear proof for anybody
to learn from. If it works, record it and the differences, if that obvious,
will speak for themselves. That being said the journalist George Monbiot
has a very appropriate quote on his website:
"Tell people something they know already, and they will thank you for it.
Tell them something new, and they will hate you for it."

_________________
Laurent Brondel
West Paris, Maine - USA
http://www.laurentbrondel.com/

Before y'all think I'm just trying to be a jark... <g>

Guys, here's what bugged me. Read my first post in this thread. It was a response to where Scott stated that sharp edges rob energy and treble. I thought "huh?", so I answered, and asked why, if my guitars have so many sharp edges, do they still exibit so much power and treble.

Scott never responded. I wasn't trying to shoot him down. He made a statement that was completely, 100% unfounded by my experience. What I expected was that he'd come back and give a plausible explanation for why, maybe, this is. But he chose to ignor it completely, because it seems to show his thinking is flawed, I guess.

that lit me up.

Again, he made a string statement that the X braces don't like anything touching them. I see this, and I gave a historical example of great guitars, the ones held as the gold standard of steel string guitars by most players. And these have shapr edges all over, also. Again, he ignores my response, and pretends nobody noticed. What i expect is that when someone shows thousands of examples that show the opposite of the his thinking is, he'd give us a reason for why it can be. If he wants to teach, he must first learn. He must first look at the other side, and see if maybe his idea doesn't hold water. With the two above examples presented, he should stop and re-think why his work works, because it clearly shows the thinking is flawed.

If you tell me farmer Brown's barn was blue, and I say it was red, and you maintain that it was blue even after farmer Brown shows us photos of the red barn to prove it was red, I'll get a little hot. Same here....

[QUOTE=Mattia Valente]
I think what Mario's argument is is that the theory is nothing more than untested hypothesis. In terms of everyday use of the word theory, yes, it's a 'theory'. In terms of the scientific use of the word, it's nowhere near that. It's a mental model that doesn't have solid empirical (let alone experimental) evidence to back the claim that it works because of that explanation.
[/QUOTE]

Mattia,

I think you are doing Scott a bit of a disservice. My understanding is that his theory of sound rings comes from him "feeling" these rings on the tops of guitars with his fingers as he builds/modifies guitars, and he has an initial mental model in cones and the resonator guitar example he speaks of. This to me is the start of an "empirical" exploration. It is a theory, that has been neither proved NOR disproved.

I think science and art have valid roles to play in helping people both understand and build musical instruments. I think you have as much chance of building mathematical/theoretical models that describe and predict a wooden guitar's behaviour as you have of building mathematical models that explain and predict the World Economy or people's behaviour (and I speak from over 30 years bitter experience here before getting a "proper job" making guitars). Any system that has so many organic variables - not forgetting the homo sapiens that interact to make them make noise - is going to be impossible to accurately model. As in modelling Economies you can get certain basic principles that help a lot, but then the "people" (or wood in the case of guitars) stubbornly refuse to obey the "assumptions". Maybe with instruments made of more predictable materials like graphite etc this is more possible, but although I welcome the predictability I certainly wouldn't want to spend the rest of my life building those.

Also speaking of Science and Religion, the current academics amongst you please correct me if I'm wrong as its over 30 years since I studied Mathematics at Cambridge University, but nearly all sciences have at their core certain axioms on which everything else is based. These axioms by their nature can be neither proved or disproved, they are in effect "beliefs" and if they fail everything built on them crumbles. Sound familiar?

imvho, the best thing to have is an open and questioning mind and the willingness to take things and experiment/develop them - and also a willingness to share and give back. I am a magpie in this respect and take things from people like Scott and lots of other great builders, inclunding thos on this great Forum, and adapt them for the guitars I want to build. I have said before, I don't really buy into the full "Sound is round" theory but some of Scott's ideas and practices have helped me make better guitars.

If you think the "theory" is bunkum and "Snakeoil" then fine, smile and move on. Unless you have a desire for such theories to be publicly expunged so that others don't come under their "evil influence" it doesn't really matter.

I think threads like this that make people think are fabuluos and endorse a huge bravo to ALL that have particpated.

_________________
Dave White
De Faoite Stringed Instruments
". . . the one thing a machine just can't do is give you character and personalities and sometimes that comes with flaws, but it always comes with humanity" Monty Don talking about hand weaving, "Mastercrafts", Weaving, BBC March 2010

Eloquently put David, here, here.

Kim

I hate to jump in and take sides on this.. but wanted to share a couple of thoughts...

It seems to me that builders fall into two basic (and highly overgeneralized) camps. Those who build from intuition and experience, and those who measure everything and try to decipher meaning from those measurement. Both are valid approaches and can (in time) lead to very good guitars.

At the core of it is a basic hypothesis / test / review / loop process that either improves the instrument or not. In time this data which is either quantitative or qualitative leads to progress (even if it is learning what NOT to do).

However......

I think before making claims and stating them as a fact you should have a methodology and data to support those claims. And.. it would be even better if the results could be independantly verified. (but now I might be dreaming)

It is very hard for the experiential camp to support their claims in a way for the scientific crowd to accept them. So, rather than debate who is right and who is wrong, it might be best to collaborate.

Intuitive builder... "Ya know... I have noticed that when I reduce all the sharp corners I notice a significant increase in the ....... sound of my instruments. I wonder if it is because....."

Scientific builders may then jump in to tell you that this makes sense, doesn't make sense, etc. And a discussion might ensue that creates a model by which to test further ideas.

It seems to me that building walls between the camps is counter-productive for everyone. I know the scientific guys have a level of disdain for the intuitive guys... and to that I would say...

"Imagination is more important than knowledge."
... didn't a slightly famous scientist say that?

And for the intuitive guys...

"Look around... science isn't COMPLETE hooey."

I think there is middle ground here. Far more so than the science and religion metaphor that has been used.Brock Poling38748.4731828704

_________________
Brock Poling
Columbus, Ohio
http://www.polingguitars.com

Greg,

I'm glad your guitar sounds better. Now, for heaven's sake, stop sanding the top!!

Scott
Scott van Linge38749.6769907407

David, That was really well put and I think you have eloquently defined the essence of any 'discussion' versus an argument. Arguments usually require someone to be right and someone to be wrong. Discussions usually include the presumption that everyone's point of view is correct - just different.

That was the reason for my previous 'crap' post - I simply want to hear it all, and see how it works for me.

Mario, I've never heard of keeping all the edges square inside a guitar. Seems counter-intuitive to me but remember I live in a growing, but very un-educated world of beginning luthery. Should I argue that square is bad, or should I discuss the possible merits of your wonderful guitars and design ideas? I prefer the later. That's why I appreciate your participation and want more of it.

Scott, I don't have any real idea what sanding the bridge will do for my guitar but I'm gonna do it. It simply intrigues me and I can't wait to see if I can tell the difference. I even built my version of parabolic braces inspired by your web site and research.

Greg, I LOVE your bridge design but don't have the balls to sand my current one that low with its exising finish but will definetly make an effort on my new guitar. The fact, or theory, or illusion, or whatever you want to call it, that you could hear a difference in the treble by modifying the bridge is very interesting - and a compelling idea that I will put in my brain for processing.

Another genuine thank you to all who have posted their ideas here.

Larry

_________________
Thank You and Best To All

Scott,

Thanks....Yeah I'm laughing at myself too, but I am finished messing with this one, not enough wood left to sand. It sounds awesome right now, I appreciate your input, it has helped.

Thanks,
Greg

_________________
Gwaltney Guitars

Larry,

You think it was low before.....scary now!!

Looks like a UFO or something, BUT, it worked!!

Greg

_________________
Gwaltney Guitars

Hi Larry,

Before you do any sanding on the bridge at all, I hope you will take the time to make a protective guard for the top. I previously described one here I make, using clear mylar, .015" thick, to fit exactly around the bridge, with a poster board bigger complete top cover. Boo boos are never fun. Probably a plastic "for sale" sign would be a good substitute for the mylar. Let us know.

ScottScott van Linge38749.6781481481

Well, considering everyone else that has walked away has come back, I might as well too. Now please, we're all trying to understand instruments better here, so no one be wounded by the rubbish I type, okay? Thanks.

Mattia, you're right on.

Dave, because you asked for an academic's view, here it is.

[QUOTE=Dave White]
Any system that has so many organic variables - not forgetting the homo sapiens that interact to make them make noise - is going to be impossible to accurately model.

[/QUOTE]

All models are wrong (i.e., imperfect), but some are still useful. To expect a model to predict a system perfectly is rarely necessary. If a model helps explain 60% of the variation in a system (say a guitar), is that useful? 25%? 85%? Up to you.

[QUOTE=Dave White]
please correct me if I'm wrong as its over 30 years since I studied Mathematics at Cambridge University, but nearly all sciences have at their core certain axioms on which everything else is based. These axioms by their nature can be neither proved or disproved, they are in effect "beliefs" and if they fail everything built on them crumbles. Sound familiar?

[/QUOTE]

Indeed, that is wrong because it is a misrepresentation of the process. You imply that these axioms have little to no foundation, when in fact their foundation is so well understood and has proven so consistently reliable as to be almost irrefutable. All of these axioms have withstood powerful scrutiny for hundreds of years if not longer, so the likelihood of a discipline collapsing because they are completely wrong is zero. They are a million miles (kilometers?) from belief because beliefs, by definition, have no physical evidence as a foundation.

As for Brock's assessment of the kinds of builders, I tend to agree. But I'll ask how many of you would allow your children to live on the top floor of a large apartment building that was designed and constructed with art as the only contributing discipline? I want some structural engineers at the table!

Sure, I'd like it to be beautiful too, but in my world the structure of a guitar is engineering, not art. The particular adornments, shape of the headstock, etc., etc., well that's art and that's important to me too. Because we cannot or we choose not to explore how or why something works, that does not change its realm from engineering to art.

My 2 cents, likely worth considerably less.