Official Luthiers Forum!

I've followed a number of threads over the months here, and have observed:
1) that in describing how different guitars "sound", we're drawn to using various adjectives like "warm", "woody", "dry", "ringing", etc. And yet none of these descriptions is precise in what it means or how it sounds. In other words, none of these descriptions of sound has an objective definition, when applied to a guitars sound.
2) many have expressed frustration with the lack of precision that these terms convey. I know I'm not the first to comment on this - and more experienced luthiers have discussed this more eloquently. But it feels to me like we can do better.

Open question/challenge. A "glossary" of terms with definitions for qualities of sounds related to guitars, so that we might have a common vernacular when describing the sound that a guitar makes. One which is precise enough that it can be "translated" objectively. If so, I'd love to see it - if not, can we agree on one?

I would think that certain types of descriptions would be more precise in this application. (I've never understood what was meant by "warm", but "ringing" seems to be a guitar with long sustain on the trebles).

One possibility is that the descriptive qualities of a guitar might depend on each of, or the combination of a set of variables which might be objectively measured, like a) sustain, b) balance ,c) volume (ditto), and ......????

Thoughts?

This is the biggest can of worms in the business.

Somebody on another list talked once about a study that was done on violin sound a few years ago. They polled people at a music school, iirc, for terms that would describe the sound of a violin, and narrowed that list down to about 70. Then they made good recordings of the same short snippet of music, played on various violins, and played them back through earphones in random pairs, to lots of people. The listeners were asked to rate the instruments in terms of the various descriptors: which one is more 'nasal', or whatever. There were some interesting findings.

One was that everybody seems to use the terms consistently, but they all mean something different. If fiddle 'A' was more 'nasal' to you then 'B', and 'B' more nasal than 'C', you'd always hear 'A' as more nasal than 'C'. However, your 'nasal' might be my 'bright'.

More interesting than that was that when an instrument had some characteristic that a person didn't like, they were generally _unable_ to rate it on other characteristics. If you don't like 'nasal' fiddles, then you'd have trouble saying whether 'A' was more or less 'even' than 'C', for example.

This sort of thing has long been an impass in many fields that rely on subjective impressions to form ratings. What has to happen is that somebody simply defines some terms, and strongarms everybody into agreeing on them. I think this has been done with wines, for example: 'fruity' has a specific meaning, and every serious wine taster knows what that is. Maybe when I get to be Undisputed Master of the World I can do this, right after requiring mufflers on Harleys and forcing Ovation to stop using Tupperware and start building real guitars.

Until then, you're just going to have to pester your customers until you understand what _they_ mean when they say 'clear'. I had one last year that wanted a 'clear' sounding guitar, and it was only after making it that I realized that what he meant was what I'd call 'dead'. Everybody likes the guitar I made him, except him.

My wife has made a 30 year career of just this sort of thing, only with respect to food. As has been suggested, rating sound is subjective and suffers from a gap between the ability of humans to "measure" sound qualities and express those qualities with words, which by nature are extremely inaccurate, because we each have our own definitions of what those words mean.

The way evaluations are done scientifically in these instances is to train a large panel of sensitive "instruments" (people) to express the same sensory experience with the same words. Not an easy task.

The science comes in when the evaluator uses a statistical analysis using a large number of "measurements" to determine and quantify the differences between "samples."

_________________
"Building guitars looks hard, but it's actually much harder than it looks." Tom Buck

I have been wondering for a while what is meant by the term "fundamental?" As in, this wood exhibits a strong fundamental.

There's this whole field of study in psychology called psychoacoustics which is about studying the subjective human perception of sound. There are many industrial applications (what's the ideal 'plop' of a trunk, what's makes an engine sound 'sporty' vs. 'luxurious', and so on. Those industries obviously have more money to study this kind of stuff than luthiers

There are a couple of software programs hat can analyze a sound clip on a range of psychoacoustic parameters, like loudness, sharpness, roughness, ... so you theoretically could use that vocabulary to describe the sound of a guitar based on some measureable dimensions.
Most customers would probably think you're crazy if you tell them that you just made them a sort of rough, but not sharp sounding guitar

I guess given enough funds, you could also come up with your own terms.

There are research measures like multidimensional scaling... you ask a couple of people to give "dissimilarity ratings" - you basically play the same sound bit played on two guitars and ask how dissimilar the two are on a scale from 1-10. You do that with a large number of guitars and a large number of participants. What you can come up with perceptual maps, indicating clusters or groups of guitars that have common attributes.

You could then ask a bunch of people what the guitars in that group have in common...if most agree on a term (warmth), or you can just assign labels yourself. You can then try to find measureable sound characteristics that define e.g., warmth (like emphasis on a certain frequency range, and so on).

All that's left is that you share your findings, everyone in the world agrees on your definitions, and the problem is solved...no more ambiguous labels

cheers, Christian

Randolph - When you play a note, it's composed of the fundamental and a series of harmonics. I think what people usually mean is that notes are fairly pure - not many audible overtones or harmonics. Look up 'musical acoustics' in Wikipedia for more (and better explained) information.

Cheers,
Bri

In the 1970s I lived in Napa Valley and just about everyone I knew worked in a restaurant or a winery. There was lots of discussions among my friends, similar to this, to standardize terminology to describe the taste of wine. I guess something came of it, but I'm not sure it was so useful, since some wineries ended up making wines that would fit the "standard" descriptors, like fruity, as Alan mentioned. Descriptors ended up driving what the winemakers were producing, creating a lot of generic wines that those less experienced with fine wines could hang a term on. Seemed like it sort of dummied down parts of the market.

This site might give us some ideas as to where this might lead.

http://zebra.sc.edu/smell/wine_glossary.html

Is this what we want? Just posing the question.

Pat

_________________
formerly known around here as burbank
_________________

http://www.patfosterguitars.com

Hey Brian, thanks for the definition and link. Wikipedia does describe the term "fundamental" as you do - the first and most basic note void of the impending harmonics. It apparently also has the shortest frequency (distance between waveform peaks). The harmonic notes following the fundamental each have successively larger repeating units of signal. Definately helps me get a picture. And now that I know that everybody elses picture will be interpreted differently than mine, I can not worry about it too much.

Alan,

I can imagine what a nightmare it can be trying to have clients/prospective clients describe what it is they are looking for in a guitar. Especially when many people tend to be a bit imprecise in their vocabulary.

What do you think of the terms posted by Todd - from Taylor's site? Thanks, Todd! They seem to me to be useful, and appear to bring a bit more clarity to some of these descriptors. It seems to me that if we all used a similar set of definitions - and confirmed that this is what we mean by "warm", etc. it might be a useful exercise.

What do you think?

Randolph wrote:
"Wikipedia does describe the term "fundamental" as you do - the first and most basic note void of the impending harmonics. It apparently also has the shortest frequency (distance between waveform peaks). The harmonic notes following the fundamental each have successively larger repeating units of signal."

Any steady sound can be deconstructed into a series of 'pure' sine wave signals, at various amplitudes and with different phase relationships (they 'zero' at different times). The 'fundamental' is the lowest pitched member of the family, and has the _longest_ wave length and period (time between zero values).

One of the things that bugs me about people describing 'tap tones' is that they'll tell you to 'shave the bracing until the fundamental goes away'. There's _always_ going to be a lowest frequency in the mix! Maybe they mean 'until it's too low in pitch to hear'? If that's what they mean, they should say that. *sigh*

Those terms from the Taylor site are interesting, but some of them don't fit with my own perceptions and measurements. For exmple, sometimes a guitar with a lot of high frequency energy in the signal will sound 'full' or 'warm' rather than 'bright'. Your ears and brain will take the high-end stuff and, if it's pretty close to being harmonics of some lower partial, assign the energy to the lower partial. I've noticed that mahogany guitars can have much less high-end energy in the signal than rosewood, but may sound 'bright' anyway. Perhaps it's the 'punch' of the starting transient driven by that light weight back? High-mid-low balance is just one part of 'tone', and maybe even not the most important.

I used to drink wine, but I quit because I don't know enough adjectives. I'm starting to think the same about guitars.

The strangest description of a guitar sound I've heard is Chocolate, I didn't know if the person describing it meant sweet, dark, soft, light , heavy, or what. Way too strange for my mind to get around as most of the wine and guitar sound descriptors.

Fred

_________________
Fred Tellier
http://www.fetellierguitars.com
Facebook page http://www.facebook.com/pages/FE-Tellier-Guitars/163451547003866

Uh-oh. It's happening.

I've heard that one for wine. And "hint of cinnamon", vegetal, herby (as in herbs), and barnyard. What would barnyard mean as a guitar tone descriptor? A honker?

Pat

_________________
formerly known around here as burbank
_________________

http://www.patfosterguitars.com