A few specific questions about testing wood influence on tone

[beaubrummels]

Re: A few specific questions about testing wood influence on tone

For this test, the question is basically "does wood matter at all?", so I'm not necessarily trying to profile the response of a species of wood, so I shouldn't have to worry about how representative a sample is of that species, or how much sap or mositure is in the wood, what part of the tree it came from, how many samples are tested, etc, because all that would matter is that there is or is not a difference.

I'm open to using a noise generator and a microphone or piezo or something, but since the practical application is guitar, I'd think a string and a pickup would satisfy the "problem domain".

If all you want to answer is if the wood contributes any difference, just take your pickguard assembly out of one Strat, wholesale, and slap it into a different Strat that has different body wood but the same hardware. If you hear a difference, I would consider that compelling evidence that the wood contributed to the difference.

 

[David Collins]

Re: A few specific questions about testing wood influence on tone

If all you want to answer is if the wood contributes any difference, just take your pickguard assembly out of one Strat, wholesale, and slap it into a different Strat that has different body wood but the same hardware. If you hear a difference, I would consider that compelling evidence that the wood contributed to the difference.

Would be nice if it were so simple, but maintaining ideal controls and objective appraisal with this type of comparison becomes much more challenging than most may think.

 

[DreX]

Re: A few specific questions about testing wood influence on tone

It does get complicated though. With nearly every step you take to simplify, there are almost inevitably some opportunities for influence in the outcome being missed. I course a line has to be drawn some here for "good enough", but if too many corners are cut, confidence in the results can quickly drop so low that you're left to wonder whether it was worth doing to begin with.

Hypothetically or otherwise, how does someone say the test was invalid because there wasn't enough extraneous elements, such as frets or a specific body shape? Would they claim that these elements alter the wood's ability to behave like wood?

 

[DarkMatter]

Re: A few specific questions about testing wood influence on tone

David is right. No more shenanigans from me. I'd genuinely like to see something come of this. Everyone, let us science types indulge ourselves here.

 

[DreX]

Re: A few specific questions about testing wood influence on tone

If all you want to answer is if the wood contributes any difference, just take your pickguard assembly out of one Strat, wholesale, and slap it into a different Strat that has different body wood but the same hardware. If you hear a difference, I would consider that compelling evidence that the wood contributed to the difference.

I'll measure and record the difference, but just listening wouldn't be objective enough, even for my own uses. I think I hear a difference when I swap pickups between otherwise identical basswood Strats, but maybe it's literally the fact that they're different colors that makes me think they sounded different.

I think setting up a simpler test with disposable wood boards is about as much work as swapping all the hardware on a Strat, so I'd rather go with the "cleaner" of the two test setups, unless there some reason the believe the wood only behaves properly when cut in the shape of a guitar and has all the associated accoutrements attached to it. I gotta be honest, a more complicated setup could break my time budget (or the amount Im willing to spend to answer this question), hence the reason for my questions.

 

[David Collins]

Re: A few specific questions about testing wood influence on tone

Hypothetically or otherwise, how does someone say the test was invalid because there wasn't enough extraneous elements, such as frets or a specific body shape? Would they claim that these elements alter the wood's ability to behave like wood?

Reliable tests could certainly be done without a complete instrument (frets, truss rod, etc). A monochord setup would likely be perfectly suitable. The biggest challenges tend to come with reliable drive and observation techniques. There are a lot of ways this can go, and when vetting your methods, many would be surprised just how challengng it can be to get consistent results with the same sample on subsequent tests. When your testing is comparitive, this stage of proving your methods is never so easy as it may seem.

I'm currently doing some tests on fret wire, primarily focused on wear rates of different materials and different treatments, but also getting in to concerns on tonal effect (which I am not terribly optimistic of finding). Over the last two months I have run millions of cycles on my test apparatus just to work out the bugs and refine it to a point where I can be confident that results are identical on subsequent tests of the same wire. It always seems simple, but once you get in to evaluating and refining reliability of methods, it never is.

I may take on a study of this sort later in the year, but doubt any sooner.

 

[Tonewoods]

Re: A few specific questions about testing wood influence on tone

Would be nice if it were so simple, but maintaining ideal controls and objective appraisal with this type of comparison becomes much more challenging than most may think.

Blinding is non-negotiable.
There's also the bottomless 'variable rabbit hole' that tonewood believers tend to use as a crutch when their beliefs are called out in practical terms. In their demands, they usually just wind up demonstrating how little they understand about science (blood pressure? LMAO)

If you're interested in the theoretical, Will Gelvin did some very decent videos.

https://www.youtube.com/watch?v=nJfeFe4CKEk - Tonewood Riff

https://www.youtube.com/watch?v=K7tcVVt3OrM - How pickups work (he has patents on pickups)

https://www.youtube.com/watch?v=Ehgl-AeDPkQ - Resonant frequency / anm actual sig generator

If you want to simplify it, here's a guy offering a free blind challenge in a random blind test. He posted it yesterday'ish.
https://www.youtube.com/watch?v=hjHt9IqDmIA
If attack subsumes tonewood as a operational variable, then the tonewood believer narrative just took a huge dump.

If you want to get into setting up an efficacious test to duplicate conditions that isolate tonewood as a concept completely away from its interplay with the instrument, we first have to establish the question we want to answer in its simplest and clearest terms.

 

[DarkMatter]

Re: A few specific questions about testing wood influence on tone

I feel that on an electric, the body has a very small effect, but the neck could potentially have a noticeable effect. What about having a test body that stays the same and then you just swap necks?

 

[Tonewoods]

Re: A few specific questions about testing wood influence on tone

I feel that on an electric, the body has a very small effect, but the neck could potentially have a noticeable effect. What about having a test body that stays the same and then you just swap necks?

That would be a good way to approach testing the neck.
The key, of course, is blinding. In the realm of subtlety and 'perception', people will hear what they want.

 

[DarkMatter]

Re: A few specific questions about testing wood influence on tone

That would be a good way to approach testing the neck.
The key, of course, is blinding. In the realm of subtlety and 'perception', people will hear what they want.

I agree that without eliminating the perception factor, testing becomes pointless.

 

[DreX]

Re: A few specific questions about testing wood influence on tone

Blinding is non-negotiable.
...
If you want to get into setting up an efficacious test to duplicate conditions that isolate tonewood as a concept completely away from its interplay with the instrument, we first have to establish the question we want to answer in its simplest and clearest terms.

I'm more interested in measuring a difference, because if it's measurable, then there's something there, regardless of whether it's audible, and if it isn't likely to audible but people claim they hear a difference anyway, I'll give them the benefit of the doubt and let them ascribe as much or little meaning as they want to it. But if I can't even measure a difference with multiple samples of wood of varying varieties, then I'm going to doubt that myself or others ever heard a genuine difference.

I like the idea of a blind test, but I'm not in the best position to conduct such a test. More time and money required than I have available.

 

[DreX]

Re: A few specific questions about testing wood influence on tone

I feel that on an electric, the body has a very small effect, but the neck could potentially have a noticeable effect. What about having a test body that stays the same and then you just swap necks?

I was thinking I should test larger boards and very thin ones, to represent the body and the neck respectively. Maybe it's the case that the neck makes all the difference and body make almost none. Maybe a 1x2 for the thin and a 2x12 for the thick.

One problem I've already run into is looking for "tone woods" on ebay, they come is many strange non-building sizes, and I don't have a planer to make a bunch of weird shaped samples into a consistent size and shape. This would be really easy test for a luthier to set up, it makes me wonder if they'd rather not know.

 

[Tonewoods]

Re: A few specific questions about testing wood influence on tone

If tonewood can withstand blinding in practical terms, we can 'give the benefit of the doubt' to its theoretical underpinnings since it will have demonstrated itself practically.

If tonewood is only perceptible under spectography, that still does not mean they can hear it, even though they sincerely believe they do. Its easy to confuse 'a certain sort of person' with tricky causal variables, so no benefit of the doubt is earned simply by demonstrating that there's a spectrographic difference, particularly of the MOE isn't normalized.

 

[DreX]

Re: A few specific questions about testing wood influence on tone

If tonewood can withstand blinding in practical terms, we can 'give the benefit of the doubt' to its theoretical underpinnings since it will have demonstrated itself practically.

If tonewood is only perceptible under spectography, that still does not mean they can hear it, even though they sincerely believe they do. Its easy to confuse 'a certain sort of person' with tricky causal variables, so no benefit of the doubt is earned simply by demonstrating that there's a spectrographic difference, particularly of the MOE isn't normalized.

But it changes the question from "is there a difference to be seen?" to "can the seen difference be heard?", and I would be satisfied to have narrowed it down some.

Maybe I could do a blind audio test with the same test rig used for measurement, but in this case even I would take issue with the rig not being an actual guitar, since there would be the question of whether, for example, a single string of fixed pitch makes the differences as clearly audible as six strings that can perform a full scale length of pitches.

 

[GoldenVulture]

Re: A few specific questions about testing wood influence on tone

Blinding is non-negotiable.
There's also the bottomless 'variable rabbit hole' that tonewood believers tend to use as a crutch when their beliefs are called out in practical terms. In their demands, they usually just wind up demonstrating how little they understand about science (blood pressure? LMAO)

If you're interested in the theoretical, Will Gelvin did some very decent videos.

https://www.youtube.com/watch?v=nJfeFe4CKEk - Tonewood Riff

https://www.youtube.com/watch?v=K7tcVVt3OrM - How pickups work (he has patents on pickups)

https://www.youtube.com/watch?v=Ehgl-AeDPkQ - Resonant frequency / anm actual sig generator

If you want to simplify it, here's a guy offering a free blind challenge in a random blind test. He posted it yesterday'ish.
https://www.youtube.com/watch?v=hjHt9IqDmIA
If attack subsumes tonewood as a operational variable, then the tonewood believer narrative just took a huge dump.

If you want to get into setting up an efficacious test to duplicate conditions that isolate tonewood as a concept completely away from its interplay with the instrument, we first have to establish the question we want to answer in its simplest and clearest terms.

LOL December 29, 2014
This is a valiant effort, but this is not ‘science’.
It could perhaps better be called ‘vaguely science’ish’ or maybe ‘sciency’ but there is just so much wrong.
First off, you start out from the vantage that wood makes a difference in an electric signal to a degree that can be heard. You basically accept an inherently invalid premise as valid, then go from there. The entire experiment is essentially an exercise in confirming your own biases, which themselves would not withstand a blindfold.
Your entire conclusion is that it has something to do with ‘overtones’ that again, are deep into the realm of magical thinking.
Lastly, everything is unblinded (you provide an answer key before the video even begins) which all but guarantees confirmation bias will be overwhelming.
Nice try, but perhaps study how medicines are studied in clinical trials- and more importantly, why they are studied that way- to shed a little light on why this kind of ‘test’ doesn’t add much to the tonewood debate, other than being a Rorschach that allows a certain sort of person to see what they want.

- http://guitarworks.thestrandbergs.com/2014/12/28/the-impact-of-wood-choice-in-an-electric-guitar/ -
:soapbox:

 

[DreX]

Re: A few specific questions about testing wood influence on tone

- http://guitarworks.thestrandbergs.com/2014/12/28/the-impact-of-wood-choice-in-an-electric-guitar/ -
:soapbox:

That's pretty close to what I'd like to do, but I'll dumb it down even further, take more pictures and try other sample methods for the FFT comparison. I'm unclear as to how they dealt with issues of repeatability and why the pick attack sounded as different as it did from one sample to the next. It looks like I can expect a wide variety of variation based on this, but I'd like to see it for myself and document the process a little better.

 

[DarkMatter]

Re: A few specific questions about testing wood influence on tone

I wonder if there is some type of relatively cheap spring-loaded device that could be repurposed as an string stimulus. Of course, proving it's repeatability may be a problem.

 

[DarkMatter]

Re: A few specific questions about testing wood influence on tone

- http://guitarworks.thestrandbergs.com/2014/12/28/the-impact-of-wood-choice-in-an-electric-guitar/ -
:soapbox:

Was one of the comments after the article you Tonewoods?

 

[Chris of Arabia]

Re: A few specific questions about testing wood influence on tone

I'm unclear as to how they dealt with issues of repeatability and why the pick attack sounded as different as it did from one sample to the next.

My guess is that it's nothing to do with pick attack, but is instead to do with the way the samples were cut from the original recording. He says in the article:

Created Fast Fourier Transform snapshots 2 seconds into each clip to provide a visual comparison

Assuming that to be a precise, but arbitrary, cut point, there is no guarantee that the signal was at 0dB there; it could be anywhere between 0dB and whatever the max signal strength is for that recording. Anyone with any experience in cutting up samples will know that you cannot just select an arbitrary point in time, you have to get right into sound wave and cut at the point where the signal crosses 0dB - using the Strandberg article as an example, that may or may not be at 2 seconds. The practical effect of not doing this is that you get something of a click as the sample is played - the amplitude of the click is very much dependent upon how close to the 0dB point the wave was where is was cut.

 

[LesStrat]

Re: A few specific questions about testing wood influence on tone

There are so many confounding variables in this proposed "study" as to make the entire process uncontrolled, and therefore invalid, unreliable, and meaningless.