Hi Marcus,
I’ve heard (and read, though I cannot yet rediscover the source) convincing arguments that it’s the microscopic differences in texture that create these tonal differences despite similar profile & voicing–i.e., they generate different turbulence at the outside boundary of the air column. I leave it to the trained scientists amongst us to use the proper terminology or tell me I’ve got it all wrong.
I’m not representing it well, but the science seemed, in my limited judgement, valid. Certainly a more valid (if inconclusive) datapoint than the various tests with musicians and testing apparatus (apparati?) that have been done to answer this question: it seems to indicate that there could be a valid, objective, and inherent property that contributes to the differences we detect. However, it’s trying to isolate one aspect in an intensely multi-variable environment.
The rest is purely subjective assessment reinforced by the anecdotal evidence of many others with similar responses. I’ve been lucky enough to try at different times the same model of flute made at the same time in different woods, and to exchange headjoints of different woods on the same flute body–and listen to others do the same: a flute “tasting,” if you will. My silly formulation is just a representation of my impression of the amount of tonal change due to the various factors.
It’s not unlike systematic wine tasting or coffee cupping - a valid model for judging these types of factors (combination subjective/objective assessment of micro changes in the thing tested, paired against the predispositions, experience, and sensitivity of the tester).
I’ve seen (long ago) tables from two different tests done, one purportedly proving your belief and the other disproving it. Given that a mechanical/automated test appliance could not adequately represent all of the micro-adaptations a musician makes during performance, and on the other hand, a musician could not approach the level of consistency required for consistent results, (even if one used double-blind techniques) I sincerely doubt a conclusive result is possible using objective and scientifically sound measures. Even if it were, would the conclusions be useful to musicians? This is without even bringing into the equation the differences introduced by different makers, different voicings, and even different sticks of wood from the same tree.
I don’t think the question as it is usually phrased can be definitively answered in objective terms. For me that’s not necessary, though–I’ve experienced enough “high positive correlation” between wood types and tonal differences, all other factors being equal (as possible), that I’m comfortable with my own broad generalizations. I like mopane, boxwood, and blackwood flutes for different reasons, and with few exceptions, I’m not surprised by the changes in my own sound when I play flutes made from these different woods.
Further, even if there were or were not an inherent and objectively quantifiable difference between the tonal characteristics of wood types, I’m still going to make my choice of instrument based upon maker, voicing style, and accoustical model before making a choice of wood (with the possible exception of certain boxwood instruments, which, for me, break out of that 10% impact on tone–either that or figured boxwood just makes me soft in the head).
