Even something as seemingly simple as this is much harder to pin down than it sounds.
Bear in mind that the pitch generated by a whistle depends on much more than the physical dimensions of the whistle. It is the resonance of the air column inside the whistle that determines the pitch, and that changes depending on the temperature . So the exact same whistle will take different inputs to play in tune at different temperatures.
If you pull the head or tuning slide to compensate for different temperatures, then you affect different notes of the scale differently, so the internal tuning balance changes for every position of the slide. Hence it also changes for every different temperature.
If a maker tunes their whistles precisely to 12 tone equal temperament at A440 right before they leave the workshop, regardless of the current temperature, then you can guarantee that the whistles will not be consistent with each other, since the temperature changes from one day to the next.
The same is true for humidity, atmospheric pressure, elevation, and variations in the players breath temperature, to name just a few other variables.
And this isnât even addressing the huge question of what âin tuneâ means.
For the purposes of testing pitch, would it be demonstrative if you always tested it at the same temperature, elevation, and humidity, to show if itâs in tune with itself at least?
If you choose a specific temperature, humidity, and pressure and demonstrate that a whistle is in tune with itself and tuned to A440, then that tells you for sure that it will not be in tune with itself at other temperatures etc. How useful is that, really?
Also if you choose conditions that are representative of your location in Utah in either Summer or Winter, they are going to be of no relevance to someone in, say, Ireland. High, dry, and either very hot or very cold, vs humid, sea level, maritime climate.
Putting numbers on something like this tends to create an illusion of precision that doesnât exist in reality and isnât very useful either.
I do not. I make my flutes such that they have tuning and playing characteristics inline with those of other highly regarded flutes (ie, ones I like). Then, after much side by side testing against those other examples, I settle on a design, which is embodied in an actual flute that I refuse to sell. This flute, which I call the âreference fluteâ for that particular model, is then used in direct comparisons with all subsequent flutes I make in that model. I tune and voice them to be as close to identical to the reference flute before releasing them. This works regardless of temperature or atmospheric conditions.
What this strategy buys me, primarily, is very tight consistency. This is useful in helping people decide whether they like or dislike the way my flutes play.
The strategy also rests on the observation that people play flutes into tune, and that there is no such thing as an inherently in tune flute absent a players input.
One simple piece of analysis is to measure the hole size and spacing of a collection of whistles and then calculate an estimate of the resonant frequency relative to the bell note for each hole with the âBenade equationsâ used for the various hole position calculators. This is doing in reverse what the calculators do - and the maths is much simpler.
For the ones I have done the result is, for example, what you would expect from looking at and playing two whistles with the same hole spacing but different sized holes. But it puts some numbers on how different makers make different choices. And you can do it for different temperatures.
For sure thatâs one of the conundrums that Low Whistle makers face: acoustics wants Hole 6 to be as big as Hole 5, and placed halfway between Hole 5 and the open end, but located there nobody could reach it.
I now have on loan 31 Low D whistles from a large number of makers and all of the makers have arrived at the same compromise for Hole 6: itâs moved up to where most people can reach it, requiring it to be too small.
The interesting thing is that a small number of makers have somehow voiced the whistle so that Low E is really close in volume and timbre to the rest of the low octave, in spite of the too-small too-high hole. With most makers thereâs a noticeable dropoff in clarity and volume when you hit Low E.
BTW on most of the Low Ds I have here Low E is a bit sharp.
I do have here, from two makers (Howard and a Carbony) pairs of Low Ds with both regular hole placement and close hole placement.
That doesnât surprise me. I think this reflects the makerâs desire to improve how the E speaks, even at the expense of its tuning.
This is something I also seen on flutes designed for people with small hands. The E tuning in this case is sacrificed for ergonomics. But of course, in both cases it is a three way trade off between voicing, tuning, and ergonomics.
This trade off seems to be easier to satisfy on a conical bore instrument than a cylindrical bore one.
I think something similar when I see a flute where the F-E spacing is larger than the G-F spacing. Itâs more uncomfortable ergonomically, but the maker wanted to put the E just a bit lower downâŚ
To return the thread to the original topic (as I understood it, at least), I think a database of some easily measureable stats for whistles is an interesting proposal albeit a lot of work.
But of course, if thereâs a will, thereâs a way and even though thereâd be variety among some of the stats due to various factors, it doesnât negate their usefulness.
I worry though, that the stats might not be easily understood by everyone who has access to the database. As long as the numbers could be easily translated into descriptive information, that would work.
I also think itâs important that readers have a point of reference for the numbers, a standard by which they can gauge the data that they read e.g. the most common, easily available regular D whistle scores the following <insert data points/ numbers etc here>.