Reel Time Tuning Analysis

(By Terry)

A Long Felt Want

For as long as I can remember as a flutemaker (probably about 32 years!) I've wanted a better system for tuning flutes. In the earliest days, I used a tone reference I built myself - the handy little tuners available now were not available then! Then I got one of those handy little tuners. When Tatsuaki Koroda came out with his PC-based AutoTuner, and was then prepared to make changes to it to suit my very specific needs, I switched to that (and still use it!). But always, always, something else was in the back of my mind.

So what's wrong with a tuner?

In a word, people! When you sit in front of a tuner, and play a note on the flute, it's very tempting to lip that note up or down to make it look more like what you're hoping to see. Even when you are trying not to, you can't be really sure you didn't. And when you have to pause after each note, and note down the reading, then find your embouchure again before passing on to the next note, it's tedious, time consuming, and another source of error. And always, at the back of the mind is the doubt - OK, now I know what I get in front of a tuner, but how do I know I blow the same way when actually playing music?

And what would be better?

What I've always wanted is a system that as much as is possible, pulled me out of the picture as observer, note taker and decision maker. Something that:

I could just play into without being made aware of any results (to keep me honest)
Let me play music, not notes (to make it realistic)
Let me play for a while (so that I would relax and play normally, and the system would have plenty of good data to choose from)
Let me play the notes in any order I wanted (so that if I play downward intervals differently to upward intervals, or wide intervals differently to narrow ones, it could average all that out)
Did all the note-taking (so I could concentrate solely on playing)
Did all the collating and presentation of results (so I wouldn't have to wade through vast amounts of data, cherry-picking the flattering ones)
Decided on an objective basis which notes were reliable enough to include and which should be excluded (to keep me honest and in case I fluff a few notes in a long test piece)
Cover keyless and keyed flutes, low flutes, piccolos, fiddles, any any anything else we might throw at it
Be quick and easy to operate
Would work for players as well as makers.

And?

I'm pleased to say I think we have it, and you can have it too, at no cost. New Zealand flute player and computer scientist, Graeme Roxburgh happened to mention the development of a new piece of music software at his University. As a data collection system it looked ideal, and I wondered out loud to Graeme whether the authors would be interested in developing it further. Turns out they are not, for the moment at least, but, when I spelled out my long felt want, Graeme offered to look into how else it might be achieved. The rest, as they say, is history.

A word on nomenclature

You'll find in the pages that follow that Graeme has adhered to the international standardised nomenclature for notes. This is a system that has to work for all instruments, including massive organs and pianos. Consequently, you might be surprised to find that the lowest D on an Irish flute is called D4, middle D is D5 and third octave D is D6. The notes above D4 are all called 4 until we get to the next C, which is C5. It might seem a bit strange at first, but you'll quickly get used to it.

Give it a go!

So, that's the brief background to how this all started. Interested? Read on ...