I'm finally reading John Adams' autobiography, Hallelujah Junction: Composing an American Life.
In the chapter on the impact of electronics in classical music, he talks about a Harvard professor who had his students listen to various instruments all playing the same pitch, except without the initial "attack" i.e. as a sustained sound. And so it was hard to actually distinguish the clarinet from the trumpet from the violin from the piccolo:
The lesson was that most of the characteristic information in a musical sound is in the onset, or the "attack." Strange but true: the attack is largely characterized by noise rather than by pitch. Each instrument has highly complex attack characteristics, and it is from here, in this split second of the beginning of a sound, that we derive the lion's share of the information.
To be fair, the book is not at all pedantic like that paragraph might imply.
And just today, Musical Assumptions happens to post about this very idea:
The edge for me is the millisecond when sound begins.
Comments