sandell@aristotle.ils.nwu.edu (Greg Sandell) (03/24/91)
weigel@DPW.COM (William Weigel) writes: > The previous post asked whether a pure sine wave Bb sounds different > from a pure sine wave A. Few people would make a distinction between the tone color of two sines a halfstep apart, I believe. On the other hand, the tone quality of sines that are farther apart HAVE been demonstrated to be different. Listeners have noticed that a sine tone at a low frequency sounds duller and darker, like an "oo" vowel, while a sine tone at a higher frequency sounds brighter, as in "ee". (Robert Cogan cites Ernst Mach as having first observed this in 1885.) The myth that pitch and timbre are completely orthogonal (i.e. that you can vary pitch and keep timbre invariant, and vice-versa) is an unfortunate propogation of the timbre-definition-that-wouldn't-die, the one by the American Standards Association (made in 1960). It says: "Timbre is that attribute of auditory sensation in terms of which a listener can judge that two sounds similarly presented and having the same loudness and pitch are dissimilar." So loudness is also orthogonal to timbre? Not so. A trumpet played mezzo forte and then forte yields two very different spectra; now suppose you equalize them in amplitude and present them to a listener. You ask him, what changed between the two sounds? Here is a possible dialogue. Listener: "The second sound was louder." Presenter: "Was there a change in timbre?" Listener: "No. It was still the trumpet." So timbre created the suggestion of a change in loudness, even though the sounds did not vary in amplitude. That phrase "similarly presented" gets on my nerve, too. In 60's vintage language, I bet that means "with the loudspeaker in the same location, in the same reverberant environment." Another incorrect assumption, this one due to outmoded technology. Spatial location and reverberation are now parameters on the timbre knob that we can twiddle just as easy as pitch and loudness. In other words, to vary a spatial location of a sound and tell a listener "Don't be fooled into thinking you just heard a timbre change; that was a change of spatial location" is a fallacy. Basically, an outmoded viewpoint is what characterizes that definition. Why, for instance, continue to use a definition that was cast in the context of a particular kind of experiment design? The ASA definition suggests a classic psychoacoustic experiment of keeping one parameter constant while varying others (and acoustics has grown up alot in the last 30 years). Does this *define* timbre, or does it merely *demonstrate* it? I think the latter. I found out recently that the American Standards Association has just drafted a new definition of timbre, and it has "gone to vote." Now I wonder who is voting on it? You'd think that they'd want to invite some of the people who have done alot of research in the area of timbre; recently I mentioned the new ASA draft to David Wessel, and he was unaware of it. Who knows what we'll get this time. It's interesting to think of them 'voting' on this kind of thing, and there's a small detail about the 1960 definition that suggests that coming to a consensus might have had some difficulty. The sentence I quoted above is the formal definition, but a tiny comment is appended to it, in a smaller font: "NOTE: Timbre depends primarily upon the spectrum of the stimulus, but it also depends upon the waveform, the sound pressure, the frequency location of the spectrum, and the temporal characteristics of the stimulus." This, of course, helps correct the flaws of the definition; the problem is, though, the formal part is what has become the definition for the status quo acoustical world. - Greg Sandell -- Greg Sandell sandell@ils.nwu.edu