Wednesday 24 September 2008

A rift in the lute?

At the beginning of this month (before this blog existed) there was a flurry of press releases about some European researchers who'd recreated the sound of the epigonion, an extinct kind of psaltery (though everyone's been calling it a lute). It's a nice story, but the sounds, to me, are rather disappointing. A clue as to why this might be comes from the project webpage:
"Although physical modeling was not a new concept in acoustics and synthesis, having been implemented using finite difference approximations of the wave equation by Hiller and Ruiz in 1971, it was not until the development of the Karplus-Strong algorithm, the subsequent refinement and generalization of the algorithm into the extremely efficient digital waveguide synthesis by Julius O. Smith III and others, and the increase in DSP power in the late 1980s that commercial implementations became feasible."
The extreme efficiency of digital waveguide modelling comes at a price. With this method it's very hard to capture dispersiveness* or nonlinearity, both of which can make an important contribution to the distinctiveness of the sound of an instrument. Of course, I'm not saying these researchers didn't address these issues, but I'll have to wait until there's a published paper to find out how.

The method's great advantage is that it's probably the only one currently able to compute waveguide response quickly enough to work in real-time, which is vital if you want to be able to change some parameter during a note while playing your synthesizer. But all the computations for this project were done off-line; apparently it took four hours to compute 30 seconds of sound. Other methods (such as finite elements) would take longer but would be better placed to capture the physics.

Even if we assume perfect (or as good as possible) modelling there's another issue endemic to musical instrument acoustics research: how detailed to make the model? Two violins that look very different can sound very different to the people who matter, the players. Wood is complicated stuff, just read chapter 6 of J. E. Gordon's excellent "The New Science of Strong Materials", a masterpiece of science writing. The chapter is headed by Charles Gurney's couplet "Plastics are made by fools like me, /But only God can make a tree". Two pieces of the same wood, the same size and the same shape can behave very differently. So when you model your epigonion how much detail do you put into your model of the wood? And what about the pick with which it was plucked, that can make a big difference too. Again, I'm not saying these researchers haven't thought about these issues, but I'm pretty sure the news outlets who've taken their press release and turned it into a "triumph of science" story haven't. I'll be interested to read the paper when it comes out, I'll post my comments here.

* Not so, see later post.

Tuesday 23 September 2008

Welcome

This is the obligatory 'hello world' post to help me get to grips with the technology. It may well get deleted later.