Tuesday, December 27, 2005
The degree to which our processors seldom these days approach their "peaks" is the degree to which the computational requirements of a certain class of musical "problems" has finally been eclipsed by the sheer computational speed of the actual machines themselves. In our current situation, unless some new paradigms of musical complexity are introduced (nothing demands this, naturally), future "error" (resulting from hitting processor "ceilings") as such will be qualified not so much in terms of its frequency of occurance, but by the degree to which it may be, for its pure ornamentality, freely introduced by the composer.
Put another way, what musical problem imaginable would justify the power of one of these monsters to solve? This argues the need to go in the opposite direction, some would say...To argue this ad absurdum leads, absurdly, to a situation where the "Digiki"s of the future make their "Polypunk 1000"s...on The Earth Simulator IV?