On Wed, 9 Jul 2003, Peter Fairbrother wrote:

the nyquist/lindquist/someone-else-who-was-pissed sampling theorems are based on the possibility of mathematically extracting frequencies from digital information in a STEADY_STATE situation.

That doesn't mean that a speaker will properly reproduce those frequencies.

Nor does it mean the op amp driving the speakers will follow them either. High speed and power are a hard combination to build.

Consider the dynamics of energy transfer. A digital signal at near-1/2-sampling frequency will have two datum points. The transitiion between them will be dramatic! the possibilities of energy transfer will not be comparable to an analogue sinusoidal waveform.

And that's why good analogue is better then good digital.

It's definitly why you need fast digital. To reproduce 20+ kHz you should use a 200kHz sample rate and have a nice filter stage before the power amp. "good digital" can do more things than good analog because the final output is good analog in both cases. The speaker driver is pure analog by definition. To produce 65kHz (for cats) my present boss prefers a 1 MHz sample rate. The guys who do bats think it's good enough for 200kHz, but my boss won't do bats - much too complex. We've got a 25 bit dac which updates at 1 MHz, but we still need a nice filter and analog output stage for 120 dB clean signals. (I'm only getting 100 dB because it costs too much to really do the best possible.) Clearly a digital system can be built that can create any wave form a speaker can follow, and it's easier to control than an analog system. The human hearing system is capable of noticing phase relations at 100kHz rates. So any sample rate faster than 200kHz is outside the range of human detection. Cats can notice phase shifts in the 200kHz range, and bats are out in the 400kHz range. Biological systems *are* impressive. But digital vs analog is a silly argument, the final stage is analog. Patience, persistence, truth, Dr. mike