Quote:
Frequency is how high or low a sound is. Dynamics measure how loud and soft it is. Distortion measures the shape of the sound. What other property is there to measure?
From my limited understanding of harmonics, discussing "frequency" in the singular may be oversimplifying things a bit. If we are merely talking about reproducing the sound of a pitch fork, or a pure electronically generated tone, then perhaps fundamental frequencies are all a dac has to handle. But with live instruments and voices, the perception of timbre that allows us to differentiate between a human tenor, a trumpet, and a guitar (as three random examples) are varying magnitudes in overtones relative to the core frequency, and (I think), some amount of phase shift between them, among other things. (
http://en.wikipedia.org/wiki/Harmonic_series_(music) )
Reproducing a single note plucked on a guitar, and making it sound like the instrument it is actually being played on, means the equipment has to account for quite a number of simultaneous frequencies, in some permutation of phases, at the correct magnitudes. Now, instead of a single note on a guitar, let's contemplate a full orchestra, along with a choir, performing something fast and dynamic like Verdi's Dies Irae. Playback is certainly going to be limited by the quality of the recording/the source data, but it seems to me that in any instant, a dac playing this sort of music is going to have to convert, and seamlessly output in analog, hundreds, if not thousands, of distinct frequencies, at different phases, and at different magnitudes in order to accurately convey the notes played and the various instruments/voices playing them.
I am totally ignorant as to whether or not there is even an accurate/accepted way to measure a dac's performance in playing back this sort of material. Maybe my speculation is totally flawed, but it seems like the quality of the hardware, speed of the chip, lack of any latency in the system as a whole, and the power supplies implemented in the output stage might make a real difference - particularly when listening to this sort of a recording? Then again, if listeners can't tell the difference in a DBT, maybe this is all irrelevant.