[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

levels of abstraction



hello all -

i'm curious about the levels or forms of abstraction that people in this
list use when working with sound.

what i mean by that maybe i can clarify a bit by describing what i often
find myself doing. generally, when i work with sound on a computer, i'm
thinking of the sound in terms or an array of numbers, and i'm thinking of
filtering in terms of mathematical operations on a series of numbers, and
synthesis as a means of genereating series of numbers. i imagine that this
attitude stems from the tools i've used - snd/scheme and matlab in
particular - but even when i'm using software that completely hides that
notion of digitial audio as a series of numbers, i still often have this
in mind.

a second level of abstraction that i often use is the spectrum/fourier
transform. i think about the distributions of different frequencies,
harmonic stacks, etc. this is also a somewhat mathematical approach, but
it jumps a step away from a single series of numbers and becomes, through
the fourier transform, a set of intensities at different frequencies which
are summed together.

so - if any of you happen to relate to this question at all - what are
you thinking about when you deal with sound? i imagine there are endless
possibilities here which are likely constrained to some extent by the
tools used.

i think the approach i described stands in start contrast to the
traditional units of abstraction in western music - note, pitch, loudness,
timbre, harmony - and certainly the two have a relation to one another,
but generally i'm simply not thinking about notes, but rather signals.

i'm curious to hear what all of you have to say - i imagine this could
turn into an argument about whether or not different perspectives are good
perspectives - but at the moment i'm not so interested in that argument,
rather just hearing what's out there.

brett.

------------------------------