[sdiy] OT: making "IDM"
Metrophage
c0r3dump23 at yahoo.com
Mon Jun 20 17:27:22 CEST 2005
--- Richard Wentk <richard at skydancer.com> wrote:
> The problem with any algorithmic approach is that it's really,
> really, really hard to do well. It's not so hard to do random-ish
> sample cut-ups, but for satisfying music you need a sense of
> larger structure and development, and that means introducing
> large-scale control structures that correlate events at multiple
> levels. This is more of a challenge than a random walk through
> a constrained space, perhaps with some fairly simplistic
> Markov-chaining.
Indeed! Of course randomness is not difficult to produce. Even
something as complex as Xenakis' "GENDYN" system still often sounds to
me like a swarm of 50-foot mosquitos. OTOH Markov chains can be
tempered enough for more conventionally "musical" sounding results.
I think that much of the reason why people use algorithmic software in
music is because of what you noted: creating control structures which
correlate events. This is very easy to do in a fully modular system,
where signals can also be control signals. Even though digital music
environments are ideally more modular, most music software has evolved
from the simple linear timeline of the magnetic tape studio. It is
difficult to get really polymetrical with most software.
The issue with randomness as concerns Autechre is more of perception.
The very same control structures which you mentioned are capable of
producing relationships which are not intuitively obvious to many
people. My experience in performing is that when I am playing patterns
which are in even meter, say 4/4 or 3/4, then most listeners tell me it
sounds "structured" and "tight" (even though some of my programming is
crap). When I am playing patterns which are in in various simultaneous
meters, and/or odd signatures, then most listeners tell me that it
sounds "random" and "crazy". I can relate very well to the frustration
of people referring to my music as random, I don't hook up a S&H and
twiddle my thumbs! I grew up on industrial music as well as more
rhythmically involved musics such as Gamelan and Indian classical
music, so this progression is quite natural to me. Even western
classical music started to (finally) develop theories of rhythm over
the last century.
That said, I am not a real percussionist or a very skilled programmer.
I have been making algorithmic rhythms for a few years, some I like,
but it does take a lot of work to make something which satisfies my
sense of creative symmetry. My past experiences have involved very
simple patches in Max and sometimes SuperCollider, with a simple
hardware controller. A lot of my Max patches were based on Karlheinz
Essl's "RTClib", a Max library for Real Time Composition. I'd make a
MIDI in, connected to a Phatboy knob controller. A MIDI out, then
connected to my 707 or SR16. Then I'd make a basic 32-step sequencer
with toggle-boxes to write patterns. Each row of 32 steps had its own
clock. Then I'd connect the Phatboy through various ranging, scaling,
and quantizing to vary the interval of each rows progression from one
beat to the next. This is a simple route to algorithmic sequencing
which even a Max novice such as myself is able to do. My modified SR16
was able to record the patterns in real-time as I played, usually in
multiple passes with different quantization. Then when I took the SR16
out, I'd sometimes change the patterns further by having some bits from
the ROM outputs trigger various events in other parts of the drum
machines circuitry. The last step was difficult to control, but I was
able to get some "predictable" results from it. Some kind of divider
chip would have perhaps been useful there too. The output of my Max
patches were quite fun at times, so I began to favor this way of
working. Too bad my PowerMac was too big to bring on the subway easily.
Other fun options are LFO control of interval rate, one could
experiment with this stuff for years.
For the past two years I have been trying to build my analog modular,
and have not been playing as much. I do at times use the Numerology
sequencer, which is a really involved bit of software for doing stuff
much like my Max patches. Not as versatile as Max, but very quick for
just jumping in and playing. Also much cheaper than multitrack
sequencer software.
Sigh.... all of this talk makes me want to make music! Now I just need
to build that MIDI interface to my laptop, and finish those panels in
the garage, and mount the hardware for my drum machine mods. And then
program some fresh sounds on my synths... AIIEEE!!!! I have not been so
spontaneous lately. Must_build_studio!
CJ
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
More information about the Synth-diy
mailing list