Machine Music
::signalPaths pt2::
Dreams of Electric Sheep
Music trackers are among the earliest means of using a computer to compose compose and sequence electronic instruments. They offer a rather matter-of-fact approach to electronic music and resemble the spirit of the particular hardware they evolved alongside of in that regard: few sounds evoke digital the way that the sounds of the 8bit era of gaming and computation seem to; likewise, few means of composing music say computer the way a tracker’s downward progressing sequence of commands and parameters does; and both just scream electronic very much in the way we’ve come to understand it in the information age. Yet it could be said that what truly manifests the quality of electronic is that which best conveys the fundamental physical aspect of the electronic signal in itself. And what is digital but for the mere processing of information? And what is this but an aspect of computation?
Seen in this light, mine own favored approach of PSGs and trackers might well seem rather ancillary or superfluous to the nitty-gritty of electronic sound in music. And indeed there are, as I see it, at least two higher or more fundamental approaches: the generative approach to modular synthesis and the coded algorithmic approach to the computerization of sound control and generation.
Generative Sound Synthesis
This is a paradigm of electronic music that’s really ever only received anything like a formal definition in computer music thanks to the apparent popularization of the generative music concept by electronic music pioneer Brian Eno. Yet within the predominantly analog world of modular synthesis, it seems to have taken on a second life. Readers of this column may recall my dropping an article in the previous issue which identifies an emerging trend among synth-oriented video-streaming producers to merely set up a patch and, for the most part, just let it run while they record the device itself as it produces a typically ambient collage of sound, animated only in the satisfying flickering of LEDs.
This technique has a very particular meaning and relevance for modular as an analog approach to sound synthesis, because while any module involved in this might have a digital or even computerized quality, they nevertheless interact in an analog fashion over the patching of those modules; and of course it’s in the patching that the generative magic happens.
This approach generally seems to benefit from an abundance of experimentation, know-how, and equipment, and thus it seems to me somewhat of a luxury for committed and well-established enthusiasts, at least on the hardware side. But whether through some modest quantity of available hardware or some nigh-infinite range of virtualized devices, I do feel the practice to be an essential diversion for any producer in the modular realm to dedicate some degree of study throughout their exploration and experimentation therein, particularly if they hope to reach into the higher potential of that particular ecosystem (which I certainly do).
Algorithmic Sound Design
At bottom, any sort of computerization in music is just a flow of binary information which is at some point interpreted into an analog signal or some likely discernible aspect thereof. This is a process which, by practical necessity, must be obscured for any possibility of meaningful human interaction with that flow.
In the realm of audio, the modern form of this obfuscation typically aims at achieving maximal complexity with maximal automation via some slick and approachable GUI, usually designed to resemble familiar hardware form-factors (skeumorphism). The forerunner of this extreme level of obfuscation is perhaps the considerably more computer-native software approach of tracker software which, though offering a bit more interactivity (albeit of a constrained scope) at the expense of some automation and auspicious similitude, nevertheless indulges a considerable level of obfuscation insofar as what’s going on under the hood.
So what is going on under the hood of any these variousclasses of audio production software? Simply put, what gets the bits churning deep down in the bare metal, what gets all those transistors flipping on and off, is code issued in some fashion assembled from a human-readable level down to a machine-readable level. And indeed, before most of the obfuscation we encounter today was even possible, let alone likely even conceived of, code was the way in which audio signals were made to conform to the logic of such machines according to the will of their users. It was only only a matter of time before this dark art was revived in the modular space:
Of particular interest to me is the Teletype, which allows for live and rapid issuance and execution of commands in coded fashion (oh, and we may soon enough get to play around with it in VCV Rack):
But perhaps of slightly greater interest to the matter at hand is a little something touched upon fairly extensively in the same article: Bytebeat.
Suffice to say, this makes more hardware-oriented approaches to low-bit, low-resolution computer sound seem a lot more analog and is perhaps no less a raw and almost brutalist approach to computer music. And of course, as featured in the previous article, here also do we have at least one option in the modular space:
The Well-Trod Path Less Taken
So what are we to glean from this little study we’ve undertaken regarding the electronic form of sound and music? Perhaps it’s that we’d be remiss to neglect or take for granted the diligent work done and still ongoing to push the envelope of the veritably cataclysmic convergence of technology of and music. Perhaps we’ve been in too great a hurry to embrace the fickle fruits of progress to stop and consider legacy borne forth therein.
We’ll take these matters into further consideration in future articles, to be sure, but for now, I’ll leave you with a clever little documentary to broaden our historical context a bit:
Haven’t read pt1? Read pt1, ya dingus:
