MUFF WIGGLER Forum Index
 FAQ & Terms of UseFAQ & Terms Of Use   Wiggler RadioMW Radio   Muff Wiggler TwitterTwitter   Support the site @ PatreonPatreon 
 SearchSearch   RegisterSign up   Log inLog in 
WIGGLING 'LITE' IN GUEST MODE

AI'tude Suite: a playable patch featuring 296e, 272v, 285e
MUFF WIGGLER Forum Index -> Buchla, EMS & Serge  
Author AI'tude Suite: a playable patch featuring 296e, 272v, 285e
Michael Tiemann
While djangosfire has been lighting it up with his 296e/285e explorations, I've been on a separate quest that has yielded some results. I just posted a 20+ minute performance of a playable patch I call "AI'tude Suite". Here's the patch description I posted with the video on YouTube:

What if an AI tried to learn music by playing a Buchla 200e synthesizer, starting with a fundamental drone, the harmonic series, and then exploring from there?



The fundamental structure of the patch (and its theory of operation) is as follows:

A 262v Harmonic Oscillator functions as a kind of aeolian harp, strummed by a Fluctuating Random Voltage (B) from a 266e Source of Uncertainty, which in turn drives the frequency of a 259e Complex Waveform Generator, analyzed by a 296e Spectral Processor. All octave harmonics (yellow banana cables flowing down) go directly to the harp, whereas all other harmonics (fifths, major third, dominant seventh, major second) feed to a 210e Control Voltage and Signal Router. By selecting which of these harmonics are selected or not (and in what order), the harp can be "tuned" to different modes (Ionian, Dorian, Phrygian, Lydian, Mixolydian, Aeolian, Harmonic Minor, Melodic Minor, Phrygian n3, Lydian b7, Major Pentatonic, Minor Pentatonic, Octaves only, Octaves + 5ths, etc.)

The harp's output can be played directly, which sounds a bit like a pipe organ in timbre. But also feeds a 285e Frequency Shifter and Balanced Modulator, which can make it sound like chiming bells. In this performance I crossfade between the harp and the bell tones (or mute them entirely).

A 261e Complex Waveform Generator provides the lead instrument sound, with variable timbre, symmetry, and high-order harmonics. The top-most unit of a 255 Control Voltage Processor selects between treating the harp or the lead as the instrument in charge. Rhythmically, the harp is driven by the randomness of a Fluctuating Random Voltage (A) from a 266e Source of Uncertainty, whereas the lead is driven by self-cycling 281e Quad Function Generator (but also governed by an Eardrill Pulse Balloon). When the harp is in charge, the lead can follow it rhythmically, or, using the OR function of the 281e, the lead pulses can be OR'd with the pulses derived from the fluctuating random voltages to create hybrid rhythms.

The 256e Control Voltage Processor processes pitch ranges and pulses to drive the Quantized Random Voltages and Stored Random Voltages of the 266e Source of Uncertainty. Different distribution and time correlation parameters govern whether the Quantized Random Voltages select truly random values from the Number of States, or whether the selection is more a question of randomizing the magnitude and direction of the next selection based on the current state. In any case, the selected voltage selects a stage of the 250e Dual Arbitrary Function Generator. Each stage provides three control voltages: time (which feeds the Decay parameter of the harp's 281e function generator), and two control voltages that represent the low and high range of the pitches that the 256e will use as its pitch range.

The 5 other parameters controlled by the 255 (as given by their index in the unit are): 4-green = pitch of the frequency driving the harp; 5-purple = pulse balloon taps to inhibit lead; 6-yellow = high order harmonics of lead; 7-brown = symmetry of lead waveforms; 8-blue = portamento (or not) of harp pitches. For this performance I used a Gray Code to cycle through 8 combinations of pulse taps/high order/symmetry before controlling them directly according to my own AI.

A 263v Quantizer/Shift Register gives us the modes and scale degrees that give shape to all the random improvisation coming from the Source of Uncertainty/Dual Arbitrary Function Generator. The shift register allows us to keep separate the pitches for the harp/bells and the lead and to keep them stable until a pulse tells us "time for the next note".

What I have learned in all of this is that a little theory goes a long way.

The video:
djangosfire
.......EPIC we're not worthy
mestlick
Really nice result!
Do we hear the 259e? Have you tried playing other sounds into the 296e?
Michael Tiemann
mestlick wrote:
Really nice result!
Do we hear the 259e? Have you tried playing other sounds into the 296e?


Thanks! You don't hear the 259e because there is no audio output coming from the 296e. I did play a little bit with the idea of using more complex waveforms, which would have resulted in more harmonic tones being played at the same time. I stepped away from that for three reasons: first, because I really liked the clarity of having essentially two monophonic instruments playing in unison and/or counterpoint. A more complex tone going into the 296e would mean more harmonics playing simultaneously, which would muddy the lines that the 262v was drawing. Second, because the 262v audio output saturates pretty quickly when many harmonics play at once, and that saturation also took away from the clarity of the tones when only a single harmonic played at a time. Third, at ~20 minutes, the suite had already become pretty long. I didn't want to further extend it with yet more combinatoric examples, so I instead tried to focus on elaborating what I had within a smaller (but still large!) range of parameters.

But indeed I think there are some fruitful ways of exploring exactly this: to play power chords through the 262v and to bring (tuned, complex) audio out of the 259e. I think one of the keys to this is to find the right way to put the sound of the saturated 262v into the mix. If I can figure that out, I'll post another performance.
mirf59
...

Amazing. Wins an award for thoughtfulness and depth. Awesome.

....
Michael Tiemann
Glad you all liked this! If I did another video, how many would prefer a different performance, vs. how many would prefer a tutorial that explains some (or all) of the elements of the patch?
monokinetic
Strange nobody else replied! I found your last instructional video about pulse trains on the 281 super useful, so I'd vote more tutorial a la "Buchla (all but the kitchen) SYNC Explorations".

But I've enjoyed everything you've posted so far! Mainly posting to say just keep 'em coming and thanks for sharing your ideas smile
pyjamarama
wonderful stuff. this is the kind of creative patching that brings out the true beauty of these instruments.
Michael Tiemann
monokinetic wrote:
Strange nobody else replied! I found your last instructional video about pulse trains on the 281 super useful, so I'd vote more tutorial a la "Buchla (all but the kitchen) SYNC Explorations".

But I've enjoyed everything you've posted so far! Mainly posting to say just keep 'em coming and thanks for sharing your ideas smile


Thanks! I'll try to crystalize a few insights worth talking about...
doepferiano
fantastic ! thumbs up
eldancer
Joli!
it remind me the feel of A Rainbow in Curved Air by
Terry Riley
Michael Tiemann
eldancer wrote:
Joli!
it remind me the feel of A Rainbow in Curved Air by
Terry Riley


Wow! High praise indeed!!

https://en.wikipedia.org/wiki/A_Rainbow_in_Curved_Air
BooleanYulian
Fantastic !!!
Michael Tiemann
Here's the breakdown video I shot earlier this month. Could only publish now because I had a 2 week trip to Dubai in the middle of the month.

Michael Tiemann
Don't be shy--would love to get some feedback on the tutorial. Too high-level? Too granular? Too much resolution? Not enough resolution?
MUFF WIGGLER Forum Index -> Buchla, EMS & Serge  
Page 1 of 1
Powered by phpBB © phpBB Group