Music 109/209N | MAT 276N

Music 109/209N | MAT 276N
Special Topics in Electronic Music
Modular Synthesis
Spring 2017

Juan Manuel Escalante / Modular Experimentation


VI.15 / Turn


“Turn” is the final study for Special Topics in Electronic Music, 2017. Of all these studies, it is the only one that didn’t use any analog synthesizers.

The Patch:
The original sound material was recorded from a generative Euro Reakt patch. Several oscillators are triggered and automated from different arrays of LFOs and one Impulse Train. One set of oscillators are filtered through a driver. Its output is quantized and linked to the pitch input of a Harmonic Oscillator. The other set of oscillators are modulated through a Modern Paul Filter, its values are also automated by oscillators 3 and 4 of the previously mentioned Harmonic Oscillator. An automated compressor and a Freeverb are also included in the chain. A Vector Mix receives 4 different signals (A from the driver, B and C from the compressor, and D from the Freeverb). Two LFOs automate dynamically the Vector Mix values.

The Graphic Score and Visual Aid:
The graphic score was used as a tool during the entire process. It provided a sense of sound transformation during the study. The video presents the evolving graphic score and additional information related to the sound: automation curves, track and buses levels for each MixDown, and a couple of Processing sound-reactive sketches coded specifically for this study’s visual aid. See the following image for an entire description.

The Mutation:
Following the idea that “[…] from one sound, we can easily create a hundred mutations” (Roads, 2015), the Reaktor patch was recorded as a ‘starter sound’ using different states (or knob placements). From this point on, the mutation began in 5 main stages:

MixDown #1 - The starter sound was trimmed and arranged into five sections and tracks. Heavy EQ was applied to remove some clipping from the original recording. All tracks are sent to 6 different buses. This sends are all automated. These buses apply reverb and compression (PuigChild 670, Waves H-Reverb, MannyM Reverb and Abbey Road Plates). Each one of them is panned differently. The output is saved and then imported into the next stage.

MixDown #2 - Certain portions of the file are extracted and exaggerated –using EQ and compression– in very specific points (11 new total tracks). A few of them are slightly looped to generate a continuos atmosphere. Two of these tracks used a Morphoder to provide a cohesive acoustic thread throughout the study. Sound is then exported.

MixDown #3 - A few more tracks were added to emphasize different moments of the piece, mostly during the first half. This mixdown now truly spatialized the piece using heavy automation along 11 new buses. Some of these buses added distortion and saturation (using FF Saturn), some others introduced new harmonics (Doubler4), some others used the Arturia MiniFilter V with different values.A final bus used SoundToys’ Tremolator.

MixDown #4 - This Mix is closer to the final result. I used a different set of plug-ins from the Waves OneKnob Series to enhance brightness, pressure or fatness. These knobs are manually automated.

MixDown #5 - Higher pitched sounds from the last two thirds of the study are emphasized during this stage. A high pass filtered white noise (not present in the original sound) was also included. Minor sound transformations were also applied.


The final MixDown was mastered using BandLab’s mastering service.
(preset: CD-Quality).

[ Download Final WAV file- 60 MB] | [ Download original Reaktor file - 8MB ]



V.17 / April 20th


[ Download WAV - 45 MB]

Description & Patch:
This study was constructed from the last 13 minutes out of a 70 minute session, recorded on April 20th at Studio Varèse. Again, an attempt to use clocks. Each track appears as a diagonal line on the video above.The patch used heavily the Pittsburgh Foundation 2.0 Synthesizer, plus MN's Phonogene, and Echophon and four 4ms’ clock modules.

Below, all the annotations from the entire session:

Post-production: Moderate.
The entire study is completely re-arranged from it’s original 13 minute form. The first DAW file spatialized sounds and provided basic EQ. The second file added a main RV plus two additional ones on each channel (L & R), saturation, tremolo and the MiniFilter V by Arturia.




VI.20 / Phone Call


[ Download WAV - 29 MB]

This study was built upon sonic leftovers. They were recorded on April 22nd (2017) at Studio Varèse. 15 samples were stored on the Teenage Engineering OP-1 synth and drum banks. These samples were sequenced and arranged on the same device using it’s digital 4-track recorder. The study was finished on May 21st.

Graphic score:
Of all the sound studies, this was the only one where the graphic score was sketched after the final mix.

Post-production: average.
The track was sliced to create the three main parts of the study. The sound that connects each part (as a bridge) was later added by looping a specific sample of the original file.

Sound from the four tracks is being sent to different buses. This sends are automated. Each bus features moderate EQ and various H-Reverbs using a binaural placement. An FFSaturn subtle saturation is added in one of the buses.




IV.21 / La Espera (On Standby)


[ Download WAV - 45 MB]

Description & Patch:
This study was built in two patches:

A. Using the Pittsburgh Foundation 2.0 synthesizer, it started as a simple exploration on how the ADSR modules work. 80% of the track was recorded on one take. The "breathing" sound was generated by controlling manually one of the VCO-s FM knob. This track provided the rhythm and foundation.

B. The Patch expanded to use an additional Oscillator and a subtle echo from MakeNoise's Echophone. Layer by layer, new sounds were added to aid the build-up, using the OP-1. I always try to sketch the track, in order to have a general idea of the complete form. Once a new sound has been overdubbed, new symbols are then added to the score.

Post-production: moderate.
It was also divided in two. The first part added compression, saturation and a three different Reverbs. The precussive repetitive noises were automated and sent randomnly to different buses in different amounts, for the sake of variation. Automation curves can be seen in the video above. The first two thirds of the track are kept in the middle, after that point, the track is spatialized and sounds start to populate the stereo range freely.

The second part added final details. Amongst them are big stomping echo (and pitch-shifting) in very few locations, to emphasize change. During some parts sound is balanced in order to sound louder, brighter, silent or fat.




V.2 /Troposphere


[ Download WAV - 34 MB]

Description & Patch:
I arrived to this patch by accident. I disconnected the clock input from a fast rhythm I was working on, as I was trying to understand 4MS’ Clock Divider, Quad Clock Distributor and the Shuffling Clock Multiplier.

An oscillator is activated by Pressure Points, which is sent to PhonoGene. Background ambience is provided by Clouds.

Post-Production: Light.
The entire track was recorded in one take. However it was shortened from its original 5 min duration. Certain portions of the sound are exaggerated using EQ and panned using automation. Slight saturation added and 5 different reverbs.



IV.15 / Square Pulse


[ Download WAV - 27 MB] | [ Download original RAW file - 25 MB ]

Square pulses were recorded using the same layering technique over 4 tracks (using the Pittsburgh Foundation 2.0 synthesizer). The result: a continuous dull set of pulses –with a specific rhythm–. I took the challenge of molding this starter-sound, into a small study of rhythm and transformation.

Extremely simple. Heavily filtered squared pulses (using low-pass and high-pass).

Post-production: high (two parts).
- Part I. Same spatial bus configuration from our previous “Fire Engine” was used (see upper line of the video). Each track was automated in order to send their signal periodically from Left to Right, creating movement. Each bus had a different type of reverb. Four main ‘waves’ of sound can be heard on the resulting audio file. This waves are ‘released’ for the last 40 seconds of the study, and the sound plays without any interruption. Some fragments were exported as loops. All the mix was then rendered and placed on a new file.
- Part II. On this new file, the exported loops were placed in different parts, in order to aid the build-up of each wave. Different plug-ins were used to generate a fuller spectrum of rhythm: three tremolators in different configurations, subtle saturation, and two types of reverb hard-panned to L and R. High-cut frequencies are also automated right after wave no.3 for a short period of time. Automations of the main track, can be seen in the video above (to the right). The final rendered study was mastered using BandLab’s mastering algorithm (preset: CD Quality).



IV.17 / The Domestication of Noise


[ Download WAV - 37 MB]

Compared to previous studies, sound is freely changed and shaped during these recordings. Each layer emphasizes one specific sound using the Pittsburgh Foundation 2.0 Synthesizer. LFO Cables are continuously connected and disconnected during the process. All sounds are recorded on the OP-1's 4 track digital recorder.

Patch: Semi-simple.
More effect modules are added into the mix (Echophon, Piston Honda, MultiMode, Intellijell's µFold).

Post-production: moderate.
The four original tracks are spatialized and sent to various buses that contain: 3 different reverbs, 2 saturators, 1 heavy compression &1 brightening effect. Specific passages are duplicated on different tracks to emphasize (using EQ) certain parts. Panning is done manually.



IV.11 / Fire Engine


[ Download WAV - 41 MB]

Simple square waves sampled from a Pittsburgh Foundation 2.0 synthesizer. Each sample (total of 6) was recorded and layered one by one using a 4-track digital recorder. During these recordings, a LPF was applied in different amounts. Sample speed and envelopes were also individually controlled. Each layer's volume from the second part of the experiment, is controlled manually (from 0 to 1). The resulting clipping sound of each sampled loop provides a specific rhythm on the first half. During the second half, the tempo used is: 93 bpm.

Sample 3.aiff Sample 4.aiff Sample 5.aiff      
Sample 6.aiff Sample 7.aiff Sample 8.aiff      

Extremely simple, square waves being triggered by an LFO (no photo available).

Post-production: moderate.
- 4 types of medium and long reverbs are applied for the first half, panned to the L, CL (center left), CR (center right) and R.
- For the second half, each of the four layers has 4 compressed busses with a specific Binaural placement on each one.
- Automation curves can be seen in the video above over the audio wave forms. To the right, the sampled waves. On the bottom each track's level (main, reverb and space location), to the right, bus send level for each of the 4 main tracks.
- BandLab's mastering algorithm was used to generate the final audio file (preset: clarity).


IV.09 / Pittsburgh Drone


[ Download WAV - 38 MB]

1st attempt using a modular system sound. Only a few basic modules were used from the Pittsburgh Foundation 2.0 system: 2 VCOs, 2 LFOs, 1 FCV. Each sound was layered using a 4 track digital recorder (from the Teenage Engineering OP-1 synthesizer). Track #1 & #2 are centered, #3 is panned to the L,# 4 to the R. No other effects added. Volume and filter cutoff controls were manually executed on the Pittsburgh.

Extremely simple. Some FM was applied using oscillator #1 into oscillator #2. All filter outputs were explored.

Post-production: low.
A subtle reverb was applied to the main mix along with basic gain and a modest limiter on the master channel.

// M M X V I I