Introduction to modular synthesis
During this hackspace I gave a brief introduction to modular synthesis. I first started by giving an overview on the differences between digital and analog audio, outlining the strengths and weaknesess of each. Then, I showed my eurorack setup and explained the principles that drove me when choosing the modules I own. Finally, the focus was expanded to the creative possibilities that modular systems offer when used in conjuction with digital audio environments like SuperCollider, Max or pd.
Digital vs analog
There is one fundamental difference between analog and digital audio: quantization. While on one hand analog signals are the result of continous changes in voltage, digital signals, on the other hand, happen at discrete moments in time - called samples. The amount of samples per second determines the sampling rate of a digital system: a sampling rate of 44100 Hz explictly specifies that 44100 samples will be generated every second by our system. The higher the sampling rate, the higher will be the resoultion of the resulting audio signal, as more samples per second will be produced, reducing the risk of unwanted effects such as digital aliasing.
All considered, what are the advantages and drawbacks of each? First of all, there is the notion of cost to take into consideration. Analog technology, while having better sonic fidelity, is considerably more expensive than its digital counterpart, as it needs the production of a single piece of hardware that it’s capable of only the one task it’s been created for. Digital systems, on the other hand, can often run on our everyday computers, without requiring any additional purchases. This also links to the problem of reproducibility. Audio software can easily repeat the same set of commands as long as the CPU can perform, allowing inexpensive - money wise - execution of, for example, 100 sine tones at the same time: definitely easier and cheaper than buying 100 hardware sine tone generators and patch them together!
The eurorack format
Eurorack is a specific format for modular synthesizers originally developed by Doepfer. The size of modular systems, generally,is expressed in terms of U (around 128mm) for height and HP (around 5mm) in length. In the case of eurorack modules, the height is fixed at 3U, while the length can vary from usually 2HP up to 40HP. As any other modular system, eurorack modules are housed in cases with different amount of rows. Eurorack modules draw power from a two-row ribbon cable connector containing either 10 or 16 pins, and they call for ±12V power, with some modules also requiring +5V. While eurorack is the most commercially common format, it’s not the only one. Other popular formats are Buchla‘s and Serge‘s.
As with other modular systems, eurorack synthesizers center around the concept of patching cables from the various modules, in order to create complex voltage interactions that will then produce the desired sounds. As previously hinted, being modular systems often completely analog, the voltage exchange between the modules does not follow a rigorous discrete order, but it all happens simultaneously. This allows for complex cross-modulations that are not possible in the digital world, as they cannot be expressed as a sequence of ordered commands, especially when introducing feedback networks in the patching (e.g. modulating a filter’s cutoff control pitch with its own output).
Hybrid audio: digital meets analog
It is possible to create hybrid systems that leverage on both the ease of digital audio manipulation and the complexity and richness of analog eurorack sinthesizers. The use of DC coupled audio interfaces, or, alternatively, eurorack modules like the Expert Sleepers ES-8, allow the possibility of sharing audio streams between the eurorack system and a computer, opening up for even more complex interactions between the two worlds.
This is a picture from my setup, which is built around the concept of being a noise generating machine. All the modulations are then routed from the computer to the system via the ES-8 module. On the digital side,my software of choice is SuperCollider programming language.
As an example, here a cubic interpolated noise function (
LFNoise2.ar ) is sent to the first output of the ES-8, and then used as FM input for the Instruo CS-L oscillator. In my setup, the outputs from SuperCollider to the eurorack system correspond to values
2..18 (16 in total), hence why here
Out.ar(2, ...) is used. The first line of code (
Out.ar([0, 1], ...) ) simply plays the output from the eurorack system to the speakers.