Magma and Sound Design for Computer Music

In the previous article, we presented Magma and used it in a very specific context: layering for lofi ambient music. If you would like to read the article again and see the video, please click here.
Now, in order to show the versatility of this tool, it’s time to move into a totally different field.

For this second Magma article, we will work with Computer Music.


Computer Music is the repertoire that includes pieces using computers for processing, transforming and organising sounds. This branch of electroacoustic music was born around 1963 with the publication, on the famous American magazine Science, of the article “The digital computer as a musical instrument” by Max Mathews. After this debut, things developed very quickly. In Italy, for example, Giuseppe Di Giugno, a physicist and researcher, created the 4A processor in 1974/1975. Capable of real time digital synthesis, the first version of this processor was able to generate 256 oscillators. Luciano Berio, who in those years was called by IRCAM by Pierre Boulez, requested an electronic instrument capable of generating as many voices as possible. So, Di Giugno went to Paris and started working on the next processors: 4B, 4C, 4X and the never realised 5A.

With CSound, a programming language designed specifically for generating and modifying sounds, I created some sound objects, made using several forms of synthesis, and imported them into Live so that they could be changed in shape using our plugin. This is very short-lived audio material. All of these objects can be used in cyclic form: while some of them are formally created and exported as loops, others can arise less traditional sound surfaces.

Let’s automate MAGMA

Magma can make these objects less “boring” by generating some small time evolutions. These evolutions will be entrusted to automations, i.e. the movement of one or more parameters of the plugin itself. The movements will be programmed, written in this case. So the DAW will later call them up by readout. The sound object we are going to modify is a texture. We start with the plugin in its default position, the one we find when we insert it into the channel. This way we can already hear a slight modification of the object, a sort of reverberating tail that makes it much less “synthetic”.

We automate the RAW button, to set a first variation in time, as if we were dealing with a hardware synth. RAW which, when clicked (it is off by default), deactivates the envelope follower applied to the input.

Now let’s move on to modifying our texture, again using automations, which can suggest a structure to the piece we are producing.

We use the EQ section by removing a quarter of the treble and adding a quarter to the bass (= 150% bass and 50% treble). Now the reverberant tail of the object will sound much more audible. Let’s increase the Memory, again through automation. To make these changes more ‘regular’ let’s program them within an even number of bars. We can count the number of loops, which, if constructed following the time grid, will be our rhythmic cage, within which we’ll be able to play by adding further elements, anticipating or delaying. To construct a rhythmic variation, we automate the Size potentiometer up to a value of 17ms. You can hear the original texture and its pseudo reverberant projection, but now it will be drier, more cavernous and more rhythmic.

Obviously, adjustments of the EQ and Temperature are also going to be influenced by the soundscape inside which such sound object is placed: it might be necessary to take the right measures when working on the frequency, in order not to create unnecessary “duplication”. The Temperature potentiometer, as already mentioned, highlights certain parts of the sound we are creating. Up to 50% it exerts a gain on the sound that leaves it apparently almost intact, but from 50% onwards it crushes the sample, giving it a very different frequency identity.

In the meantime, thanks to other two objects, I built a little bit of context, consisting of our texture varying throughout time, a bass guitar and a very small rhythmic-melodic phrase. In this simple piece, automations build an actual arrangement for the fabrication of musical phrases and semi-phrases.

Returning to the expressive possibilities that we can create with Magma, we can also automate the Rate with a narrower, shorter automation, so that the sound will have a different pitch.

It is possible to play with the Mid/Side pot to alter the stereo image of what has been created, to add more movement. On that note, we can add that all of Magma’s parameters can be automated, so by automating the Mid/Side on a sound object, we can give it more air or make it heavier, adding even more character to what to it.


We can now rely on the wilder side of Magma, literally raising the temperature, to attempt to dissolve the definition of our object. We take the Wet pot to the extreme and close down the Dry. Before the last minute of this little piece we use a very powerful expressive tool: silence. We create a stop and do this with the help of our plugin. Our texture, bass and pseudo-rhythmic will fall silent for a short time, but they won’t be leaving a digitally empty space, precisely because of the coda we have created with Magma. It will be a little bit more than a breath. Again, showing the thousand possibilities of this plugin, possibilities that we can choose to tame or set free, to surprise us when we least expect it.

Do you often use Magma with synthetic timbres? Let us know which are your favorite sound sources to process throughout Magma.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.