On Bronze

Lexxx is a music producer, Gwilym Gold a musician. Together they co-developed ‘Bronze’, a new music format that builds and transforms every aspect of a piece of music as it plays, to create a unique version on each listen. Gwilym Gold summarises what lead to the production of Bronze.

I realised the power of Bronze when I walked into the room to introduce the first public playback of the piece we had been working on, Tender Metal, and I found myself lost for words with nerves and excitement. Although I had performed in a fair number of circumstances in my life, I had never found myself struck this way. Never before had the presentation to people of a piece of my music (in which I had no active involvement, apart from to press start and stop) felt so distinctly like a performance. Is this because in a way it was a performance? And is it possible for more, if not all, ‘recorded’ music to exist in this way?

You’re probably wondering what I’m talking about. Let’s go back a little bit so I can give you a start on the thinking. Coming from a background in Jazz and improvised music, it had always seemed clear to me that a discrete sequence of events is not necessarily what defines a song or piece. In most circumstances, (and certainly before the advent of recorded music), when performing a song, a musician will – rather than trying to fulfill an exact sequence of events – be working intuitively within a set of rules. Such rules can have a bearing on everything from the detailed nuances to larger aesthetic decisions, depending on the parameters. A number of factors, including the circumstances and the nature of the piece itself will define how stringent these rules are. It is of course possible that the rules can be so open you end up with a distinctly different piece from one performance to the next. It is also, though, possible that if the parameters are set in a certain way, each performance, though different in detail, will retain some kind of identity from one performance to the next. I would venture here to say that the identity of a song or piece lies not in a specific sequence of events but within a margin of possibilities.

With that in mind, let’s go back to the piece I mentioned at the start, Tender Metal. In the early stages of production, myself and Lexxx were using a multitude of software processes alongside a lot of performance ideas in search of some kind of sonic aesthetic and mood to accompany a set of songs. Although what you might call a sonic palette and some kind of mood were building fast, it became increasingly clear that no one performance of a song felt definitive. When ideas, sounds, rhythms and textures were being performed and manipulated live around a song, it often felt distinctly more vital than when listening back to recordings of what we had done. It was also becoming apparent that the identity of each song lay in something more undefinable than a particular sequence of parts or sounds. It was through this process and these realisations that it became clear that the music should exist as a performance, as a living breathing entity and not as simply a document. It became starkly apparent that traditional recording and presentation of this music (as much as we love a lot of what that can be) was simply not the most compelling way for this music to exist. The existing formats began to feel very limiting for what we wanted to achieve. In fact, when considering the power of the devices that most people use for listening these days it began to feel increasingly strange to us that music formats have developed little, fidelity and length aside, since the gramophone. That people still go and see live music is proof enough that high-fidelity stereo sound files are not, and should not be, the end of the line for recorded music.

In many ways, our new music format Bronze was a natural extension of other software systems developed during various production processes to manipulate / augment / create sounds, and in fact all of these initial pieces of software were designed for one purpose: to release things from the ‘exact repetition’ that computers tend to impose upon a creative work. Everything within a conventional software environment is entirely repeatable – within any digital system, two identical inputs will both cause identical outputs. In the analogue domain, or with any process outside of a computer, nothing is perfectly repeatable, which means that throughout the thousands of processes encountered during the making of a record, each would be significantly shaped by factors outside of one’s control. In short, it always seemed to us that working outside of software almost always rewarded us with something deeper, less controlled and more unique, shaped not only by our ideas, but by all elements of chance and randomness that exist in the real world.

It was only later, during some early experiments that we discovered that software is also very good at performing something other than exact repetition. Software is perfect at making impartial random decisions – it can create the exact opposite of repetition without the biases that human judgement imposes – software is able to reliably replicate the chance and randomness that was missing from the existing software we had been using. The first pieces of software were all experiments with rhythm – applying both subtle and drastic variation to the timing of rhythmic elements to simulate the looseness in timing you might expect in a real performer, but applied to sounds only possible with digital technologies. We quickly discovered that whilst pure randomness was a great tool to introduce variation into precise systems, it was only when combined with the influence of a human bias that this random variation produced musical results. What we had been looking for was a combination of factors – human creative judgement augmented by authored random variation. 

Of course terminology gets vague and difficult when talking about this sort of thing but I hope you get the idea. I imagine everyone is pretty familiar with and can understand why recorded music came about and why it became so popular so quickly, but I think a more relevant question is why that is still the case when other possibilities are now available. Music itself is not something you look at or touch and is not something you can really own. Music delivery formats merely provide a vessel for the music to live from. This is becoming even more abstracted with the rise of music in ‘the cloud’; whatever your choice of delivery, the music is projected into the air around you and then it is gone.

Throughout the course of creating both the album, Tender Metal and Bronze, we were almost entirely unconcerned with the way in which we achieved what we wanted to hear, assessing only the nature of the listening experience. Any refinements to the software would be judged solely on their abilities to transform the experience in a way that felt effortless, so that the workings and process would gradually fade into the background and only the piece would remain. This may explain why Bronze has turned out so differently to many other generative technologies; we were never guided by a particular functional aspect, a specific process, or any reactive / interactive element. Instead, we continually refined how Bronze could allow us to improve the music we were making. 

To compare the capabilities of a delivery format to a human performance is of course a reductive and shaky comparison, but I think the point holds. Just as physical spaces are built and rebuilt (and debated over!) should we not be looking more closely at creating more fertile and flexible digital spaces in which music can exist? In both cases, all that really exists of the music is the sound while it is playing, and the impression it has left in our head, ears and body when it stops.