Abstract: This paper argues that despite the increasing popularity of analogue (subtractive) synthesizers, in-depth studies of synthesizer performance practice in popular and experimental music are under-represented in academic research. Special features of the analogue synthesizer are briefly defined within the broader context of electronic music instrument research. By examining recordings of live and online video synthesizer performances, the paper explores different strategies for technology-centred audiovisual analysis of music. Where instruments can be clearly defined (as in the case of most non-modular or semi-modular synthesizers), in-depth analyses of performance practice are presented. Each analysis reveals the specific framework and affordances of the synthesizer used in each case, and the different musical approaches to the same instrument across a range of video performances. Three main modes of performance practice are identified: keyboard-centred, parameter-centred (spectral) and sequencer-driven. Musical examples are used to demonstrate how modes of performance practice overlap, often incorporating elements of two or more categories.

Bio: Ewan Stefani is a composer, musician and academic based in the UK. He has been composing with synthesizers, computers and tape recorders since the late 1980s. As a composer and performer he has worked closely with a broad cross-section of collaborators including Christophe de Bézenac, Stephen Altoft, Paul Hession, Will Baldry, studio Césaré, and Centre de vidéo-danse de Bourgogne. His solo pieces have been performed on BBC Radio 3 and at various international conferences and festivals. As an academic, the main focus for his research is on the analogue synthesizer viewed from the perspectives of composer, musicologist, and performer.


  1. The next phase of the analysis process (not included in the video as I ran out of time):
    – audience analysis (the videos are YouTube, and the comments reveal the ‘enthusiast’ character of the audience);
    – the hidden digital features in the performance (each was made using Logic to send MIDI notes to the synth and to apply delay/reverb for example);
    – the aesthetics of the second performance play into the deliberate retro/vintage imagery and sound of the synthesizer (e.g. the black and white filter; shots of the Revox tape recorder, etc.);
    – timbre analysis of the instrument, to pin-point specific sonic characteristics in relation to other synthesizers;
    – analysis of what the positioning of the hands tell us about the performer’s thought process (e.g. the hands move over controls as the performer considers the next move);
    – linking together different methods of analysis to create a more complete multi-dimensional representation of each performance;
    – creating a searchable database of analyses and observations in the style of the rhizomatic methods of instrument definition as proposed by Weisser & Quanten (2011) and Magnusson, 2017).

    I’m planning to look at more complex instruments (including modular systems) and ensemble synthesizer works next too.

    I’d be interested to know of any good studies of analogue synthesizer practice that I may have missed out there!

  2. Thank you! This was fascinating! I don’t really have any good questions to ask, but just wanted to give you a huge thumbs up for the way you used the video format for analysis. One of the great struggles in doing music analysis in writing seems to me to be this disconnect from the music’s unfolding in time – and this format with annotations over music is a great example of how changes in media can really aid our analyses, both in content and clarity.
    Thanks again!

  3. Hi Ewan – likwise, this more of a statement about how interesting this is rather than a question.
    Having said that, though, there is a kind of question – in that I’m interested in the annotated video approach which quite a few people seem to be using. I have several PhD students exploring similar techniques and I’ve used some myself in an output on the C21MP webpage (http://www.c21mp.org/simon-zagorski-thomas/). I wonder if you have thoughts on the ways in which these techniques can / should be used as a tool of analysis as well as data capture?

    1. Hi Simon, I think these techniques are becoming increasingly important as more of the music (that I’m interested in) is being made for online distribution. These ‘enthusiast’ videos are great for analysis as they clearly show what’s going on. Other more artistic efforts usually obscure the details. I’ve been experimenting with 360-VR video and audio capture when working with small synth ensembles and that seems to be promising. The cameras are becoming cheaper and allow you to move around the 360-degree panorama quite nicely. If the group is set up carefully around the camera, it’s then possible to focus on each player to see what they’re doing without being overly obtrusive. I want to set-up performances with the specific aim of just capturing a lot of data, building up a large database of performances to compare. Until then I’m going to scour YouTube. I need to get up to speed on what’s been written about working with new media formats..

        1. Interesting! That’s a much nicer artistic, visual experience than my approach, which is really just for functional analysis of individual parts within ensemble imrpovisations. E.g. https://youtu.be/4IHxV_AT2Gk

          I like the idea of splitting 360 video into separate screens though for performance.

  4. Hi Ewan, cool paper, thanks for this. In my PhD project, a big piece of my work is going to be on the performance practices of dub sound systems, and working with an analog synthesiser interface has a lot of similarities with all of the channel strips and bits and pieces of hardware that soundmen typically use. In other words I think your work will have some trasnferability/relevance to other fields of electronic music, not just directly synthesiser music. Are you planning on publishing the work anywhere?

    1. Hi Ivan, there are definitely interesting links with dub and Jamaican sound systems practice; I agree! I’m reminded of Mulder’s NIME 2010 paper on ‘the loudspeaker as musical instrument’ for example. But much more work needs to be done on the use of acetate discs, and the sound sytstem in general. We had a PhD student (Ray Hitchins) in Leeds a few years back working on bass frequencies in particular, and his work looked interesting too. I’m planning to publish more in this area, starting with a book chapter in a book called ‘Rethinking the Musical Instrument’ coming out next year.

  5. Hi Ewan, I enjoyed this a lot. It’s a very interesting project! I have a couple questions:

    1) When you’re watching a performance video to ascertain how the performer is using the synthesizer, what is the relationship between visual, sonic, and embodied elements? The videos you showed here obviously prioritized the visual, with the full-screen shot of the synthesizer and the text; but as a composer and performer yourself, are you also using your ears? And how does your existing embodied knowledge of synthesizers (how you personally interact with a synthesizer) come into play?

    2) At the very beginning of your presentation, you mention that synthesizers are ‘simple devices that are fun to play’. Fun and enjoyment are motivating factors for a lot of musicians, all across the amateur-professional spectrum – as I’m sure you know from your own experience. Is there a way of incorporating pleasure into this kind of analysis? (Pleasure in listening, pleasure in composing and being creative, pleasure in interacting with a fan community, etc.)

    1. Hi Kayla, thanks for the questions – this has made me think..
      1. the visual information is being used in this case for confirmation of the parameters being manipulated on the instrument. So in some cases it confirms what can already be heard quite clearly. However, there are a few moments when the audible effect of the changes are so subtle that they might not get picked up from aural analysis alone. My existing embodied knowledge is definitely useful when doing aural analyses (where I can’t see what’s going on, usually), but I prefer to rely upon visual information primarily whenever possible as even with a good knowledge of the instrument it’s possible to be fooled by what I can hear. Some effects sound very similar, and I want to be as accurate as possible. Ultimately, there will be times when I have to make a good guess though, and that’s when the embodied knowledge will be most useful. One of the reasons to look at instruments that are still available now is precisely to be able to learn how to use them and experience what it’s like to play with them. I feel sympathy for those academics looking at instruments that no longer exist..

      2. Yes! This is a huge part of the ‘analogue experience’, and I need to look into the implications for flow and creativity more in the next phase. Definitely an important factor, and hopefully someone has already provided some nice frameworks to pick up and apply to these instruments. This will help to answer the question ‘why analogue’?

      1. Hi Ewan, thanks for the thoughtful replies. The hierarchy of input/senses that you’ve described is especially interesting and is making me think about how I take in this information myself. Great paper, and a really great project!

  6. Tags for this video: analog, synthesizer, performance, practice, Vermona PerFOURmer II, pop music, experimental music, audiovisual analysis.

Leave a Reply