I configured my Dolby Atmos Renderer for 11.1.10 playback at Stanford CCRMA in preparation for my talk later this year. You too can go beyond the 9.1.6 configuration (which most everyone assumes the Renderer is maxed out at) – you can actually push it to 11.1.10.
Industry standard meets ultra-custom immersive audio setup
The dichotomy of academic research focus and industry standardization is almost amusing to explore. Have you ever walked into a prestigious academic research institution where experimentation is everything, and felt like uttering an industry-standard format is blasphemy? And vice versa, have you felt way too nerdy to bring up your experiments in highly-customized systems (that would never fly in the commercial world) at an industry event? If you answered yes to either, you know what I mean.
11.1.10 Dolby Atmos speaker configuration at Stanford CCRMA – from front to back view. In preparation for my Spatial Composition talk. Photo by Franck Martin.
Navigating between these two professional environments requires respect, understanding, tact, translation (sometimes literally) – and a “why not!?” attitude. But if you navigate them with success, the results may give you the best of both words: experimental yet practically applicable.
When I configured my standard Dolby Atmos Renderer for 11.1.10 immersive playback at Stanford CCRMA in preparation for my talk later this year, I knew it wasn’t a “first” by any means. Yet, it sent minor shock waves through the community of immersive audio professionals in LinkedIn. As it was received with a mix of surprise and doubt, I will detail the setup below.
Production and Playback in Dolby Atmos: Why Stop at 7.1.4 or 9.1.6?
The CCRMA Stage offers 56 discreet channels (64 monitors) in a highly flexible arrangement – but doesn’t follow any standard formats, like Atmos (for good reasons, mainly freedom in research). Contrary to popular belief, you can go beyond 9.1.6 monitoring through the standard Dolby Atmos Renderer.
11.1.10 Dolby Atmos speaker configuration for my Spatial Composition talk with 22 discreet channels.
Thank to several ABV-networked interfaces, the wonderful colleagues at CCRMA and I were able to configure 22 discreet channels for a quite detailed Atmos playback. As of this writing, we are thinking about adding 12 more monitors to reach the 34 limit of the (non-cinema) renderer in array mode. (More about the pros and cons of this here).
Myths vs Actual Capabilities
Just like most colleagues, I also thought up until a few days ago that Dolby’s “normal” (not Cinema) Renderer only supported up to 9.1.6 discrete channels and up to 35 speaker outputs in array mode – and then still capping discreet outputs at 9.1.6.. This is false. Yes, 9.1.6 is a standard channel configuration, but “standard” does not mean “maximum”. This week at CCRMA we set up and tested 11.1.10 in the Dolby Atmos Renderer without using any arrays, only utilizing discreet outputs – i.e. not achieving extra spatial locations with proximation or “phantom sides”, but proper 11.1.10 playback. The Renderer allows this as long as the audio interface (or networked interfaces) are allowing it. (CCRMA’s Stage is using 6 MOTU M8s networked via AVB). While in my presentations at various Dolby Cinemas I had the luxury of using the Cinema Renderer, for this configuration we used the “regular” Renderer that we all know. (The Cinema version provides expanded channel count, however, it is only available for theatrical installations.)
You Can Use 22 Discrete Channels Right Now
Go to your Dolby Atmos Renderer’s Room Setup and enable all speakers (without turning on non-array mode). You might be surprised to see that the distribution of speakers on the interface is front-heavy with those extra addressable channels. I believe this is meant for top high-end home cinema playback, i.e. film sound.
11.1.10 Dolby Atmos Renderer maximum speaker configuration for Spatial Composition at 22 discreet channels. Note the non-array mode, the discreet output channels per speaker, the output name abbreviations, and the clustering of speakers in the front – necessitating the physical re-arrangement / re-calibration of speakers (otherwise the front area is too dense for music playback). Our playback tests confirmed the discreet playback to each speaker as shown in the renderer configuration here.
It can make sense for music, though, in two instances:
• if you are using it throughout your composition and production phases and take advantage of a more detailed stage up front. It also gives you increased spatial resolution in the overhead channels. It translates perfectly to other 11.0.10 or higher systems, and should play back well in 7.1.4 and 9.1.6 as well. Just have to remember to compensate with the higher sum of energy in the front cluster.
• if you can intentionally reconfigure the whole room for an evenly-spaced arrangement, i.e. repurposing those extra FR channels and adjusting your physical setup accordingly (also re-calibrating it). This is the way to go if you don’t care about translation to other systems, i.e. you are composing for a specific site (e.g. music and sound installations).
We went for the 2nd option and somewhat rearranged the playback space. For this reason, the Renderer’s UI shows the theatrical 11.1.10 configuration, but the physical arrangement of speakers looks a bit different. In the future, I will likely compose a piece with the first use case of 11.1.10 in mind.
11.1.10 Atmos configuration at Stanford CCRMA – front stage and front height channels view. In preparation for my Spatial Composition presentation.
In Summary
I am not the first one to make this configuration happen by any means – others have done it with very expensive home theater receivers and AVB network switches. What this experiment taught me is that custom high-end systems can allow us to ignore some standard practices and push the capabilities of the software to its actual limits.
My talk at Stanford CCRMA will be on November 4, 2026. Until then, some more testing is planned – and I will update this page with new documentation and audio+video.
Tl;dr – it’s quite easy for anyone to get into spatial composition that reaches the audience today. While other formats are also available, Dolby Atmos covers the full pipeline from synthesis and composition all the way to distribution and playback. It yields the power to express and absorb original musical ideas, using space as the 5th element of music. This immersive music guide explains, that in order to get into spatial music,
• Creators only need a DAW that integrates Atmos (free in Pro Tools, Logic, etc.)
• Listeners don’t need anything more than what they are already using to listen to music (even just a pair of regular headphones)
• Professionals wanting to go beyond the 9.1.6 configuration (which most assume Atmos is maxed out at) can actually push it to 11.1.10 (see Part 2).
Since the completion of my Spatial Composition research, I have been focused on developing related new techniques, refining aesthetics, and demonstrating its potentials at various spaces around the country – including at Dolby’s own three screening rooms (currently the only three in the world) in San Francisco, Hollywood and New York City.
As I’m planning to take the concept on an international tour, today I have stopped to reflect on how (relatively) easy the road has been to get here – although it was a long journey that started around 1994. What started it was the impressive, now-famous Dolby Train trailer. It was screened before movies for several years in movie theaters. It’s a long story of mine that I tell in my talks, involving the first surround sound-equipped movie theater back in Budapest (Sport mozi), a projectionist, some wine and my emotional reaction to sound (barely music) moving in space. I always saw space – space as in spatial location and movement – as a synthesis parameter. The problem was that no way to control it in a usable, sharable form existed up until 7-8 years ago.
(One notable surround synthesis exception one might think of is the Hartman Neuron synthesizer from 2001; it had spatial joysticks and 5.1 outputs, it was compact, far more user-friendly than software environments like Max/MSP, but also rare, expensive, under-supported… and after all, most surround music created stopped after the production stage before reaching distribution, as there was no feasible path to take surround compositions to consumers to experience its spatial design). While I had the concept of spatial synthesis and composition in mind for a quarter of a century, I wasn’t able to express it for my audience.
At Dolby’s New York City screening room before by Spatial Composition talk
Accessible Immersive Music Tools
Jumping forward by a few decades, I got re-energized by the possibility to compose in- and for Atmos – way before it was done by anyone (it still isn’t really), while I was mixing the forgotten future W1 album in Atmos (which was the newest thing to do) back in 2019. Sitting in a studio at Dolby HQ, I had realized how much the meaning of textures and melodies changed depending on their location and movement in the 360-degree dome space. This is when I set out to shift my focus from mixing in space or arranging for space to using location as a core, conceptual part of composition, not an afterthought during the execution of such concepts.
Aside from the delays caused by the covid lockdown, the journey has been quite smooth. The tools weren’t as user-friendly in 2019 as they are now, but fortunately Atmos was created with the user, not the researcher in mind. So, if you really wanted to figure it out, you certainly could. It doesn’t take more to get into Atmos or other immersive formats today than it takes for someone to learn a new DAW (of course, knowing a DAW first is expected). The power of the tool – and the ecosystem that it opens up for immersive music – is no joke. So much so, that the data from my research led me to think of (and subsequently publicly promote) SPACE as the 5th element of MUSIC (joining rhythm, melody, harmony and timbre).
This is big. Maybe almost as big as music concrete was 75 years ago (which eventually birthed electronic music). Of course, this isn’t a new invention: I’m merely picking up the true concept of spatial composition where Varèse left it off; I’m objectively testing, pushing the boundaries of the concept and applying it in practical, accessible ways using today’s new technologies and ecosystems.
For Creators: a New Way to Create Original Content
This section assumes that you are already familiar with the basic concept of object-based spatial audio. If not yet, head here first.
Why do it?
Instead of trying to race against the ever-expanding camp of amateur and professional songwriters, lyricists, composers, musical experimenters on their quest to craft an original message using the traditional elements, you might want to add an entirely new category of expressive ingredient to your writing. (Not to mention racing against ai, which recycles the traditional elements of music in new combinations – but would never think of going outside of sounds, lyrics, melodies, rhythms and harmonies.) Adding the 5th element, space, to your writing (not just production!) process opens up a whole new dimension (pun intended) in the expression of emotions and thoughts through music. This is not something that can be recycled or machine-learned, since the number of historical precedents of the conceptual use of spatial movement in music is close to none.
What do you need?
Only a DAW that already includes the Dolby Atmos Renderer. Currently, Pro Tools, Logic, Cubase, Nuendo do this natively. For users of other DAWs, you can get the standalone Dolby Atmos Renderer to connect with your DAW. (If you’re not using any DAW just an instrument or paper & pencil, talk to your engineer 0 see below).
This solution will give you up to 11.1.10 channels and 34 speakers for composing. How many have so many speakers, you might ask. The good news is that you have a binaural option, that is, working in your regular headphones and getting the impression that your sounds, musical elements surround you. Binaural render algorithms have been getting really good, too. While the binaural experience (which the Atmos Renderer also provides) isn’t the same as the physical experience of 34 speakers around you, remember that the concept behind spatial composition is all about original expression, not production. And, in my experience, binaural monitoring can take you into that 5th dimension 80% of the way as you are conceptualizing and writing your composition.
How about Engineers?
Aside from the basic technical know-how, the main skill you need in order to mix spatial compositions is communication. That is, listening to the creator, understanding and accepting their vision, their spatial concept. If they composed in- and for space in a DAW, take their spatial automation (export from their DAW into yours). Can they only describe their concept? Listen and take notes, then apply it to the mix. If the author doesn’t have a concept at all, take the opportunity to co-create one with them. (Side note: in my research I put down a solid logic as to why engineers should get writer’s co-ownership for co-creating spatial content – another discussion for another time).
For Listeners: How to Experience Immersive Music?
Why should you care?
Because it can be more expressive in ways that you haven’t experienced before. Also because it can enable your favorite composers, songwriters to create original content that otherwise they would not think of creating.
Isn’t immersive music just a gimmick?
Well, this is where the difference between a spatially composed piece, a smart immersive mix and an average spatial mix differ in big ways. A mixing engineer working in Atmos (or Sony 360 or Eclipsa or ambisonics or in any of the other formats for that matter) has the option to move musical elements around the space as they wish – even randomly. They have the option to move instruments around to their own taste, but do so without following any concept (defined together with the creator or solely by the the creator of the piece) will be amiss. This approach may weaken the original concept (if there was one) – and an acoustic guitar or a shaker flying all over in space for no apparent reason can definitely feel gimmicky. And very often it does and becomes nothing more than a senseless showcasing of what technology can do.
But, following the concept of the author during production can help express and even strengthen their concept. A good spatial production materializes and locks in the emotions via space, that the composer initially felt during the conception phase. This is how an immersive mix can reach its potential and provide a stronger, aesthetically far more cohesive experience for the listener – exactly the way the composer intended to convey them.
What do you need?
Similarly to creators, you have a couple of options: lots of speakers or headphones. You don’t have to know which music file to choose: streamers recognize what system you are using and select the correct format for you ((e.g. the spatial audio version for an Apple TV or the binaural version for your headphones). Tune into Apple Music or Tidal (make sure that your subscription covers this feature and that the track you select is actually available in Atmos), and you’re off to listening to immersive music. Well, at least impressively-sounding audio. True spatial composition? That;s much less likely, as there is hardly any available, so far, but it’s coming.
Until then, a few music recommendations, in growing order of proximity to true spatial composition:
Dave Grusin: West Side Story (1997) – surround arrangement and mix
Jean-Michel Jarre: Aero (2004) – surround re-mix / 3 tracks of surround concept
Robert Rich: Nest (2012) – surround concept and mix
Peter Gregson: Patina (2021) – spatial concept / arrangement and immersive recording, mix
Justin Gray: Immersed (2025) – spatial concept / arrangement and immersive recording, mix
Isao Tomita: (1976/1991 release) – spatial synthesis and re-composition (adaptation of Holst)
forgotten future: Circular Time (2023) – spatial synthesis
Julius Dobos: Spatial composition research (2025) – spatial composition and mix
Edgard Varèse: Poème Électronique (1958!) – spatial composition and performance
(unfortunately only approximate recreations are available)
There is a lot of talk about how a.i.-enabled tools will destroy music and the music industry – maybe even more conversations that the topic truly deserves. Most articles on “will a.i. replace real musicians” fail from the start by not defining what success means for musicians. Number of streams? Audience recognition? Getting an award? Artistic fulfillment?
Let’s state the obvious: generative a.i. tools certainly make it easy to generate the kind of homogeneous musical content that such tools were trained on. They are excellent tools for hobbyists to enjoy the process of generating an output that they (more and more comfortably) label as music. Their main features, convenience and ease of use, naturally come with a consequence: these tools excel at their ability to increase anyone’s productivity – as long as we measure “productivity” by quantity, speed and efficiency. And while the industry might use such metrics to measure success, it is not the primary criterion for most self-respecting artists.
Tech companies are the least impartial to express how musicians think and offer their advise on how creators can succeed. Some of their business narratives are so narrow-minded, demagogic or deeply misinformed (like this one), that reading them makes me feel frustrated and ashamed to be a human at the same time. I don’t have a direct stake in the a.i. game, but as a composer whose personal and professional life has revolved around the musical arts, and also has has been using, testing, developing, lecturing about music technology for over thirty years, I felt inclined to share my angle. The goal of this article is to examine everyone’s unique role in shaping the future of music and the music industry – even the role of the audience.
Trends: Music and the Music Industry
To begin, we must separate the concept of music and the music industry – which is highly relevant to separating how “success” in music is measured by different stakeholders. While music and music industry are often discussed together, these are quite distinct concepts; sometimes even seen as dynamically contrasting ideas. It is widely known among music professionals and even among today’s well-informed listeners, that behind the phrase ‘music industry”, more “industry” than “music” has been hiding for over a century. Most musicians argue that this uneven ratio has gotten even more out of balance over the past decades – and often blame the industry’s greed for it. (I don’t disagree about the detrimental effect of corporate greed, but would also add most creators’ egos to the core causes.)
The critics of this view regularly cite classic-era composers to prove their point, arguing that the “industry” aspect has always been part of the success of well-known composers. While it is true that networking, promotion, etc. were partial reasons behind their popularity, the music marketers of today often overlook some important details. First, Haydn, Mozart, Beethoven (or Bach, for that matter) might have been excellent service providers and promoters, but they also had to be quite good composers to begin with. It wasn’t their tools, rather their talent, skills, personal approach and experience that gave them the ability to create original, quality content. Most average composers, even if they had excelled at networking and promoting their average content, wouldn’t have survived in the music business of the 18th Century.
Secondly, the opposite is also true: there have been tons of original, high-quality content that didn’t see the light of day due to lack of promotion, keeping their creators hidden in the shadows of music history, yet, their lack of fan-base did not diminish the greatness of their work. Unlinking quality from popularity shows how the success of a composer measured in original and high quality output has little to do with the success of the same composer measured in financial goals and popularity. Therefore, the effects of a.i.-enabled tools on music vs the music industry should be also measured distinctly.
Change in Music Technology is Inevitable, but…
I happen to believe that it isn’t technology itself, but the ways new technologies are designed and encouraged to be used are what have resulted in the generation of more junk than even pop culture can swallow. Most (not all, but most) toolmakers focusing on profit via providing musicians with convenience and efficiency has been driving the direction of the music tech industry for at last 40 years. Whether we blame the designer or the user for this direction, a.i.-enabled tools are taking this to the extreme. They will erase some musician careers but might also be part of the solution for a much needed curation in the musical arts – a sort of natural selection. Let me explain.
In the 1960s it took as much (or more) skill to use a new guitar or a new synthesizer than a guitar or piano a hundred years before that. However, this can’t be said about the recent decades; too many tools since the 1980s have aimed at convenience, replacing talent and skill. For composers, performers, songwriters, such idiot-proof technologies have come in many forms (e.g. midi-loops, chord analyzers, auto-tune, trackspacer, auto-mixers, etc.). The democratization of distribution services (redesigned to support streaming, the most destructive model ever created), while beneficial for professionals, saturated the market with heaps of sub-standard content, diluting quality and the perceived value of music itself. (Spotify took this to the lowest level by enabling users to record any full-length album for 1.8 cents).
The most frequent argument I hear in favor or a.i. tools in music is that technology has always shaped music, whether it’s the evolution of electronic instruments or digital production tools. Some use cliche examples, like drum machines that didn’t replace drummers, sequencers that put keyboard players out of work, auto-tune that didn’t replace vocalists, and so on. These are quite poor comparisons that got old years ago. Where they all miss the point: virtually all of these past technological inventions changed the tools we used to create music, or effected how we used them. However, a.i. is not changing how a creator works, but replaces the ideators, the sources of musical concept, the creators themselves. It is not technology that reforms “the old ways” – or any ways; it is technology with the power to replace the user behind it.
Let that sink in for a second, please – this is why a.i. tools are in fact dissimilar to all disruptions we have dealt with before. Mix in the modern human’s desire for convenience, instant gratification and hunger for self-validation, and the result is a further shift of the (already upset) balance from carefully-crafted musical art to efficiently mass-made products. But, there is also a silver lining.
Use or Resist a.i.-enabled Tools? It Depends on Your Goals.
Different people have different relationships with music. Their goals and the means to reach them can greatly vary as well.
I’ve discussed a.i.’s effect on music on various industry panels (most recently in AI vs the Audio and Music Professional in 2020 and in 2024, resp.) with diverse sets of experts. They included Maya Ackerman (computer scientist), Jonathan Wyner and Shahan Nercessian (iZotope), Alex Wankhammer (sonible) Nahuel Bronzini (mixing engineer), David Bowles (recording engineer), the late Steven Albini (record producer), Ken Felton (immersive audio technologist), Jean-Marc Jot (immersive audio technologist) and other professionals from music, technology or the music business. Back in 2020, we disagreed about the looming negative impact of a.i. on music, which became reality by 2024, the year of our second panel. Our second panel’s most pessimistic predictions became reality by mid-2025. Most tellingly, proponents of a.i. tools in music focused on benefits like convenience, business opportunities and the democratization of access (music as hobby), while panelists opposing the use of such technologies were more concerned with its negative effects on deeper human expression and ultimately, cultural standards.
I’m all for the democratization of tools of self-expression, but I do not support the false premise that new tools will turn casual users into artists. While I do not think that art should be reserved for a qualified few to create, I do believe that random products of casual self-entertainment should be distinguished from thoughtfully conceptualized & expressed works of art. Users who rely on cliche formulas to generate shallow imitations should not wear the same title (e.g. composer) as creators who invest in developing and refining their skills and commit to their craft for life. It’s time to acknowledge that creating music is not for everyone.
I was fully against a.i.-enabled music tech and saw no benefit to anyone back then. I understood the point of both its haters and supporters – but had three issues with it:
1. The suppression of originality / devaluation of the arts: generative music a.i. is built on the porridge made from the most promoted (therefore most over-played) music of the past, hence it serves up the cliches we had created over the past decades – or at best, repackages them as reusable building blocks. The inherent qualitative failure of a.i.-generated “music” lies in the way it’s constructed: the more widely such building blocks have been used (i.e. in more songs that became popular), the more likely they will end up in a.i.-generated content. It’s a self-feeding mechanism: it utilizes the most popular solutions, generates the safest, least innovative variations, then feeds the results back into the algorithm to fine-tune itself to offer increasingly more predictable popular variations. Do we really need more of the same aesthetically super-narrow cliches that labels have been showing down our throats for decades? I don’t think so – but the general audience who likes convenience and familiarity, might disagree. (This is also why the argument that a.i. algorithms learn audience preferences is a very backwards one.) In this aspect, the music industry wins, music and culture suffer.
2. Lack of concept, cultural context and personal expression: a.i. might be capable of generating entertaining content, even better quality content than some amateurish human-made music can be. However, a.i. does not “believe” in what it creates (unlike our best musicians). It might serve up logical solutions based on the context it is fed, but it has never experienced any cultural context or even artifacts. Where is the history, suffering, love, devotion, etc. – all that we experience – expressed in generated content? Personal interpretation and expression through art is what makes each artist unique – it’s not about logic and data, it’s about the expression of a concept formed by our individual lived experience. (And such personal expressions are even present in the millimeter-accurate and millisecond-timed decisions one makes when playing an instrument).
3. Emotional quantization: like in all digital systems, values are assigned on a stepped numerical scale in generative music. However small the steps are, there are always gaps between them (fractions of the smallest values in any code). The nuances of human emotions, and their expressions, are not based on values but are fluid and infinitely refined, “analog”. Just think of an already-nuanced feeling like bittersweet; each time you experience it, it is something different. Also, these multi-dimensional qualities are cross-dependent on one other (what type of bitter, what kind of sweet), dynamically influencing the outcome, often without logic. And then we haven’t even added the personal-contextual variables which further limit their assessment and imitation via code.
While I used to see these as “issues”, I see them now as reality that might actually lead to natural selection in today’s impostor-ridden music landscape. The question is: who will be effected and how.
Music and a.i.: Natural Selection
How about, for once, not ranking composers, musicians, songwriters, performers based on their position in the hierarchy of the music industry, but based on their level of artistic development? The ongoing changes are already reshaping the three fundamental creator segments:
• Hobbyists. The output quality of this suddenly expanded segment is getting close to the advanced creators’ output. Given that they see music as a delivery system for entertainment and/or for their self-expression, it is an opportunity for them and for the tech industry to offer them a.i.-supported services (e.g. writing assistance) – and with that, the illusion of possessing writing skills.
• Advanced creators. They are at the highest risk as the value of their advanced skills is diminishing. The a.i. music revolution is their new competition, so in order to keep their position, they are forced to adapt and utilize the new tools. Unfortunately, this comes at the price of (knowingly or unknowingly) lowering their standards as they their segment will devolve artistically. Otherwise their options are limited as they must cater to a pragmatic clientele or to an unassuming general audience. However, the most committed creators in this segment also have the opportunity to move “upwards” by developing into a top-tier artist (see diagram below).
• Top-tier artists (defined here in terms of level of artistry, not revenue). For this narrow segment, a.i.’s effect is minimal; if anything, it’s pushing them to raise artistic standards. It is their suddenly accelerated opportunity to deepen their artistry NOW and evolve their individual style without a.i. (rather than re-invent themselves using a.i. – contrary to the generic advice of tech professionals). They only risk sliding down into the new abyss of well-packaged generic content if they get complacent with their own personal expression that has served them this far.
Maybe all this isn’t a bad thing for music overall. On the upside, the best will have to further improve, the mediocre will use a.i. to compete in a faster-paced, less-musical new segment. The respect of real music might return to where it was in the mid-20th Century, driven by an technological over-compensation for democratization. I think of it as a sort of music market correction. On the downside, the new music industry will likely flood the mainstream with an unlimited supply of a.i. mediocrity, regurgitated cliches and shallow imitations.
Natural selection and music market correction in an a.i.-infested music industry shift.
For conceptual originality and the expression of individuality, a human is required. The fate of the music industry is not in the creators’ and not even in the companies’ hands: it’s in the listeners’ hands. Whether that’s a good or bad thing, time will tell. Different listener segments expect creations at different artistic standards from different creators – therefore, every stakeholder has a different reason to adapt to the changing landscape.
Music is heading for an a.i. “Market Correction”
Since the 1980s, technology has made it “too easy” for many untrained writers working only on the surface of the quality and skill level that was previously expected in the musical arts. It has become the new norm to create seemingly original musical content that seemingly expressed deep layers. In reality, modern musical arts have become complacent with shallow imitations of the actual artistic depth and nuance found in sincerely designed and expressed works of top-tier creators. For them, music was never just a delivery mechanism but the core of their personality. Those who desperately relied on the assistance of technology due to their lack of skills, talent, artistic concept or commitment, quickly evolved into Advanced Creators (see the segment above). Conversely, the same technology only enhanced real artists’ concepts and sincere expressions, representing the best of the modern music culture.
However, over time, the lines got blurred between these segments; as advanced creators claimed their place in the top-tier artist segment, the general audience was getting heavily conditioned by packaging (production) and popularity (industry). The result: the majority of today’s listeners think that the more products an artist sells, the better they are at their craft (i.e. top-tier creators are likely with major labels). Yet, reality shows us that there is close to no correlation between a musician’s artistic quality and their position in the hierarchy of the music industry.
Now with a.i. bringing generic solutions and leveling the playing field for all but the truly original outstanding individual creator, top tier artists must work harder to separate themselves from the new, unified segment of mediocrity. While the advanced creators understandably fear a.i.-tools’ impact on their position, risking to be consumed by the new segment of mediocrity, they are faced with the true value of their art, their own humility and commitment to improve it.
How? Sincere and committed creators can grow by seriously improving the originality of their concepts, the individuality of their expressions, and the contexts within which they create – the areas a.i. tools are at an inherent disadvantage.
However, another important step for the new top-tier of artists to prevail is the elevation of the audience’s standards. This is not just an opportunity for them but for the the musical arts and culture. This can only take place through education – which is possibly the biggest challenge, given our recent trends of sinking appreciation of the arts and pragmatic priorities. The a new audience, the task is not unlike learning what to appreciate in a Monet painting beyond the obvious. Characteristic brush strokes are an easily observable, learnable and even imitable technique – relatively easy tasks for the general public or for an advanced a.i. engine. But, internalizing the painter’s personal intent with making the painting and every bit of passionate detail and emotional context behind each brush stroke on it takes far more attention, context, and a deep sensitivity – certainly more than code ever will be capable of. It requires a first-hand experience of the human condition.
Immersive audio is an important concept in our listening experience as we enjoy films, games and other audio content. In musical works, while the concept of multichannel formats is not new, spatial audio technologies have been far underutilized in the inception of content; technologies such as ambisonics, wave field synthesis, object-based audio are, with an overwhelmingly wide margin, used in the recording, mixing and mastering stages of music production today. These spatial audio technologies are usually applied during music production; after the music creation but before the exhibition (performance or playback) phases, with spatial decisions typically made during the mixing process, often in absence of a spatial-musical concept. Here we propose the consideration of space as a compositional, rather than production element of music. We examine how its use in the earliest phase of conceptual design has a substantial effect on the composition itself.
Focusing on the creative possibilities offered by spatial audio design, what if decisions such as an instrument’s location or its movement in space were not made during audio production by engineers, but rather, they were design choices made by composers prior to production, during the composition phase? Composers utilize melody, harmony, rhythm (and often timbre) to express a musical idea and evoke emotions – what if such ideas could be also shaped or even inspired by spatial location, area, position or movement? Would such space-derived decisions about notes, timbres and the roles of musical parts be the same if they were made at the time of composition, rather than during audio production? And, even more interestingly, would making spatial decisions during the composition process have a bearing on the musical message itself?
This paper explores, through an elaborately conceptualized and documented composition test process, via the analysis of the resulting stereo and spatial compositions, and through the evaluation of focus group feedback from comparative listening sessions, how various spatial environments can shape compositional design. The paper discusses the consequences of using immersive technologies in the earliest phases of music production and recommends best practices to impactfully express spatial-musical concepts. It concludes with the analysis of the advantages and drawbacks of using space as a compositional element and proposes approaches to consider in the future practice of spatial music composition.
[Read the paper preprint here] Published: November, 2024
Last revised: September, 2025
[TL;DR – additional notes]
Here are my main conclusions from the research:
Space is a potent and influential element to use in composition, indeed, now practically joining melody, harmony, rhythm and timbre
Focus group participants preferred music composed in a spatial environment vs stereo environment
Listeners found the spatial composition to be more exciting, while the comparable traditional (stereo) composition more beautiful
The more spatial elements are used in composition, the more likely melody and harmony will receive less focus
Spatial composition encourages more rhythmic and melodic movement
Composing in/for traditional (stereo) space promotes less melodic movement but more reliance, therefore focus, on melodic expression
Spatial composition is more likely to inspire fragmented melodies (shorter melodic arcs, shorter phrasing)
Spatial composition promotes the use of complex polyrhythms
Timbre is the intrinsic musical component least effected by the introduction of the spatial element
Space is still not an intrinsic quality of a note, but it is indeed an intrinsic quality of the relationship of ≥2 notes
Spatial audio productions should prioritize conceptual choices made by music composers over mixing choices; engineers should enhance musical spatial concepts during production
While the broad acceptance of space as a conceptual element of composition grows slowly, we can already see its non-conceptual (and in some cases even harmful) use in production
Spatial composition offers opportunities for future real-time spatial performances
I get a fair amount of questions from the students in my UEMP (Ultimate Electronic Music Production) course each year. Often these are questions about specific genres or production approaches or synths. This Fall, though, I got some bigger-picture ones, so why not share a couple of the answers here.
Q: “Would it be possible to create another branch on the tree [of origins of electronic music] in our modern world of electronic music?”
A: Technically and in theory everything is possible. Practically, quite difficult; it’s difficult to not be fundamentally influenced by what we already know. I believe a completely new, thick branch (directly growing out of the double-trunk of the Electronic Music Tree rooted in France and Germany) will / can only be grown by a non-musician artist, from another discipline, like painting, sculpture, dance or poetry. One could argue that Schaeffer, Stockhausen, etc. made significant progress thank to their not-purely-musical interests. The less someone is exposed to music, the more likely they won’t be influenced by the genres already out there (any existing “branches”), especially the 1950s-70s roots, and can think artistically but freely. The only time the origins of an entirely new direction in electronic music was not tied to the 70s andwas a musical in its source (Jamaica), was Jungle (then Drum & Bass) in the early ’90s. The unrelated components being in the right time and place and actually cross-influencing one another – such coincidental cultural collaborations are uncommon in today’s too-multinational world. Though, maybe we could still give a Model D and a zoom recorder to the indigenous tribes of the Amazon who live far away from modern civilizations, collaborate with them and see what comes out of it…
At the same time, a new smaller branch (a branch growing out of a large branch of the tree) might come from a musically-oriented (trained or not) person in a different musical environment (possibly not even electronic music). Again, the less they know about electronic music, the less likely they will be effected by what has existed before them, thus, the more likely they will create something uninfluenced by existing electronic music. Still, it’s likely that we will be able to trance the results back to one of the main trunks of the tree (i.e. GRM, BBC, JMJ, Kraftwek, Vangelis, Tomita, Eno, etc.).
Like I often say in the beginning of UEMP semesters: the technology to create something groundbreaking and originalintentionally break has been in front of us, for many years. Powerful computers and more free plugins that one could explore in a lifetime are now in our pockets. Affordable technology to write the music of the future is reality, even for the average college student. But you have to fight the part of your DNA that always makes you create music similarly to the way you normally create and listen. Producing Musique concrète has the ability to take you very far from all that… listen to your Musique concrète work a half year later and you might be surprised by what you had created. An inspiring point to start from.
Q: “If Kraftwerk hadn’t popped when they did, how do you think things would have turned out for electronic music?”
– and – Q: If Kraftwerk had never existed, where do you think electronic music would be today? Would it be at a previous point on the electronic music timeline or would it have gone in a completely different direction?
A: Well, I’m not that old to have been there… I was a small kid playing the piano in the 80s when Kraftwerk had been already around for a decade and well-known in Europe… so their influence on most everything-in-4/4 in electronic music was a done deal by then. But I’ll take a stab at hypothesizing about this fascinating proposition.
It is truly interesting to think about what would have happened to electronic music if Krafwerk’s Autobahn wouldn’t have received much attention, or hadn’t even been released… or if Ralf und Florian wouldn’t have met or even existed. Well, let’s break this down based on the components that Kraftwerk combined into a pioneering new direction in electronic music:
The 4/4 time signature would have become popular in electronic music in some fashion anyway, since it was the middle ages’ 2/4 folk dance rhythms then the ’60s disco’s straightforward patterns that evolved into 4/4 dance-able rhythms – the point being dance-able – in other genres as well.
Sequenced, quantized basslines and arps would have been sooner or later migrated from Jarre’s and Schultze’s kitchen to other, more rhythm-based genres. Kraftwerk was inspired by the persistent, “quantized” action of mechanical machines, but certainly not the only one who got inspiration from heavy industries.
Vocoders were in use even before Kraftwerk, although without the band and its influence on others (e.g. YMO, Moroder), vocoders might have actually disappeared earlier.
Minimalism in itself was around in painting, sculpture, architecture, film-making, etc. before the 1970s (let alone the concept of wabi-sabi from centuries before). Terry Riley was already active before Kraftwerk and other musical creators of the ’70s were also making very different type of music on minimalist principles (e.g. Brian Eno’s ambient).
The idea of “robots” making music… now this was a big one. It was big because this concept was able to connect the minimal aesthetic with most people, who were curious about the future. And when I say people, I mean nearly everyone – just think about all the major sci-fi movie franchises that originated in the 70s. Shortly after the Moon landing and with portable computers starting to appear, it was the perfect timing for everyone to entertain the idea “The Future is Now”. Thank to people’s curiosity about robots and robot-like musicians, unlike “untouchable” high-artists and abstract minimalists, Kraftwerk was able to bring minimalism into the mainstream. This is the aspect without which I’m not sure if the band’s music would have had the same enormous influence on popular electronic music. And without that, the genres that later evolved (and still evolve) from Kraftwerk’s aesthetic (first and foremost techno and its derivatives) might exist only as minor, underground sub-genres these days.
Yellow Magic Orchestra was, self-admittedly, the Japanese version of the German robots. What if these gentle “robots” infused in jazz chords would have broken into the mainstream before Kraftwerk, and Dusseldorf would have only remained a national treasure?
I believe that without Kraftwerk, the precision and angular German-ness might be present only in a smaller proportion in electronic music. Four-on-the-floor wouldn’t automatically mean that it’s quantized kicks on downbeats, sequenced basslines wouldn’t be necessarily repeated as 1/8th or 1/16th notes or as octave-alternating straightforward sequences, bass lines wouldn’t predominantly have short decay and very short-release VCA envelopes – as they often do today. In other words, a higher proportion of electronic music would be less “tight”, and maybe more… …French. Maybe softer, likely more textural, probably groovy and rhythmically varied yet easily dance-able genres would be the default “EDM”. That is, at least until 1990 and DnB in the U.K. and later in 2010 Skrillex bringing rhythm-based electronic music into the U.S. mainstream. (But wait, would DnB have been the same without the counter-culture role it had in Eurodance times? And, would Skrillex have made the same debut album in 2010 without the influence of Kraftwerk-and-techno on baby Sonny during the 1990s? This is the paradox of synthception… noone knows… but likely, he’d be shredding metal today.)
In any case, while Germany’s influence was successfully pushed back in other areas during the 20th Century, they might have unknowingly won the long-term cultural influence on electronic music. (And Dusseldorf’s direction was relatable by many, far more than Berlin’s high art.) Their sharp character has influenced other areas as well; just think about cars. The passionate and adventurous Italians’ slogans are “Only those who dare, truly live – We are the Competition.” (Ferrari), and “Expect the Unexpected” (Lamborghini). Meanwhile, BMW’s slogan since WW2 has been “The Ultimate Driving Machine” (despite of this being a enormous lie). The automotive world could have been as graceful without the past 30 years’ German cars as the world of modern Japanese robots is without German robots in it.
I truly love the music of Kraftwerk and I’m glad that it got popular. Though looking at their influence today, maybe too popular?
An even more interesting conundrum is the role of counter-culture. Let’s say, if without Kraftwerk, electronic music would have evolved to be less quantized and less “raw” during the past 5 decades, would various types of non-electronic music have tried to counteract it less? With Disco Demolition Night never taking place, Afrika Bambaataa not having anything to steal, the Pet Shop Boys playing bass on bass guitar, and DJs having to know at least a little bit about harmonies, electronic music skeptics haters would have had to look for different targets… or just chill and make even more rock and pop. If a softer, more romantic form of electronic music had been the norm throughout the 1980s and ’90s, would have a different set of genres become more minimal, tight, quantized, aggressive by now? Or would we have reached the current aesthetic of block-y, rigid on/off notes in electronic music anyway, given the black-or-white binarity that’s already woven into so many areas of our overly-practical life today? This is as big of a philosophical question as you can expect to face when you hypothetically re-write the history of electronic music… so I’d rather look for the answers from Nietzsche Descartes.
