FOR most of the thousands of years that humans have been listening to music, the only way to hear it was to be there when it was played. The earliest musical notation would take millennia to emerge, and the first recordings were made only about 150 years ago. You might think that since then we have got steadily better at capturing the grace and richness of music. But for almost 40 years, we have relied on the same technology to produce the vast majority of the music we hear. And it’s badly in need of an upgrade.
Finally, though, a sweeping overhaul is in progress. The ramifications for music-making will be huge. When digital music entered into its own in the 1980s, it quickly began to shape what people listened to, ushering in waves of creativity and whole new genres. Now that its foundations are being reset, the same is sure to happen again. And it isn’t just music; the way we experience sound – on television, in cinemas and beyond – is set for an upgrade. So steady those ears: they are about to experience sound as never before.
The first sound recording we know of was made by a device called a phonautograph in 1860, and features a rendition of the folk song Au Claire de la Lune. The machine, a brainchild of French inventor Édouard-Léon Scott de Martinville, transcribed sound waves into a line traced on smoke-blackened glass or paper. It may seem primitive, but conceptually, things stayed the same for more than a century. If you listen to Nancy Sinatra singing These Boots Are Made for Walking, for instance, you are hearing sound waves recorded on a specific occasion in 1966 when she sang that song.
Advertisement
A revolution arrived not long after, although you could see it as a return to a former era. Long before the phonautograph we had developed another way to capture music. Instead of recording sound waves, we wrote notes on a musical stave. This notation encodes the music, distilling the essential elements of a tune into instructions that can be used to recreate it.
“You could record a tune on a keyboard in the morning, and then tweak the code later to make it sound like a guitar”
In 1983, a new method emerged of encoding music digitally. To see how it works, imagine tapping a key on an electric keyboard. Unlike a piano, this doesn’t hit a string and produce sound waves. Instead, information about the key pressed and the intensity and duration of your touch is converted into a digital code, which tells speakers or audio software what sounds to produce. Whichever keyboard you happen to use, the code will be the same: the Musical Instrument Digital Interface, more commonly known as MIDI.
There were similar ways of digitising music before the 1980s, but MIDI was transformative. It became the universal standard, guaranteeing that instruments of any kind could be connected to synthesisers and computers. It also made digital music easily editable: you could record a tune on your keyboard in the morning and then tweak the code later to make it sound like it was being played on a guitar – or almost any other instrument you fancied.
The birth of the home studio soon followed. “It became possible for a single person to compose, record and produce music,” says producer , who has worked with artists including Placebo and Boy George. By making it cheaper and easier to create electronic music, MIDI fuelled an underground explosion of dance music that led to mainstream success. And as the use of electronic instruments proliferated, MIDI’s benefits would establish these sounds as the backbone of pop for years to come.
Cue the harmonica
Most remarkably of all, MIDI is still at the core of music production, though less so in the case of genres commonly recorded live. It has had a hand in the making of almost every pop song from the past 30 years, typically as part of digital audio workstations used in production. “MIDI is a very flexible tool that offers a vast array of possibilities to its user,” says Natalia Rodrigues Milanesi, a freelance music producer based in London. Kane agrees: “It’s a tribute to the original designers and engineers who worked on the version 1 spec that its first revision is 37 years later.”
That’s right, cue the harmonica: for the times they are a-changin’. In January 2020, the MIDI Manufacturers Association agreed on plans for MIDI 2.0. For, as the years passed, cracks in the code had begun to show. “When MIDI was created, it was for a very different world of electronic music,” says Will MacNamara at instrument manufacturer ROLI. “It’s like a 1983 car that people are still buying in the billions now.”
One of MIDI’s biggest limitations concerns how it quantises data. In order to be captured digitally, each of the infinite gradations of a note or sound, including pitch, volume and timing, must be captured on a finite digital grid. MIDI 1.0 uses strings of seven digital bits to capture these parameters, meaning there are only 128 possible values on the grid. That is more nuanced than paper notation but still lacking. “There’s so much musical expression between those lines,” says Milton Mermikides, a composer based at the University of Surrey, UK. MIDI 2.0 will be capable of capturing far more nuance, because it will be 32-bit as standard, allowing variables to take more than 4 billion possible values.
This is good news for all sorts of reasons. For one, it means music will sound richer and less clunky. Virtual acoustic instruments will be perceived as being more realistic because gradual shifts in pitch known as pitch bends will be captured properly. The same goes for the bends that appear in emotional vocals. “There’s huge expression there that’s very hard to map,” says Mermikides. “MIDI 2.0 could do that really, really well.”
MIDI 2.0 will also suit music that strikes Western ears as unfamiliar. The existing MIDI standard is based around the 12-tone equal temperament (12-TET), the division of an octave into 12 equal semitones as found on a piano. This has been the default tuning in most Western music since the 18th century, but it isn’t the only way to organise things. One alternative, called pure intonation, involves tuning instruments based on ratios of note frequencies. This is common in classical Indian music, for instance, but MIDI isn’t set up to capture it. Doing so requires you to either use expensive plug-ins or hardware.
“The Midi 2.0 Revolution promises to allow anyone to make music with a wider Palette of sounds”
You don’t have to try out anything too wild before running into difficulties. Take the band Radiohead. The intro of their song How to Disappear Completely, released in 2000, features wavering strings that blend together notes only differing in pitch by tiny amounts. Anyone hoping to recreate something similar in a home studio would struggle, because these microtonal variations can’t be captured by normal MIDI. “Microtonality’s impact has been subdued in the age of digital music tech,” says Sean Archibald, who creates microtonal music under the name Sevish. “There is a sense that the technology needs to evolve so that musicians can easily employ alternative tunings.”
The MIDI 2.0 revolution promises to allow anyone to make music with a wider palette of sounds, even with a limited set-up. “All of these barriers have been falling year by year, and it’s increasingly easy for people to discover their musicality with just a laptop and a free programme” says MacNamara. “MIDI 2.0 is going to make that even easier.”
Mermikides thinks this will result in artists and creators pushing the boundaries, producing an explosion of new genres akin to the one that followed the introduction of MIDI in the 1980s. “Just like we have dance music and techno because of technological shifts, which are massively embraced, we’re going to get new music forms that are embraced as well,” he says.
The most exciting effects of MIDI 2.0 might not even be felt in music, but in the soundscapes you hear in a film or virtual reality device. The conventional way of creating realistic soundscapes has been to position several speakers around a listener and play different channels of sound through each, as in cinema surround sound. These days, immersive sound can be achieved more cheaply and easily using soundbars. These speaker devices sit in one place and bounce sound off the ceiling and walls of a room so it feels as though the noise surrounds you. The most advanced units fire white noise around a room to calibrate themselves and can place “sound objects” at specific positions. Imagine a buzzing bee hovering over your shoulder and then flying off towards the window.
“The most exciting effects might not be in music, but in soundscapes in film or virtual reality”
It is in situations like this that MIDI 2.0 could really shine. All those extra data points with which to encode sound will mean the bee’s buzz is pin-sharp. “As a creator, you want to master and finish a sound in such a way that the consumer can listen to it in the closest possible way to how you heard it,” says Scott Marshall at Bamsound, a sound design studio in London.
This isn’t limited to home cinema. Immersive audio can also be produced by making a recording with more than one microphone, creating what’s known as binaural sound. This can be enjoyed with earphones; no need for fancy equipment. Many think pairing this technology with MIDI 2.0 will be important in everything from virtual-reality gaming to live events. “Immersive audio is going to become a much more invisible technology,” says Marshall.
How quickly we begin to see the effects of MIDI 2.0 depends on how fast it is adopted. There is already one keyboard available to buy, with other prototypes in development. “I think by the end of this year, we might start to see some consumer-level products with interesting new functionalities,” says Kane.
Forecasting the effects of MIDI 2.0 – and predicting how noticeable they will be – isn’t easy. But a good way to imagine the change is to think of a clay animation monster in a decades-old film; at the time it would have looked great, but to our eyes it can seem awfully dated. Perhaps we will soon look back at the past 40 years of music history and wonder how we ever thought this was all there could be.
