Music and technology entered my life separately.
Music came first. Some of my earliest experiences were at Christ Church Cathedral Choir in Oxford: rehearsals, services and performances, but also recording sessions and television appearances. I learned music inside a tradition that stretched back centuries, while microphones and cameras carried it into another world.
The technology was present, but other people controlled it.
I also began composing and coding while I was young. At first, they belonged to different parts of my life. One meant notes and instruments. The other meant instructions, systems and making a machine do what I intended.
Then a Yamaha V50 arrived in our house.
When the two strands met
My father was teaching at the Royal College of Music. Around 1990, the college loaned him the synthesizer as part of its attempt to keep pace with a rapidly changing digital world.
I explored what it could do.
The V50 combined a keyboard, digital synthesizer, drum machine and eight-track sequencer. I used it to transcribe parts of Brahms’s Clarinet Quintet and the theme from Jeeves and Wooster. Every note had to be entered, corrected and balanced inside the machine.
Later, I acquired a MIDI interface and connected the V50 to a PC. The keyboard became the input device; Cakewalk took over the sequencing. The same instrument now sat inside a larger system.
That was the point at which composing and computing began to occupy the same space.
At first, the computer did not make any sound. Cakewalk stored MIDI instructions: which note I had played, when I had played it and how hard. The sound still came from the V50 or an external module.
An E-mu Proteus 2/XR gave me a larger orchestral palette. Its sounds were dry, so I passed them through an ART effects processor to add reverb. Composing meant coordinating a collection of machines, each responsible for one part of the result.
More equipment meant more possibilities.
It also meant more cables.
When recording still meant tape
When I studied film and television composition at the London College of Music in 1997, the studios used TASCAM DA-88 digital tape machines.
The recording was digital. The working method remained linear.
Tape moved forwards or backwards. Finding a moment meant locating it in time. Replacing part of a recording meant working within the physical sequence already on the tape.
Writing to picture added another tape. A VHS copy of the film carried timecode on one audio channel. A separate interface read that signal and kept the sequencer running alongside the picture.
Everything had to agree about where it was.
Hard-disk recording changed that relationship. Audio no longer had to be approached as a strip of tape passing a fixed point. A passage could be opened directly, moved, copied, shortened or replaced without winding through everything before it.
Video followed the same path. Once the picture lived as a file on the computer, I could move to any frame and hear the corresponding point in the music.
The machine stopped behaving like a tape recorder and began behaving like an editable map of the piece.
That changed more than speed. It changed how freely I could revise.
Bringing instruments into the computer
The next problem was sound.
Hardware modules could imitate orchestral instruments, but their memory was limited. Hardware samplers improved the situation by allowing recorded instruments to be loaded and played from a keyboard. My setup included an E-mu ESI-32, a SCSI CD drive and an Iomega Zip drive.
The process now sounds archaeological. At the time, it opened another door.
Recorded performances could be mapped across a keyboard. A single instrument might contain different attacks, dynamics and articulations. The keyboard no longer triggered a synthesised approximation of a violin or clarinet. It triggered recordings of the instruments themselves.
Software samplers moved that work onto the computer. GigaStudio used hard-drive streaming to handle libraries that would not fit inside the memory of a hardware sampler. The available orchestral palette grew rapidly.
So did the demands.
Larger libraries required faster drives, more memory and more computers. New sounds promised greater realism, but every improvement came with another installation, another configuration and another expense.
By 2010, I was running a modest orchestral template across three computers.
The machinery had not disappeared. It had multiplied.
Returning to a live orchestra
In 2007, I recorded with the London Symphony Orchestra at Abbey Road for the first time, for the Philips Aurea campaign, Seduction by Light.
That might look like a return from technology to traditional music-making. It was not. The score had already passed through years of changing systems: MIDI sequencing, orchestral samples, synchronised picture and computer notation.
The technology allowed me to build, revise and present the music before the recording session. It helped everyone arrive with a shared understanding of the structure, timing and orchestration.
Then the musicians changed it.
A sample can show whether a line works. It cannot reproduce the collective judgement of an orchestra: the way players balance while listening to one another, alter the shape of a phrase or respond to a conductor in the moment.
The machines had brought the music to the room.
They were not the reason the room mattered.
Less machinery in the way
Over time, the separate parts of the studio began to collapse back into the computer.
The sequencer absorbed audio recording. Software replaced racks of sound modules and samplers. Video became another file in the project. Faster processors and drives took over work that had previously been distributed across several machines.
The room became quieter.
The promise of music technology was never simply more equipment.
A new machine offered more tracks, more memory or a more convincing sound. It also had to be bought, connected, configured and learned. At its worst, technology gave me another obstacle to master before I could return to the music.
Its real value was practical: removing inconvenient steps between what I had and what I needed.
MIDI let one performance control different instruments. Hard-disk recording removed the need to wind through tape. Non-linear video let me move directly to the frame I was scoring. Software samplers placed sounds inside the composing system instead of across a rack of external machines.
Each development shortened the distance between an idea and something another person could hear.
The best technology did not make the musical decision for me.
It left less machinery in the way.