Each is based on a collection of instruments playing five patterns of three seconds each. These patterns (aside from untuned percussion instruments) have been transposed into twelve musical keys and play randomly.
It can be found at extendedplay.filipnest.com
I’ve wanted to do this project for a long time, first making attempts some time around 2010/11 with a one minute ukulele track cut into two or three second pieces (the original idea was one second but MP3 players can’t shuffle in time). I for some silly reason left this proof of concept as it was for years rather than trying to do it properly. The concept was that you could listen to the same piece of music for much longer than its running time if it were shuffled randomly and very short. Parts would combine into new parts.
The original idea was for a series of EPs, each about a minute long and focusing on a specific instrument or style. These would then be uploaded to something like Spotify and people could make playlists from them, shuffle them, share them and make an infinite amount of music from something very short and finite.
Five years later, I tried this out and realised two things:
1) Cutting up a one minute track doesn’t produce the best results. The best results were from actually composing or playing separate three second pieces.
2) The single one minute piece, however complex, could still be detected when shuffled. Our brains are pretty good at detecting repetition it seems, even if it’s in a random order.
So I decided to modify the concept in two ways:
1) I would transpose the melodic parts into 12 keys using MIDI. Sadly I originally wanted to do it all with acoustic instruments rather than samples but it turns out pitch-shifting 12 semitones distorts the sound of things too much.
2) I would allow listeners to play more than one track at the same time.
Those who know me will know I’m terrible at arithmetic so, on my first attempt at recording, I miscounted and recorded a minute and a half of music. At this point I decided that, as I was doing the transposing thing anyway, the minute limit would not stand.
I’d love to try it in the future with a huge number of instruments, perhaps a whole orchestra. And also syllables recorded through a vocoder to see if I can make some random words form through separated/joined vowels and consonants.
I suppose the lesson is that concepts aren’t always as good in practice as in theory but I hope my tweaks make it work a bit and enjoyable.
I could re-record the musical parts forever (already took ages) but instead I’ll just keep giving it a go and release more EPs of it in the future. Hopefully they’ll get better as they go on.