20m experiment
By Nueli on July 18, 2026 7:48 pm
"Can I do the whole WeeklyBeats in a day?" my brain asked? This corresponds to about an hour of music (52 tracks of at least a minute), so each minute of composition per track results in an hour of work. So no, the answer is no, I cannot reasonably stay that concentrated for that
But that gives me an interesting challenge: can I make a track in under 20m? Let's try and document the process.
I need a plan before diving in, a rough outline of what I want. My usual "try stuff, see what sticks, multiple projects in parallel until one hatches" will not cut it. So here is the plan:
- everything in the m8 (as per my year constraint)
- only 1 instrument, two tracks (melody and bass), max 5m to find good synth parameters
- make a loop, copy-past until it reaches 1m, tweak tweak tweak
And here is of it actually went:
- setup a track with a C4 and a C6 alternating
- wavsynth (morphing), fiddle with parameters, ahd envelope, reverb+delay to fill up the space.
- random generation of a rhythm. Random notes up then down to make a melody out of it. Adjusted the rhythm.
- steady bass. Play with transpose (TSP) rather than creating new phrases.
- make an intro with just the melody for 4 phrases
- copy-paste the loop as intended.
- make 2 variants of the melody and alternate between them every 4 phrases.
- make a second section with no melody and faster bass
- add more high notes in second section each phrase to build intensity
- Alarm rings!
Sounds alright. I think the B part should take its time more.
Maybe next week I clean this up this and dedicate it some well deserved time?
Audio works licensed by author under:
CC0 Creative Commons Zero (Public Domain)
