When I was a kid, my mom made me play piano. My piano teacher wasn’t very strict either. At my elementary-school peak I could show off by playing some Richard Clayderman pieces, and I passed the grade 4 piano exam. But once my mom was no longer around the house, I stopped playing.
I love music. I just don’t love being forced to play pieces I don’t like.
By college I’d basically forgotten how to play. Going purely on my own interest, I bought a Yamaha digital piano for my dorm and played a bit of the Joe Hisaishi I love, on and off, like Hisaishi’s nocturne (because Chopin’s nocturnes are too hard).
I never properly studied music theory either, but I’ve always stayed interested in making music. My head is always full of it.
A few years ago, back when I still had a day job, I was following Andrew Huang. He happened to launch a course with a big advertising push, and I got hooked all over again. Another burst of three-minute enthusiasm.
I whipped out my credit card on the spot and signed up for Andrew Huang’s online course, determined to speed-run learning music production in Ableton and finally become the musician I’d always admired from afar.
The course was structured around making three songs in a month. But just following along felt boring. I’ve always loved going off-script. I still remember a team offsite where we learned to paint San Francisco’s Golden Gate Bridge, and I insisted on painting a monster underneath swallowing the whole bridge. I just don’t want to make the same thing as everyone else.
Learning music was no different. I didn’t want to just take the course. I wanted to learn while drawing on the music I loved, and make a song that was mine.
For those few days, I was running on pure enthusiasm. Every day after work I’d come home and study, going until three or four in the morning. My head was so full of ideas I couldn’t sleep even after I finished, and the next morning I’d still drag myself up and go to work, thinking the whole time about coming home to make music.
The course encouraged us to share our progress on the community page each time, so I still have a lot of the original files from back then.
A 10-day music production log
Monday, March 30, 2020
First attempt. No idea what I was doing. Just followed the course to get familiar with Ableton and made some kind of oscillator (?) effect. A level 1 player fighting level 1 monsters.
Tuesday, March 31, 2020
Fiddling around, for some reason I stumbled onto a melody that sounds like A Man of Determination. But paired with a slightly Hiroyuki Sawano-style bass, I thought it actually came out pretty well.
Changed the length of each note to see if I could tone down the “Man of Determination” vibe.
Thursday, April 2, 2020
Added an intro, some layers to bring in the drums, and a second overlaid melody. But it sounded messy.
Friday, April 3, 2020
Adjusted the layers a bit.
- Inspired by Spangle call Lilli line - Sea, I added piano at varying intervals
- Inspired by Covet - Falkor, I added a fast drum fill before the final chorus
Through constant trial and error, slowly turning a small idea into a complete song. It was genuinely a joyful experience.
Wednesday, April 8, 2020
I was probably heads-down working in between, but didn’t share anything, so the log jumps straight to here. I added a lot of parts, mainly the second and third choruses, the final D melody, and the outro, stretching the whole song out to 3:53.
Thursday, April 9, 2020 (final version)
With my tin ears, I did my best to master this song: adjusting the volume, repositioning each track, adding lots of little tweaks and ear candy, and even trying to record a bit of vocals. It stretched out to over five minutes. A tribute to Falkor’s seven.
That’s the 10 days of music-making I still have on record. I’m turning it into a blog post before I forget it.
Making this song was full of surprises. At the start I had no idea what kind of song I wanted, so I just went with whatever felt right, picking out every track’s melody one key at a time on the keyboard. Adding bits here, trimming bits there, until it became what it is now.
I love listening to it, even if it’s pretty niche. I made it taking after the math rock I love, and at the time my thinking was the more complex the better, so there are a lot of different tracks running at once. It’s chaotic. And with my tin ears, I didn’t know how to mix the levels well. The musician friends I played it for didn’t think much of it either: the drums are too loud and monotonous, and your ears start to hurt after a while.
But over the years, I still pull it out to enjoy from time to time. After all, it’s tuned entirely to my own taste, and listening to it brings back the flow state of those 10 days.
It’s not necessarily good, but it’s a work only the person I was back then could have made.
After finishing this one, my enthusiasm had actually burned itself out. I was completely sleep-deprived and crashing on dopamine. It was like sprinting all the way to the summit, too spent to climb down, just wanting a cable car or a helicopter to carry me back.
By the second song, Andrew wanted us to go record our own everyday samples to make music. I found that too much hassle, so I paused. The third song required adding my own singing voice, and writing lyrics was so hard that even after I’d caught up on sleep, I had no energy left to do it.
After that I still tried to make something here and there, but it was all small stuff. I’d usually work for an hour or two and then give up.
Some time passed, and when I reopened Ableton, I’d stare at the complicated UI with no idea where to start. I’d completely forgotten how to use it, and I didn’t have the enthusiasm to take the online course over again.
I thought this was the first song I’d ever make, and also the last. That I was done with music production for good.
When AI music arrived, it didn’t immediately spark my interest either.
When I thought about why, I realized I much prefer tweaking each note in a DAW. In that process, every decision is mine: heavier kick drum here, softer synth there. The final product might not sound as “professional” as what Suno spits out, but every note has my fingerprint on it. — Vibe coding is a slot machine
Until… Suno v5 came along and proved me wrong, reigniting my enthusiasm just a little. I used it to make a lofi jazz album for working to, and a bossa nova virtual singer.
And with that, I started “playing” with music again.
Next up: One song, seven ways: covering my own music with AI.
