Like I said in the previous entry, this 30-day course involved doing 3 different projects. Each project taught a few ideas and required students to create a track that incorporated those ideas. So now it’s time for me to show my work.
Actually, it’s time for me to go on a series of meandering digressions about creativity and music, but I’ll post my work somewhere in the middle of it all.
I used to share my music tracks on a regular basis. This was fun at the start of my musical adventure. I’d post something, and get a pat on the head from the public because I was doing okay for a beginner. But as time went on, two things happened:
- As I spent more time with it, I could no longer justify amateurish work with the excuse that I was new.
- I raised my expectations and thus became increasingly critical of my work.
I’ve had several fits of frustration over the last couple of years where I very nearly purged my Soundcloud account to delete all of my early tracks. My inability to improve has created this perverse hatred for my early stuff, as if the only way to get better is to bring up the overall average quality by destroying the worst bits. Yes, I realize this impulse is horribly vain.
(Strangely enough, Soundcloud seems to have solved this problem for me. Their shifting pricing plans led me to drop my subscription, which means my account reverted to a free account, which means I’m over budget in terms of upload time, which means that a lot of my early tracks are disabled.)
I have this terrible self-indulgent cycle I go through where I get frustrated with my work because I don’t sound like Armin Van Buuren or Deadmau5. Then I realize it’s completely unreasonable to expect to be able to create on the level of superstar artists, and even more unreasonable – bordering on delusional – to expect to be able to work on that level when you’ve only been composing for five years. And only part time. And you don’t even play an instrument. And your musical diet is fairly narrow. So then I give up for a few months. Then I hear something that tickles my brain and makes me want to try again. I’ve got sky-high expectations and rock-bottom skills, which isn’t really a good recipe for a healthy and rewarding hobby.
Track 1: What’s So Funny?
This project was a sort of catch-all introduction to the course. It started with the basic theory surrounding chords and chord progressions and then dove into sampling.
The project had two stages: The first was to just come up with a chord progression and pick out some instruments to make a basic loop. The second stage was to round up random audio samples and incorporate them into your song. It seems to be a pretty big jump to go from “here is what a chord is” to “take disparate audio samples and make them fit within the framework of a full track.” But whatever. It seemed to work out.
This course was taught by Andrew Huang. He’s a master at using audio samples and it’s a big part of his YouTube channel. He spends a lot of time making music with non-musical sounds. As part of the lectures for this first project, he produced a track. I wish the track was available in public so I can show you what he was doing, but right now it’s stuck behind the Monthly paywall. He took a ton of completely random samples and packed them together around the more musical parts of the song. He calls these sorts of injected samples “ear candy”, although I don’t know if that’s an accepted musical term. It seems like it would be a recipe for chaos and dissonance, but he knows how to make these random sounds make rhythmic and melodic sense. Pop Culture by Madeon is the most well known(?) example of this technique that I can think of, except Huang’s track used random non-musical samples rather than iconic bursts from popular music. An older and more obscure example might be Lesson 6 – The Lecture by Jurassic 5, which is a really interesting track if you’re not into the whole dance-ability thing.
I got to the second half of the project and I realized I have no idea how to do this. I watched Huang do it, in real time, over the course of two hour-long lectures. You can teach the technique, but you can’t teach the intuition that guides the process. That comes with experience. (Talent doesn’t hurt either.)
While Huang is fascinated by sampling, I’m obsessed with oscillators. My favorite part of producing a track is building my own sounds out of various filters and sine wave generators. My favorite OSC these days is Helm, which looks like this:
I’d already constructed most of the oscillator-based track when we got to the “adding random samples” stage. I really struggled with making this work and I couldn’t figure out why mine sounded so terrible. It was pretty late in the project when I realized that adding samples is easier when you’ve left some room in the mix. Like usual, I had managed to fill up the track with notes and I didn’t have any breathing room where you could drop in extra bits. I spent a lot of time fiddling with it in frustration. I eventually ran out of time and was forced to post what I had. If not for the class, this track would have been abandoned in some sub-folder of my hard drive and never heard again.
In the end, I just had some random laugh sounds sprinkled around without any sense of purpose. It… doesn’t work:
I guess I like the intro, and the bassline might be worth swiping for a future project, but the rest of the track is fairly dull, repetitive, and not mixed particularly well.
It was the easiest assignment in the course, but I struggled with it more than any of the others. Looking back, I think it would have been better to throw most of the track away and start over when we got to the sampling stuff.
Track 2: Veloscillator
I’ve talked before about trust in the storyteller. That’s where a writer will – accidentally or on purpose – introduce an element that doesn’t quite fit with what we’ve already been shown. If we trust the storyteller, then we assume this tension is deliberate and the pieces will fall into place later. If we don’t trust the storyteller then we assume the writer is a lazy screwup who can’t keep track of their plot elements.
Over the years I’ve realized that this same idea applies to a lot of art. If the audience trusts you, then you can use that trust to create additional tension and make the work even better.
Imagine a movie that opens up with the camera peeking under a garage door that’s opened to shin height. We see the feet of two characters through the gap and they have a conversation. The camera holds this frustrating position for the entire exchange, never taking us inside or showing us who is talking.
If we’re watching a Spielberg movie – and if we know we’re watching a Spielberg movie – then most people will trust that this shot exists for a purpose. This is a setup for a coming reveal or punchline. Maybe we have two men planning a murder, and when the conversation ends they open the garage door and we see that they’re both preists or police officers. Or maybe we realize one of them is a famous actor playing against type. Or they’re dressed as sports mascots. Or they kiss and we suddenly realize these two are a couple. Whatever. The frustrating shot existed to get the audience curious and thus lead to a bigger payoff when the reveal happens and the conversation takes on a new dimension.
On the other hand, if we’re watching a student film with atrocious production values then the under-the-door shot will just be annoying. The audience won’t assume this is a setup for something clever, they’ll assume the filmmaker is trying to be “artistic” in the most pretentious way. Or they’ll assume the weird shot just exists to hide scenery and lighting problems.
I ran into this problem while making this track. Deadmau5 is a big fan of using rhythmic tension through the use of polymeters. You play two different parts in a way that makes it feel like they aren’t keeping proper time with each other. They seem to get out of sync and everything feels a little off-kilter. You can hear this in the track Brazil starting at the two minute mark. The song spends just under a minute in this unstable state where there’s no drum to help you feel the pulse of the song.
Like the garage door shot, this creates tension. If the audience trusts you, then they know they’re being set up for a moment of resolution when the beats align and everything falls into place. If you’re Deadmau5, then you can leverage that trust to make the tension last even longer, and you can make the resolution stronger by having the rhythms get even more disjointed. The same is true for really slow sections. In the context of dance music, a long, plodding intro might create tension or it might frustrate the audience, depending on whether or not they trust you’re going to reward with something really good. You can see Deadmau5 use both of these techniques in the track Strobe. That track spends a full three minutes floating along in a dreamlike state, and then it spends two more stuttering through polyrhythms before it finally settles into the meat of the track at the five minute mark. All by itself, the build-up section of Strobe is longer than most tracks!
But you can’t get away with those sorts of musical shenanigans if the audience doesn’t trust you. Specifically, a lot of people are not open to being challenged by a class project from an amateur composer. Instead of feeling like the music is building up to something cool, they’ll worry they just clicked on some pretentious avant-garde arhythmic nonsense.
Other students were going to hear my song in the context of a feed filled with dozens of amateur tracks. Since this class included some beginners, several of these tracks were quite rough. It stands to reason that people working their way through this feed weren’t going to be open to long buildups with unresolved rhythms.
I was painfully aware of this – perhaps even too aware of it – when I made this track. I originally had a minute-long build up at the start where I slowly introduced the parts and used changing timbreUse sweeping filters and messing with the release time and delay effects. to make you wonder what was going to happen next. Then I started to worry that this would come off as a bit too showy and self-indulgent in the context of a music class. So I hacked it down to the bare bones. in the end it takes just 11 seconds for the music to solidify in a rhythmic and harmonic sense.
Looking back, this was a terrible over-correction. The song basically resolves the tension before the listener even feels it.
I suppose I could always make an extended remix later.
Track 3: Lovell’s Journey
(Above YouTube video is barely related to the topic at hand, but I’m posting it because I feel like Henry Rollins is calling me out. Warning: Contains naughty language and Henry Rollins.)
I knew at the start of the course that I probably wasn’t going to do the third project, which was focused on writing and performing traditional pop song with lyrics and singing.
I’ve said before that I use music as a concentration tool. You know how your brain will suddenly ambush you with annoying crap?
Remember how you let your friend down that one time? Oh, you do? Okay. Just checking.
Wow. You sure are behind on your work / bill paying. There will probably be consequences for that. Maybe you should stop what you’re doing and worry about it right now.
Hey, remember that really embarrassing thing that happened when you were 12? Wasn’t that just THE WORST?
My brain does this to me constantly when I try to write, and I have no idea why. It doesn’t happen when I’m (say) in the shower or trying to fall asleep at the end of the day. During those times, I can concentrate on work just fine. But when I’m actually at the computer and trying to do work, the Larry David part of my brain becomes active and starts making me feel awkward, anxious, and inadequate.
I’ve found that music is really useful for taming that annoying part of my brain. If I have music on, I can concentrate more or less normally. The trick is, it only works if the music is instrumental. I can’t write / code if there’s vocals going onRepeated out-of-context sound samples are fine.. My brain stalls out and gets stuck thinking about the lyrics. This means that even though I love They Might Be Giants, I never listen to them as background music when I’m working because it’s literally too interesting and my brain fixates on it.
(Oddly enough, this lyrics business seems to be a problem that I developed in my 30s. Back in the 90s, I remember coding while listening to lyrical music with no problem.)
So these days I favor non-lyrical electronic music. I knew I wasn’t going to write lyrics or sing for this track, but I did add some voice-based content in an effort to sort of loosely acknowledge the intended project. I have snippets of an interview with Jim LovellHe’s probably best-known as the commander of Apollo 13. and some synth voices singing “Ooh” in the background.
I’m generally happy with how it turned out. The track could stand to be longer and do more different things. A key change here or there might be nice. In any case, this is probably the most solid track I produced in the course.
Anyway. So that’s what I did with October. I might have one more follow-up post where I talk about some of the stuff I learned in the course. We’ll see.
 Use sweeping filters and messing with the release time and delay effects.
 Repeated out-of-context sound samples are fine.
 He’s probably best-known as the commander of Apollo 13.
Dead or Alive 5 Last Round
I'm not surprised a fighting game has an absurd story. I just can't figure out why they bothered with the story at all.
The Mistakes DOOM Didn't Make
How did this game avoid all the usual stupidity that ruins remakes of classic titles?
Crash Dot Com
Back in 1999, I rode the dot-com bubble. Got rich. Worked hard. Went crazy. Turned poor. It was fun.
Do It Again, Stupid
One of the highest-rated games of all time has some of the least interesting gameplay.
Could Have Been Great
Here are four games that could have been much better with just a little more work.