I would absolutely hate to be making music for video games in the 1980s.
Don’t get me wrong. I love classic game music. The hours spent playing and listening to classic NES, Super Nintendo, and Gameboy games had an enormous influence on me. However, the restrictions composers such as Koji Kondo, Hip Tanaka, Kazunaka Yamane, and Hiroshige Tonomura (just to name some of my favorites) had to deal with to create the iconic pieces of music we know and love are so mind-boggling, I’m thankful to not have to deal with them. Here’s a crash course on what making music for video games used to be like, as seen through the lens of our recent Super Adventure Box release in Guild Wars 2.
I’ll focus primarily on the Nintendo Entertainment System, since it’s probably what the majority of people think of when they think of classic video games. For a much more technical breakdown, see the Wikipedia page on the NSF Sound Format.
Limited Choices
Back in the day, the sound chip inside the NES console was only capable of playing back five channels of audio at one time. There were four very basic synthesis channels and one channel that could be used to play back very low resolution samples. The sample channel was used much less frequently than the other four channels, so I won’t focus on it much.
The remaining four channels were as follows:
Channels 1 and 2 – Square (or Pulse) Wave
These channels do the majority of the heavy lifting, providing the main melody, counter-melody, harmony, and arpeggios. They are the most flexible channels, in that you can play around with the overall timbre, volume, and vibrato for each note as well as do some other fun stuff like pitch bending and even simulated echo and reverb, if you get a little tricky.Channel 3 – Triangle Wave
This channel is almost entirely used for bass lines. It’s possible to use it for other sounds, but it’s not nearly as flexible as the square wave channels. You can’t play around with volume or timbre.
Channel 4 – Noise
This channel generates a white noise hissing sound. It’s possible to alter the pitch a bit, but it never really becomes a proper note. It’s either just “lower†or “higher†noise. It’s used almost exclusively for creating drum/percussion parts.
And that’s it; that’s your total sonic palette. No eighty piece orchestras, no awesome band or electronic artist to write some tracks for you. Just four channels. There was also no polyphony, or multiple notes played simultaneously by one instrument, so no chords. Finally, every piece of music and sound effect had to fit into 128k of memory. Good luck on your adventure.
Tracking Numbers
Next, let’s talk a little bit about how these folks had to actually compose the music. Nowadays, a composer generally writes and records all their music into a computer program called a DAW (Digital Audio Workstation), either recording live instruments or using virtual instruments loaded on the computer for each part. There are very few limitations placed on the composer, and they have more to do with time and budget than with the technology available. However, back then, it wasn’t so easy. I’ll pass it off here to my colleague Leif Chappelle, who also wrote some pieces for the Super Adventure Box content.
Leif Chappelle:
In the days of the NES and other 8-bit systems, there were no standard ways of getting music into a game. Composers either had to program hexadecimal data that called up the sounds we know, or write their own program to create that data. We call these programs “trackers.â€
One of the main tools I use to write chip tunes is a program called Famitracker, which is a more modern approximation of a tracker used by 8-bit-era composers. Because it’s not the most traditional means of composition, let me go into a bit of detail on my process.