Game Sound — Dinaburg Writes

by Karen Collins

Through the ease and ubiquity of youtube playlists with increasingly bananas titles like “relaxing SEGA megadrive US genesis songs”, “another relaxing and *~*~*~beautiful~*~*~* sNeS RPG music vol. 3”, and “best video game songs to feel good and fill your heart with memories,” madeleining my way to the soundscape of my preteen bedroom is a lot easier than digging up the betamax tape upon which rests the video of me beating Sonic 2 (for swag) or entering the dark world in Link to the Past (so i could idle the music without leaving my SNES running).

There is more to game audio than nostalgia, though: I live and die with ambient ocremix–a videogame music remix website–running at home, both because I like it and because instrumental-only music helps toddlers focus on rhythm and instrumentation without splitting their limited cognitive function trying to parse the meaning of the words flying around. I’ve also heard, though cannot state definitively, that the relatively simply square wave is more pleasing (/comprehensible?) to a newborn than complex and shifting waveforms of non-synthesized music.

I picked up Game Sound: An Introduction to the History, Theory, and Practice of Video Game Music and Sound Design to answer questions such as why I kept telling my wife, “Newborns love the square wave’s simplicity, that’s why chiptune remixes are good”, or asking myself why I always know when an NES song was from a game published by Konami. Beyond these quests for particular facts, I love deep dives into how the physical limitations of hardware shape the world as we receive it:

It [the Sega Genesis] had an equivalent PSG 3 + 1 chip to handle effects and the occasional music part ( a Texas Instruments SN76489, which had been used in the ColecoVision), as well as a Yamaha FM synthesis chip, a YM 2612, which had six channels of digitized stereo sound, and one PCM 8-bit sample channel (the same chip used in the popular Yamaha DX27 and DX100 music keyboards).
The Sega Genesis was the most popular game console to employ FM sound….Sound on the Genesis was rather difficult to program for, however, requiring assembly language code gto engage the two chips…Each FM channel had four operators (waveforms) assigned, each with a frequency and an envelope with which to modify the input.

Is this why Toxic Caves from Sonic Spinball sounds so good? One can only describe the Genesis sound profile as “crunchy,” and as I’ve continued to listen to VGM podcasts like KVGM, Legacy Music Hour, and Pixelated Audio I’ve realized the Genesis’ FM synthesis is more interesting, to me, than the SNES wavetable synthesis. Game Sound gave me a way to not only interpret that hardware distinction, but internalize the difference.

The opening half of Game Sound focuses on hardware breakdowns, similar in style to what I loved from I AM ERROR. It did everything I wanted it to: a detailed history of videogame generations–it even outscooped my living memory, as I confused the isometric proto-roguelite adventure game Toejam & Earl with its sequel, Panic on Funkotron, which the book correctly names a “music side-scrolling platform game”–and all the hardware is discussed chronologically in as discrete hardware cycles as can be made.

When things veered away from consoles to the PC, a truly wild world of massive hardware differentiation was revealed. Did your SoundBlaster card crash out every time you tried to run Quest for Glory 4? Did you have a MIDI player running old game music while you tried to apply a fan-translation patch to FF5? I sure did. MIDI meant almost nothing to me—I did not realize it was a file-type distinction until right now—it was just words, like WinAmp or REALplayer.

The most significant advance of the 16-bit era sound, however, was the adoption of the Musical Instrument Digital Interface (MIDI) protocol.
MIDI was defined in 1983 to allow musical devices (synthesizers, keyboards, sequencers, mixing decks, computers, etc.) to be compatible in standardized format. Only code was transmitted, rather than actual sounds, meaning the file size was very small. A MIDI command might, for instance, tell a synthesizer when to start and stop playing a note, at what volume and what pitch, and what voice (instrument) or sound to use. Initially, some of this information would vary greatly depending on the devices used, which complicated programming for soundcards, but in 1991 a General MIDI (GM) standard was agreed upon….Another problem with MIDI, however, was the fact that most MIDI hardware playback devices sounded quite different. A slap bass on one person’s soundcard might sound very different from a slap bass on another person’s soundcard, and what sounded good on a composer’s specialized soundcard might sound very poor on a cheap card.

I’m getting shades of the realization that emojis were different across phone platforms, and how these unknown distictions might skew tone as messages hopped between hardware ecosystems.

Solutions composers came up with to manage hardware-mediated outputs resulted in More Work for Mother: code that would read which hardware setup you had and would call different files (thus multiple versions of the same composition were required) or having a generic recommendation for what to use and just hoping for the best.

There is more to the book than awesome hardware charts and historical nuggets, though: chapter 6, for me, felt especially kludgy. Once we broke into the “industry” side of things, it unfortunately was more homework than not. Perhaps it felt that way because I was assigned Everything You Need to Know about the Music Business for my Music Licensing seminar, which covered a lot of the same basic material. Reading it in 2023 made the lift even heavier: once the book stretched beyond the historical record, the material began to show its age:

In terms of playback, it may be that we will see games where users can insert and tag their own playlists. Users can already insert their own music into many games—all games on the Microsoft XBOX360—but as yet they cannot specify where they wish the songs to play back. Mood tagging, for instance, may allow players to choose what battle music they want to hear.

This was during the mainstreaming of Web 2.0, where users were expected to perform much of the labor surrounding the “expression” of massive database organization. Couched in terms of “You get to have it your way!”, crowdsourced acts like tagging and uploading songs and creating tangles of “mood catalogues” was simply captcha technology across a different industry; work captured by large companies by unwitting, unpaid, human participants. This has largely fallen away—“lean back” choices like algorithmic recommendations and created playlists (that likely draw their conclusions from the work done during Web 2.0 crowdsourcing booms) have conquered each walled kingdom of the corporate-mediated cyberlandscape—an outcome no one could have foreseen in the mid-aughts. Perhaps Fortnite and Roblox see their massive userbases inserting their own playlists and songs, but such scope is outside the realm of this book and this review (side note: I am very interested to read academic breakdowns of Fortnite once that cohort ages into phd candidacy).

It is not incredibly fair to judge when Game Sound sticks its nose into the business side and tries to be more forward-facing, because we have fifteen years of hindsight available; can’t blame Game Sound for not being Cassandra. Around the time the book was released, games didn’t really know what they were doing with regards to popular music. Example: this was the dawn of the looter-shooter (unfortunately [ironically?] now named the shlooter) genre, popularized with the hit Borderlands. Its iconic music-video-style opening still lives in my head when I hear the song:

In what can only be coined a cosmic coincidence, the professor that taught my Music Licensing seminar and assigned me Passman’s Music Business book was also the manager for Cage the Elephant, the band whose song introduces you to the Borderlands. Perhaps he read Game Sound when it came out.

By and large, Game Sound is a fantastic look at exactly what its subtitle claims: it is an introduction to the history, theory, and practice of video game music and sound design. It is rude of a reader to expect to enjoy the sections on history and theory and practice in equal measure. Game Sound is the premier place to start if you’re curious at all about how a listener might, for example, be able to distinguish a Capcom track from a Konami one upon first listen.