Sound in videogames has made as many leaps as the visuals have over the years. The earliest games had only bleeping sound effects, but the addition of musical scores and, much later, voice acting, immensely strengthened the emotional impact of narrative in games.
Early musical scores were necessarily crude, as 8-bit systems typically only had one sound channel, meaning only one note could be played at a time. Musical chords and unison were impossible, and the tiny storage capacity of the early cartridge games meant that even that one note would be tinny and poorly sampled. Matters improved somewhat in the 16-bit era, when composers like Yuzo Koshiro could overcome even the SNES’s lack of a dedicated sound chip to produce scores such as Actraiser’s orchestral-sounding sweep. But it wasn’t until the CD era that games’ full sonic potential was unlocked.
All games up until that point had been classified under the acronym PCM, which meant that they were stored as bits of code which would instruct the machine to play this particular instrument sample at that particular pitch and frequency, allowing for themes to be compressed to the point where they would stay within their allotted space both on the storage medium and in the system RAM. But since CDs were originally developed as a music playback medium, they were inherently suited toward much more robust sounds than had previously been possible. It was a simple matter to fit the already-extant playback chips inside the new game machines, which meant there was no reason why a game’s music couldn’t contain anything you might hear on a regular CD.
There were two drawbacks: first, CD audio (also known as "redbook" audio still took up a lot of space—at the time, it was common for games to feature a dozen or two songs, whereas there was space for about 9 or 10 redbook tracks if the designers still wanted to leave room for the game’s code. The other problem was that because redbook file sizes were far too large to reside in system RAM, they had to be streamed off of the CD just as a regular CD player would. But because there’s no way to know how much time a player will spend in any given area of a game, the background music for that area must loop to prevent it from stopping altogether, and the joints in redbook audio were obvious when a track ended and needed to be respooled.
Advances in system memory size and the move to the DVD medium helped matters somewhat, today the format of music used by a game is usually determined by what sort of game it is. Long games in the role-playing genre might have as many as 100 songs, and use advances in PCM audio to sound almost as good as live, prerecorded songs might. Arcade games, on the other hand, are increasingly turning to redbook audio (and its modern cousin, the MP3 format) in order to feature licensed high-energy music: the Crazy Taxi series of games uses songs from the punk outfit Offspring as its signature tunes, while the racing game Wipeout 3 highlighted techno acts selected by Sasha, the noted DJ. Microsoft’s new Xbox console goes so far as to allow the player to transfer MP3 versions of songs to the system’s hard drive for use within selected games, giving the player to choose his or her own soundtrack.
The new ability to store prerecorded audio in compressed formats also gave rise to voice acting in videogames, with all the benefits and hindrances that sound gave to film. On the one hand, well-acted characters have an extra dimension and depth in dramatic roles, and a skilled comedic actor can make so-so material better than it really is. On the other hand, poorly-acted characters can make a good game almost unbearable to play through. More and more games are getting the voice acting treatment, though it’s still not feasible to give full voice to every character in the longest games.