How Old School 8 Bit Sound Chips in Phones Worked

First and second generation featured an array of wonderful old school ringtones. Usually, the sound of these tones were generated by the type of 8-bit sound chips that were installed in most personal computers and game consoles of the time. You know the kind, remember the SuperMario theme with it’s beeping symphony of tones?


Well, have you ever wondered about the technicalities of how these 8-bit sound chips actually worked? We have found the best explanation, in simple terms put forward by The 8-Bit Guy. It’s so good that we had to transcribe the whole video to make it easily readable. ┬áHere is the video and the full transcription is below.

Video Transcription

Hello, This is David from the 8-bit guy and today I’ve got a special guest star with me, Rob from the Obsolete Geek.

Today we’re going to talk about how old-school music works. In the early days of home computers, most of them just had a simple beeper speaker. Examples would be the IBM PC and the Apple.

So these speakers were controlled directly by the computer CPU. The speaker could make clicking noises and the CPU would have to time the clicking noises exactly right in order to produce certain tones. If a programmer wanted to spend the time, some very advanced sound of music could be produced this way, but the trouble was it would require all of the CPUs runtime to accomplish that, leaving nothing left over for the computer to do anything else.

Dedicated Sound Chip Of A ComputerSo by the early 1980’s most computers and game consoles had dedicated sound chips to take this load away from the CPU. Each system in that time had a very characteristic sound. Along with its unique style of graphics which helped to get each system its own personality.

Okay, so the first thing you need to understand is that different systems had different numbers of voices or channels. To better explain how that works, let me show you this old musical keyboard. It only has a single voice and as you can see, it cannot play more than one note at a time. In order to do that you would need to have at least two voices.

Now this keyboard here is considerably more advanced than the other one and it has a total of four voices. That means you can play up to four notes at the same time. So having multiple voices is great, but those voices also need to be flexible. They need to enable the keyboard to create different types of waveforms. For example, this keyboard can change the waveform of its voice to produce different types of sounds.

Okay, so let’s take a look at two of the more popular systems from the 1980’s in order to see two different approaches to creating music. The Nintendo Entertainment System had five voices and the Commodore 64 had three voices.

The Commodore 64 Personal Computer
Commodore 64 home computer from the early 1980’s

Now you might immediately assume the Nintendo was better but actually, it wasn’t. Here’s why; The voices used in the NES were, for the most part, stuck making one type of sound. The first two voices can only produce square waves that sound like this. The third voice can only produce a triangle and is typically used for the low bass notes. The fourth voice can only produce noise and the Fifth voice is for PCM sampled sounds, which was rarely used but a good example would be Super Mario Brothers 3. You can clearly hear the steel drum sound being used.

Because the way Nintendo music worked all the music sounded pretty much exactly the same. The tune might be different but it was like they used the same instruments, so to speak.

The Commodore 64 had three voices. It could produce four different types of waveforms, Square, Triangle, Sawtooth and Noise or any combination of those.

In the early days, most programmers would just assign a certain sound to a particular voice and just leave it that way throughout the entire song. That was the simplest thing to do.

Here’s an example from the game M.U.L.E. But not long after, some clever programmers realized that it was possible to dynamically reassign the voices to other waveforms on the fly. This gave the illusion of having more than three voices.

Take this example from Commando. In order to better understand what’s going on here try listening to one voice at a time.

A few years later the IBM PC finally got a decent sound upgrade in the form of the Adlib card which used the Yamaha ym3812 sound chip. Shortly after the market share was lost in favor of the Soundblaster card which also used the same ym3812 sound chip. So this chip was basically the foundation of computer music in the IBM PC world for the next 10 years.

The ym3812 had nine voices and much like the Commodore 64 the voices were independently programmable. Incidentally, this chip was also used in several Yamaha keyboards such as this one. Take a look inside and you’ll see the ym3812 sound chip. It’s almost as if you could take a Soundblaster card and attach keys and speakers to it and you could play it like an instrument.

So take a listen to this little sample of music from the game Ultima 6 and now listen as I recreate that same sound on this keyboard.

All right, so let’s talk about sampling for a moment. Now one of the neat things about this keyboard that I haven’t shown you yet. This came out around 1985 and has four voices but one of the things that makes it interesting is it’s a sampling keyboard and let me show you exactly what that means. “8-Bit Guy”.

So this was not the only device to come out in 1985 that featured a four-voice sampling system, the other was the Commodore Amiga. The Commodore Amiga was the first affordable home computer that featured a four-voice stereo sampling sound system and with it came a new type of computer music known as the mod tracker. These were music files that contain samples of different sounds and their associated music information.

The original mod tracker used a four-track system designed around the Amiga sound chip but later versions eventually added many more tracks for more sophisticated sound cards. This type of format is still in use today as a method for composing new music.

The mod tracker format is not used quite so often anymore, with the abundance of storage and memory on modern machines they pretty much just forego all of that type of music synthesis in favor of just using one gigantic sample, usually in the form of like an mp3 file or something like that.

Alright, well that about wraps that up. I hope you found that interesting, maybe learned something you maybe didn’t already know.

It also may have come to your attention that I have changed the name of my channel. The reason is, it’s been pointed out to me on several occasions, that I haven’t made really videos about Apple iBooks in quite some time, so I change the name from the iBook guy to the 8-bit guy because it’s a little bit more representative of what I actually do here.

Also, I want to take a moment to thank Rob for being on my show and why don’t you tell us a little bit about your channel for a moment. “Well thank you for having me on your show, I’m a big fan of your channel and one of the things I like to talk about on my channel are a little bit more obscure, a little more unusual Hardware as it relates to video games. Like this sharp x68000 computer from Japan, one of my all-time favorite systems.”

And if you want to see a little bit more about this piece of obscure equipment there’s a link down in the description field you can click to take you over to his channel, and he’s got a lot of other really obscure stuff in his collection that you can have a look at.

Alright, well also don’t forget to visit me on Facebook and I’ll see you next time!


Leave a Reply

Your email address will not be published. Required fields are marked *