You could represent more distinct values with more distinct voltages, and you could use voltages far enough apart to avoid ambiguity.
But circuitry would be a lot more complex.
With binary you have a positive rail and a negative rail, and they run everywhere. Transistors can easily switch to pull a signal to the positive rail to represent a 1, or the negative to represent a zero. To have more states, you’d need more rails to be available everywhere in the circuitry.
Also, some components depend on a “breakdown” voltage where a reverse voltage causes no current flow until you reach a certain threshold, and suddenly the behaviour changes and you get current flow. With a single voltage system, the voltage is either above or below that threshold, which lends itself very strongly to binary.
Also, some circuitry looks for a rising or falling edge of voltage, ie a transition from 0 to 1 or 1 to 0. The voltage doesn’t necessarily rise or fall cleanly, it can be a little jaggard with small ups and downs along the way, so there’s circuitry to wait a moment and decide if the voltage is *really* moving from 0 to 1. Using a multi-voltage system to represent mutiple numeric values would mean this kind of circuitry would need to figure whether the voltage which has genuinely passed the first voltage is now *really* passing the next level too, or if it’s just a bit of noise. And again for the next level after that.
Latest Answers