I realise (/think?) that CPUs essentially treat two different voltages as a 1 or 0, but what stops us from using 3 or more different voltages? Wouldn’t that exponentially increase the CPU’s throughput by allowing for decisions with greater than two outcomes to be calculated in one cycle? This would presumably mean that a LOT of stuff written for base 2 would need to be updated to base 3 (in this example), but I can’t imagine that’s the only reason we haven’t done this.
I feel like I’ve explained that poorly, but hopefully you get the gist.
In: 12
> Wouldn’t that exponentially increase the CPU’s throughput by allowing for decisions with greater than two outcomes to be calculated in one cycle?
Yes and no. It would allow the same hardware primitives to represent more information, because rather than just measuring “is the voltage over or below X” and getting a 0 or a 1, we could measure “is it below X or above Y or in between the two”, and get a 0, 1 or 2.
But to do that, those primitives would have to be bigger and more complex, and changing or reading their state would have to be more precise, which means it’d be slower. So overall, you wouldn’t really gain anything from doing it.
One of the obstacles keeping us from making computers faster than they are is that it takes time for the voltage to stabilize in a transistor, when its state changes. And that problem would be worse if we needed the voltages to reach more precise levels. As it is, there’s just one threshold it has to be above, for its value to be counted as a one. So it’s ok if it’s still a little bit wibbly, as long as it isn’t so wibbly it risks taking us across that line.
But if we add more thresholds (below this value means zero, above that and bellow *this* means one, above *that* means two) then we also require the voltage to be controlled more precisely. We can’t allow it to be as wibbly as it was before. And that means we have to be more careful and take a bit more time to set it.
Latest Answers