Why we use binary and not trinary

[chrisname]

I was thinking, "Why not use trinary as an internal numeral system instead of binary?" Trinary is surely superior: a 4-byte (signed) integer in trinary could store 6.17673396 ^ 10¹⁴ instead of the relatively tiny 2,147,483,648 we have now.

I have found the answer. A byte is made up of a number of bits,"bit" being an acronym for binary digit. Well, think about what would happen if you tried to shorten ternary digit. That is why they decided to use binary instead of ternary.

[firedraco]

Uh...ok....I guess...that makes sense.... <_<;

[Return 0]

lol

[jRaskell]

I do like your answer, but it really has more to do with how the hardware itself works to handle the data. Digital electronics work on handling high and low states of voltage. Low state being 0v, high state being say, 5v for TTL signalling. In fact, TTL standards define the middle range between .4-2.6v or .8-2.0v (depending on current rating) as undefined, neither high nor low. These high and low states ultimately correspond to the 0 and 1 values of binary data in a digital system.

Sure, hardware to handle trinary data can, and in fact has been built. It's generally not only more complicated, but also slower and more sensitive to noise.

[chrisname]

I like your answer too. I suppose trinary is "too analogue" to be practical.

Having said that, imagine if every time you tried to describe your code, you had to say "this variable can store 32 tits"...

[helios]

You could say "trit", "terit", or "ternit", instead.

Trinary is surely superior: a 4-byte (signed) integer in trinary could store 6.17673396 ^ 10¹⁴ instead of the relatively tiny 2,147,483,648 we have now.

So why not just go back to decimal computers, then? Oh right! Because they suck.
Also, fiber optics with dimmer.

[tummychow]

LOL.
But the real answer, AFAIK, has to do with the fact that transistors really have no third state. (Or that it's not worth the loss in speed/efficiency.)