Why aren't bits, bytes, etc... base 10?

[Relative]

Why aren't units of information base 10?

eg: 10 bit = 1 byte instead of 8 bit = 1 byte

[lurker]

(Bytes technically don't have to be 8, but that's irrelevant) The reason is very simple. Lengths that are powers of 2 make things MUCH simpler internally.

[Argh]

Because binary switches have only 2 states, on or off. So, everything is a power of two, in computer-land, to keep things rational.

So... why is a byte 8 bits? It's 2^3. A nibble is 2^2. Kilobytes are 2^10. Get it?

[lurker]

To more specifically answer the question of 'why 8', it's because it's the smallest that can hold a single character. And I think argh just wants to draw attention to kilo being 2^10 in compuland, but to be clear kilobyte is 2^10 bytes.

[Peet]

(and the 'correct' representation is KiB, pronounced 'kibibite', referring to 2^10 rather than the 10^3 of the metric K)

[REVENGE]

Fuck switches.

(I kid, I kid.)

[Teutooni]

...or rather, why aren't numbers normally base 8 or 16 instead of 10.

[Pressure Line]

fingers and thumbs on a pair of human hands

[SinbadEV]

fingers and thumbs on a pair of human hands

Hence the work "Digits" having the duel meaning of "Fingers on a 10 Fingered Hand" and "Numerals in a base ten Number System"

[smoth]

tonight...base 2

[Panda]

...or rather, why aren't numbers normally base 8 or 16 instead of 10.

Because people only like to use base 16 numerical systems for things like digital computing where hexadecimal is used. So,

do
{
cout << Base10Numerals;
}
while (cin >> Base10Numerals);


Else, you could try to use a vigesimal (more traditional, base 20) numeric system and there's no C base.

[jcnossen]

Why aren't units of information base 10?

eg: 10 bit = 1 byte instead of 8 bit = 1 byte

The question is inconsistent. 1 byte being 10 bits, would still make it base 2. The base 2 is a property of bits having 2 states, it has nothing to do with the size of a byte.

[Tobi]

Indeed, and the size of a byte is rather arbitrary and depends on context.

For example in old serial lines a byte can be 9 bits, 8 data and 1 parity. Similarly, basic ASCII is only 7-bit.

According to Wikipedia the size of a byte is usually in the range from 5 to 12 bits

Also in more specific context it can be a lot bigger. This says in certain C++ implementation a byte is 64 bits even.

And from same site in particular the PDP-10 seemed to have weird sizes, with 36 bit bytes