On 4/13/05, Adam Wozniak <adam@xxxxxxxxxxxxxxxx> wrote:
> > Okay, we all know that every bit is sacred, but, out of curiosity, why
> > are there 8 bits in a byte as opposed to 10 or 3?
>
> 8 is a power of 2. 2^3 = 8
>
> 4 is also a power of 2, but it's too small to represent a character.
> 16 is also a power of 2, but it's way overkill to represnt a character.
>
> Of course, that was back when everyone spoke english.
Heh, good point.
Actually, and forgive my brutal summarizing and half-forgetting,
(corrections welcome) once upon a time, like the 50s and 60s, comuting
hardware had two distinct branches: business and scientific.
Business WAS generally base 10 -- good for lining up business ledger's
dollars and cents.
Scientific was the binary we all know and love.
Overtime, it seemed that the scientific machines could do the work of
the business machines better than the other way around, and the binary
approach started to win out.
By the time the 6502 rolled around it was even more obvious that
binary worked pretty dang well for a lot of things...or at least
efficiently. I still think it's funny that you can't store many
fractions, even certain "hundreths", correctly in binary--people who
don't know what they're doing in business software can get in trouble
with that.
I think some computers had non-powers-of-2 "word sizes", like 13 bits.
This is why in C, sometimes they don't define exactly how big an
"integer" is, it's assumed to be whatever is convenient on that
machine, so that the compiler can do "whatever's natural" for integer
math rather than going nuts making everything to the power of 2.
Archives (includes files) at http://www.biglist.com/lists/stella/archives/
Unsub & more at http://stella.biglist.com
