I don't get it, why is metric better?

[Bill]

For instance, if you wish to write down a 32 bit string, 8 hex digits is the "nicest way to do it, for a person.

[Scott Lurndal]

Only on the old BCD mainframes was any form of base-16 used, and there, values of A-F weren't available for use (they were called undigits on the Burroughs systems and would cause a fault if used in an arithmetic operation - integer, fixed-point or floating point.

All modern processors do arithmetic in binary[*]. Don't confuse the storage format with the human representation of the storage when printed.

[*] IBM's Power processors also support decimal floating point (a la BCD).

[Scott Lurndal]

And if you have 12-bit systems, 3 octal digits were the nicest way to do it (e.g. PDP-8, PDP-12).

For BCD machines, decimal rules. Makes it very easy to read core dumps.

[Scott Lurndal]

Ugh, 4 octal digits, of course.

[Bill]

I understand many (most) traditional pocket calculators use BCD.

[krw]

Of course but because machines like binary and we don't, doesn't mean it's the only use.

[krw]

It reduces error accumulation. Many computers were BCD, as well, for pretty much the same reason.

[krw]

Huh? How can you have base-16 arithmetic and not use A-F? That paragraph makes no sense.

It depends on how you look at it. The hardware uses binary logic, sure, but the arithmetic is purely hexadecimal. Normalization is done in hexadecimal digits and the "binary point" is actually a "hexadecimal point", for instance.

Again, who was talking about BCD?

[Bill]

That's not the reason it is used in pocket calculators (where there is virtually no "error accumulation").

[Puckdropper]

krw wrote in news:dfdptb93eie6dngsb31g0i0cum6cr97qaa@

4ax.com:

If we'd simply stop counting our thumbs and use them as status bits instead, binary would come a whole lot more naturally. Teach your kids to count properly: One, two, three, four, overflow, sign, five, six, seven, eight.

The status bits might need a bit more thought.

Puckdropper

[krw]

Think about fractions (as mentioned earlier in this thread).

[krw]

How about using them for hexadecimal. It might take some Vulcan coordination, however.

Status bits seem pretty simple, at least for your base-8. Increment right to left, decrement left to right. Overflow then becomes the count after either pinkie (pinkie and ring change together) and sign becomes a decrement or increment past zero.

[Bill]

Nope, that's not the reason either. Guess again.

[krw]

You're wrong.

[Bill]

Hmm.. The reason calculators don't use binary is because the translation from decimal to binary to do a calculation, and then convert to decimal output again is generally less efficient than using BCD. Just saying...

[krw]

As I said, you're wrong. Conversion is a trivial matter (modulo divide by "10" and post the answer to the display - repeat). The problem is adding 1/3 + 2/3. People understand that .3333333333 is

1/3 and .666666666 is 2/3 but they don't like the answer to be .999999999. The logic to make it "right" in every case wasn't trivial for early calculators.

[Len]

[Len]

You guys ( and ladies) are making a mountain out of a mole hill. You eithe r accept one or the other system. No need to compare "equivalent" values. The British system is based on 12. The metric system is based on 10. It is that simple! If you can multiply and divide bye 12 in you head you're d oing very well and are just a good as the guy using metric values. Of cour se, since you are both using a decimal system to perform the division, the metric guy has the advantage. Most people can multiply and divide by 10 a lot easier and, probably with less errors, than the person doing it in base 12. I grew up with computers; base 2, 8, and 16 are very common there. S ometimes I can add in those systems, but subraction, multiplication, and di vision totally escape me.

I am not very smart, so I will stick 10.

Len

[J. Clarke]

advantage. Most people can multiply and divide by 10 a lot easier and, probably with less errors, than the person doing it in base 12. I grew up with computers; base 2, 8, and 16 are very common there. Sometimes I can add in those systems, but subraction, multiplication, and division totally escape me.

If the Imperial system is "based on 12" then explain relationship between feet and yards and the relationship between ounces and pints.

And how does metric let you, for example, calculate the quantity of carpet you need for your living room any better than Imperial does? Is your living room exactly 10 by 10 meters or something?

[krw]

Nah, it's based on the factors of 60. Each unit uses different ones. ;-)

Well, when one has a living room that's 5yds, 2ft, 3-37/64in by 4yds,

1ft, 7-21/64in, sure it's hard to come up with the area. Then you have to figure out whether to include the base molding! ;-)