Path: utzoo!attcan!uunet!ogicse!ucsd!ucsdhub!celit!dave
From: dave@fps.com (Dave Smith)
Newsgroups: comp.arch
Subject: Re: Decimal Arithmetic
Message-ID: <7413@celit.fps.com>
Date: 19 Mar 90 22:12:08 GMT
References: <27696@cup.portal.com> <76700180@p.cs.uiuc.edu>
Sender: daemon@fps.com
Reply-To: dave@fps.com (Dave Smith)
Organization: FPS Computing Inc., San Diego CA
Lines: 54

In article <76700180@p.cs.uiuc.edu> gillies@p.cs.uiuc.edu writes:
>
>nEven today, maybe we don't know the full intellectual costs of using
>binary numbers in computers.  I imagine Von Neumann weighed the
>technical advantages, and ignored the human factors.  I imagine he &
>his group did not forsee these costs:
>
>(1) Professional programmers must be fluent in 3-4 bases
>    (binary, octal/hex, decimal), boolean logic, and arithmetic systems
>    (sign-magnitude, one's complement, two's complement).  This is no
>    trouble to Von Neumann, but try to teach most high-school kids this.

Base switching can be confusing, especially when someone has an octal number
print out with no indication that it's in octal.  Boolean logic is the
basis of the system, however.  How would using a decimal system make
it different?  We'd still be and'ing and or'ing things and doing and's and
or's in decimal is decidely non-trivial to do in your head.

The system almost has to be binary at the base levels, unless you want to
run multiple voltage levels to represent the various numbers.  That's more
of a nightmare than I would like to think of.

Trying to teach most high-school kids their ABC's is a major problem.  
Having a different number base for computers wouldn't make them any easier
to understand.

>(2) Numerical algorithms must deal with binary roundoff.
That's the machine's problem.  You'd have to deal with decimal round-off
otherwise.  Once you know the why's and the algorithms, it's not tough to 
cope with.

>(3) Accounting algorithms must hassle with representing .1 in binary

They have to hassle with representing 1/3 in any format.

>(4) The industry struggled for years with 6/7-bit characters (too
>    little), and finally settled on 8-bit characters (too much).  
>    In decimal, 2-digit characters would have made sense on day 1.

8-bit characters are still too few for many languages.

>(5) The necessity to learn / program conversion algorithms.
>
>Can anyone think of other human costs of binary numbers?

I can think of one advantage.  Since I know octal, I write all the
PIN's for my cards that I don't use often in octal on the cards.  I can
do the conversion quickly but anyone who steals my cards will be quite
confused.
--
David L. Smith
FPS Computing, San Diego
ucsd!celerity!dave or dave@fps.com
"We are a bigger musical genius than any Bob Dylan" - Milli Vanilli