EXPRESSING NUMBERS USING VARIOUS BASES

When expressing any integer N it is mainly done using a base ten numbering system.
This system involves ten characters 0-1-2-3-4-5-6-7-8-9 and was undoubtedly developed
historically because early man used his ten fingers to count. There are of course also
other number systems including the important binary system using a base two and
requiring only the symbols 0 and 1. We want here to show how one can express numbers
in various bases and then show how one carries out the basic operations of addition,
subtraction, multiplication, and division.

We begin with the decimal system which uses ten as its base. In this system a number-

N  3108  3x103  1x102  0 x101  8 x100

That is, 3 thousands+1 hundred+0 tens + 8 ones make up the number. So for abbreviation
sake one simply writes down in order the numbers which multiply the various powers of
ten. To add two numbers in decimal we simply add up the numbers multiplying various
powers of ten. That is-

37+81=(3+8)x101+(7+1)x100=11tens+8ones=1hundred+1ten+8ones=118

To subtract we have-

87-49=4 tens-2 ones=3 tens+8 ones=38

Multiplication and division follow by operating with the various powers of ten. So-

23x 46  ( 2 x 4) x102  (12  12) x101  18 x100  103  0 x102  5 x101  8 x100  1058

and-
1
135/9={1x102+3x102+5x101}/{9x100}=1x101+5x10=45

The next number system is the binary one. It was first developed by G.W. Leibnitz some
three hundred years ago and has turned out to be essentially the system used in all
electronic computer calculations. The binary system is based upon the powers of two and
uses only 1 and 0 as symbols. To write down a number N in binary we expand a typical
number given in decimal such as-

371x25+0x24+0x23+1x22+0x21+1x20=10010

The binary version of 37 is thus read off by writing down the symbol multiplying the
power of twos in descending order. It is very easy to generate these numbers by hand and
even easier to make use of our MAPLE command-
 convert(N,binary);

Here N is given in decimal and we are converting to binary. One can go between any two
bases by use of the command-

convert(Nb1,b2,b1);

Thus we have, for example, that-

convert(10100110011,decimal,binary); =1331

Here we have carried out the reverse procedure of going from a binary number to its
equivalent decimal form. A quick table using the conversion operation –

convert(N, binary, decimal);

follows-

Decimal N Binary N
0 0
1 1
2 10
3 11
4 100
5 101
6 110
7 111
8 1000
9 1001
10 1010

From this table we see at once that two to a given power n has the equivalent binary
form 1 followed by n zeros, That is, 641000000 and 102410000000000 . Also
one notes that doubling a number just adds a zero at the end of the number. The addition
of two numbers in binary, say 3 and 6 reads 11+110= 1001. That is-

0+0=0 , 0+1=1 and 1+1=10

so that 111
+11
---------
1010

The subtraction rule is seen to be- 0-0=0 , 1-0=1 , and 1-1=0

So that 1001
- 101
______
100

Thus any number in decimal can be represented in binary. To treat decimal points in
binary we look at the negative powers of two represented in binary. Thus 1/20.1,
1/40.01, and 1/80.001 . So 1/2n0.00….001, where the number of zeros after the
decimal point equals n-1. Also the decimal number such as 2.456 converts to
10.01110100 in binary.

In looking at both the decimal and binary form of representing numbers one can see that a
large base b requires more characters to represent a number but offers the advantage that
the length of a given number will be shorter than a base with fewer characters . We have
constructed a small table which gives the length of a number in decimals compared to
that in binary. Here is the result-

Numberof digits in Number of digits in Ratio of digits decimal

Binary Decimal to digits binary
3 1 3
6 2 3
9 3 3
13 4 3.25
16 5 3.20
19 6 3.1666…
23 7 3.2857…

Although not exact, the table indicates that digit length of a binary number is about three
times longer than its decimal counterpart and that this ratio increases very slowly as the
decimal length becomes even larger. The number 2300 has 91 digits in decimal and 301
digits in binary. The ratio between the two is 3.3076.. and so lies slightly below
log(10)/log(2)=3.3219 which appears to be the upper limit.

Another often used number base is the hexadecimal system. It expresses numbers in
powers of 16 and usually uses the following sixteen symbols-

{0, 1, 2, 3, 4, 5, 6, 7, 8, 9, A ,B, C, D, E, F }

to represent a number. Consider the following identity-

357=1x162+6x161 +5x 160 Its hexadecimal form therefore reads 165

One of the advantages offered by the hexadecimal system is that 16=24 and so can
easily relate to the binary system and hence be of use in computer operations by reducing
storage requirements. For example the hexadecimal notation 165 with 3 symbols can be
used to represent 101100101 in binary form. This works as follows-

165162+6x161+5x160=28+26+25+22+20=101100101

The digit length ratio between a hexadecimal base and a decimal base is about
log(10)/log(16)=0.8304. The ratio between a binary to hexadecimal base digit length is
log16/log2=4. Some even larger bases relying on powers of two are the bases 32 and 64.
Their use in computer technology is clear. Again the advantage of such power of two
bases is that it allows storage of large amounts of binary data in a much more compact
form.

There are many additional number base systems but these play only historical roles and
are no longer in common use. These include the base 20 system (vigesimal) of the Maya
undoubtedly established by counting of ten fingers and ten toes. Also the ancient
Sumarians and later the Babylonians used a base 60 system (sexagesimal) probably based
upon the solar year of 365+1/4 days. The number 60 can be considered as a rich number
since the number of divisors 1,2,3,4,5,6,10,12,15,20,30,60 is large.This makes the
number 60 ideal for expressing numbers which are fractions and multiples of sixty. The
sexagesimal system is partially retained to the present day in angle measurements and in
the relation between hours, minutes and seconds. Both base twenty and sixty systems are
now obsolete for mathematical manipulations required in everyday commerce. Already
the base ten system of the ancient Romans using only combinations of the symbols
I,V,X,L for the first ninety-nine integers was quite impractical for carrying out
multiplications and divisions. Their system also lacked the concept of zero which was not
introduced into Europe from the Middle East until about 12 hundred AD.

Summarizing we can say that any number N can be expressed in a base b system as -

0
N   an b n  a p b p  a p 1b p 1  ....  a0b 0
n p

So we have, for example, that-

2x101+4x100=1x24+1x23=4x61+0x60

In abbreviated form this number then reads 24 in decimal, 11000 in binary, and 40 in
base six(heximal). A base six system has so far found only very limited application. This
may change in view of our earlier work on integer spirals where all prime numbers p of
five and above satisfy the condition that p mod(6)=1. All Mersenne Primes M[p]=2p-1 ,
with p >3, have the form 6n+1 and all Fermat Primes F[N]=2^2^N+1, with N1, have the
form 6n-1.

U.H.Kurzweg
Dec. 19, 2016