Number Bases

Unusual number bases, including Fibonacci and the irrational Phi bases, factorials and binomial coefficients bases, rational number bases, negative bases, negative "digits", with interactive Calculators for all of these.
The calculators on this page require JavaScript but you appear to have switched JavaScript off (it is disabled). Please go to the Preferences for this browser and enable it if you want to use the calculators, section links and other interactive features, then Reload this page.
Contents of this page
The Things To Do icon means there is a You Do the Maths... section of questions to start your own investigations. The calculator calculator icon indicates that there is a live interactive calculator in that section.
With thanks to MikeMcl and his BigNumbers JS library (readMe hosted on github). It is used here in the JavaScript Calculators for arbitrary precision number calculations.

Representing whole numbers in different bases

If you are familiar with bases 2,3,... then skip to the next section. Otherwise begin here for basic information on bases.
We normally write numbers in base 10 so that each digit counts a power of 10 in the value:
2017 = 2×102 + 0×102 + 1×101 + 7×100
The word digit means finger In the American and UK systems of measuring liquids in pints and gallons, there are 8 pints to 1 gallon, so 20 pints represents 2 gallons and 4 pints.
It used to be the case that 8 gallons make 1 bushel in the UK but the American and British systems are now different. This is a base 8 system where we counts in 8s:
90 pints is 1×82 + 3×8 + 2:
90 in base 10 is 132 in base 8

Digits

We see that in base 10 (the decimal system), we need only 10 symbols, the digits 0, 1, 2, 3, 4, 5, 6, 7, 8 and 9.
In Base 8 (octal) we need only the 8 symbols 0, 1, 2, 3, 4, 5, 6, 7 because 8 in one column will convert into 1 in the next higher column.
Similarly in base β we need only β symbols for the digits, one digit per column.
So what if the base is bigger than 10?
Computer engineers often use this since computers work in Base 2 (Binary) using binary-digits 0, 1 and it was often more convenient to use base 16 (hexadecimal) using the "digits" 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, A, B, C, D, E, F.
In base β we need only β 'digit' symbols
To show what base a number is in, we write its base as a suffix after the number representation, e.g.:
90 = 1328
10 = 128
7 = 125
If we omit the base then the base used is 10.

Names of Bases

base 2: binary
base 3: ternary, trinary
base 4: quaternary
base 5: quintal, quinary
base 6: sextal, senary
base 7: septal, septenary
base 8: octal
base 9: nonary
base 10: decimal, denary
base 11: undecimal
base 12: duodecimal
base 16: hexadecimal
base 20: vigesimal
base 30: trigesimal
base 40: quadragesimal
base 50: quinquagesimal
base 60: sexagesimal
base 70: septuagesimal, septagesimal
base 80: octagesimal
base 90: nonagesimal, nonogesimal
base 100: centesimal
The vigesimal system was used by the Mayan civilisation in central America and also in Britain where 20 is called a "score". Other cultures used a base 5 system (New Hebrides) or a binary system. 12 (duodecimal) was also popular because it is easy to divide by 2, 3 and 4 and the ancient Babylonians used base 60 (sexagesimal) because it was also easy to divide by 2, 3, 4 and 5. The system of money used in Britain from the 8th century until 1971 had 12 pence (d) in a shilling (s) and 20 shillings in a pound (£).

An easy method to convert a number to and from base β

Let's look at an example.
To convert a (decimal) number to base 6, say, keep dividing the number by 6 recording any remainders. The remainders become the base 6 representation of the number.
We start with 6 dividing 5755 and work upwards , repeatedly dividing the quotients until we reach 0
0rem = 4
4rem = 24 = 6×0 + 4
26rem = 326 = 6×4 + 2
159rem = 5159 = 6×26 + 3
959rem = 1959 = 6×159 + 5
5755  5755 = 6×959 + 1

To see the base 6 representation, write down the remainders from the top to the bottom:
5755 = 423516
This works for any numeric base and shows that the 'digits' of base β must be 0,1,2,...,β-1 because they are the remainders on dividing by β.
Because we repeat the process on the quotients until we reach 0, we can see the 'digits' of the quotients too:
4 = 46
26 = 426
159 = 4236
959 = 42356
5755 = 423516

How to find n in base B Calculator

Show Base conversion maths C A L C U L A T O R
in base

R E S U L T S



 
calculator: Show maths Factorial Binomial Rational Multiples Base converter

Converting a base β number to base 10

To convert a number from base β to decimal we reverse the process, starting with the leftmost digit and continually multiplying it by the base β and adding in the next digit:
Here we convert 42356 to base 10, accumulating a final sum as we go from left to right across the 'digits':
To convert 42356 to base 10
'digits'4235
6
×sum
04=
24
26=
156
159=
954
add 'digit'426159959
The final sum is the base 10 value: 42356 = 959
Convert between base 10 and other integer bases in the Base Converter Calculator below.

Horner's Rule

To evaluate a polynomial, instead of evaluating
P(x) = anxn + an−1xn−1 + ...+a1x + a0
by multiplying each power of x individually by its coefficient, it is much easier and more efficient if we accumulate the sum by using Horner's rule:
P(x) = ((..( ( an) x + an−1 ) x + ... + a1) x + a0
So to evaluate 42356 = 4×63 + 2×62 + 3×6 + 5, we calculate instead:
((4×6 + 2)×6 + 3)×6 + 5
which is what we are doing in the table method of the previous subsection.

Factorial base

Permutations

Instead of using powers of a number as the column headings, we can use other series of numbers too, for instance the factorials.
The factorial numbers count the number of permutations of n objects.
In general, whatever the objects, they will be in first place, second place and so on in the permutation list.
So we can just use their positions 1,2, ... ,n as the objects and name the objects 1,2,...,n.
So the ways to permute 3 objects (to arrange them all in some order) are: 123, 132, 213, 231, 312, 321.
If we were permuting the letters of CAT we use 1 for C, 2 for A and 3 for T. Ther 6 permutations are therefore:
123=CAT, 132=CTA, 213=ACT, 231=ATC, 312=TAC, 321=TAC
Why are there exactly 6 permutations?
Because there are 3 ways to choose the position in which to put it including where it is now.
There are 2 position for the second object and then the third object goes in the remaining empty place: 3x2x1=6.
Similar reasoning shows there are 4×3×2×1 permutations of 4 objects.
For n objects there are n! = n×(n−1)×...×3×2×1 permutations, called factorial n, also written as n!.
n12345678910
n!126241207205040403203628803628800

A Factorial-base system

The factorial numbers can be used as the columns of a mixed base system, arranging them in reverse order: ..., 5!, 4!, 3!, 2!, 1!.
The digits allowed in each column depend on the column number: the first column (1!) can have 0 or 1;
the second column (2!) can have digits 0, 1 or 2;
and in general the n-th factorial column can have 0,1, up to the column number n.
This is because n in column n means n×n! but (n+1)×n! is (n+1)!, which "carries" to the next larger column heading (on the left).
To indicate numbers in this base, we list the "digits" and use the exclamation mark "!" as the base.
For larger number using ten or more columns, the value in a column may be 10 or more.
Here we will write the values (numbers) in each factorial column as a list of numbers enclosed in { } brackets.
6 = 1×3! = {1,0,0}!
7 = 1×3! + 1×1! = {1,0,1}!
8 = 1×3! + 1×2! = {1,1,0}!
9 = 1×3! + 1×2! + 1×1! = {1,1,1}!
10 = 1×3! + 2×2! = {1,2,0}!
23 = 3×3! + 2×2! + 1×1! = {3,2,1}!
24 = 1×4! = {1,0,0,0}!
Here are a few numbers in the factorial base:
n123456789101112
n!110112021100101110111120121200

Easy methods of converting to and from the Factorial Base

To convert a number to base !:

We can adapt the easy method to convert a number to base β above for converting a (decimal) number to base !. This time we start by dividing by 2 and the divisor increases each time.
Here is an example to convert 69 to base !:
69
34rem=1 69 = 2×34 + 1
11rem=134 = 3×11 + 1
2rem=311 = 4×2 + 3
0rem=22 = 5×0 + 2

To see the base ! representation, write down the reminders from the bottom to the top:
69 = 2311!
To convert a base ! representation to decimal:
  1. First write the multipliers above each digit by starting from the right with 2 and proceed leftwards with multipler 3 then 4 and so on. We will not use the leftmost 'multiplier' above the leftmost digit.
  2. Start with the most significant digit (the leftmost digit) which is the initial 'sum'
  3. Multiply the sum by the next column multiplier on its right ...
  4. ... and then add on the 'digit' under that multiplier to make the new 'sum'
  5. Repeat the previous two steps of multiplying and adding until you have a 'sum' under the final (rightmost) digit which is the decimal value
For example, using the same numbers as the example above, we have: To convert 2311! to decimal:
multiplier432
! digits2311
sum×
multiplier
83368
sum2113469
2311! = 69
Convert between base 10 and other integer bases in the Base Converter Calculator below.

Tests for Divisibility in Base Factorial

Every factorial from n! onwards is a multiple of n.
So to test if any given factorial base representation is divisible by n, we only need to test the number represented by the last n−1 'digits'.
For instance, to test if n! is divisible by 2=2!, look only at the last 'digit'. If it is 0, n is even, divisible by 2; if it is 1, the number is odd.

For divisibility by 3, test only the last 2 'digit's. if they are {1,1}! or {0,0}! then n is a multiple of 3; otherwise it is not.

There are 6 possibilities for the final 3 'digits' to test if a value is a multiple of 4. Show them

0 = {0,0,0}
4 = {0,2,0}
8 = {1,1,0}
12 = {2,0,0}
16 = {2,2,0}
20 = {3,1,0}
For larger divisors d, convert only the final d digits to base 10 and then test.
These divisibility tests are an advantage only if we are dealing with very large numbers and testing for small divisors.

An Application of Base Factorial

The 24 permutations of 4 objects can be listed in lexicographic order which means that the permutations, as numbers, are in numerical order or dictionary or alphabetic order. For instance here is the order of all the permutations of 1,2,3 and 4 sorted into lexicographic (dictionary) order - think of 1 as "a", 2 as "b", 3 as "c" and 4 as "d":
0:1,2,3,4
1:1,2,4,3
2:1,3,2,4
3:1,3,4,2
4:1,4,2,3
5:1,4,3,2
6:2,1,3,4
7:2,1,4,3
8:2,3,1,4
9:2,3,4,1
10:2,4,1,3
11:2,4,3,1
12:3,1,2,4
13:3,1,4,2
14:3,2,1,4
15:3,2,4,1
16:3,4,1,2
17:3,4,2,1
18:4,1,2,3
19:4,1,3,2
20:4,2,1,3
21:4,2,3,1
22:4,3,1,2
23:4,3,2,1
If we want to choose a random permutation or if we want to make sure we go through all permutations, we need a way of changing the number of the permutation in the lexicographic order to the permutation itself - and this is where base Factorial comes in.
The base factorial representation of an index number n in the lexicographic order is easily changed into the permutation itself.

Calculator for Factorial Base and Permutations

Factorials are exact up to 500! which has 1135 digits. After that, just the number of digits is shown.
You can use this Calculator to show that the number of digits in 1040! is itself a number with 42 digits(!)
Factorial Base and Permutations C A L C U L A T O R
Number of objects:
Perm. index #:
up to
Permutation:
()

!
R E S U L T S


 
calculator: Show maths Factorial Binomial Rational Multiples Base converter

You Do The Maths...

  1. Which is the first factorial that is a multiple of
    1. 8 ?
    2. 17 ?
    3. 100 ?
    8 = 23 and three 2s come from 2 and two more from 4, so 4! is the first multiple of 8
    17 is prime so if first appears in 17!
    100 = 22 52 and these factors will come from 2, 4, 5 and 10,so 10! is the first factorial to end with 00.
  2. 2! is the first even factorial,
    3! is the first facorial divisible by 3
    4! is the first divisible by 4 and
    5! the first that is a multiple of 5.
    Which is the first divisible by 6?
    And which factorials are the first divisible by 7, by 8, by 9 and by 10?
    3! is the first multiple of 6
    7 is a factor first in 7!
    8 in 4!
    9 in 6!
    10 in 5!
    The sequence of n where n! is the first divisible by 2,3,4,... is: 2,3,4,5,3,7,4,6,5, ... called the Kempner Numbers A002034
  3. How many zeroes are at the end of:
    1. 10!
    2. 20!
    3. 100!
    10! = 3628800 = 2×3×4×5×6×7×8×9×10
    ends with 2 zeroes
    20! = 2×...5×...10×...15×...20×
    ends with 4 zeroes
    100! has 5 as a prime factor from 5,10,15,20,...90,95,100 and more than enough even numbers to match these.
    Each of these 20 numbers has one 5 as a factor except 25, 50, 75 and 100 which have 5×5 as a factor:
    20 + 4 = 24 zeroes at the end

The Binomial Representation using Pascal's Triangle

Pascal's Triangle

Here is Pascal's Triangle but pushed over ("right justified") so that the 1s at the end of the rows are aligned in a column:
r
876543210n
10
111
1212
13313
146414
151010515
16152015616
1721353521717
...
Notation:
(n) = Binomial(n,r)
r

= n!
(n-r)! r!

= n(n−1)...(n−r+1)
r(r−1)...3×2×1
Here we will use Binomial(n,r) but other notations you will see in maths books are nCr, nCr or Cn,r and it is pronounced "n choose r".
Each element in Pascal's Triangle above, which is right-justified here, is the sum of the element above and the element to the right of that one where blank entries mean "0". The right-hand elements in column 0 are always 1:
(n) = (n − 1) + (n − 1) , n ≥ r > 0
rrr − 1

(n) = 1; (n) = 0 otherwise
0r
Pascal's Triangle has many interesting properties including coefficients of certain polynomials and has many applications including probabilites.

The Binomial Representation

To use Pascal's triangle as an integer representation method, you first decide on the number columns you want to use.
Then you choose one element from each column and, to make the representation unique, as we go to a smaller column number, we must choose an entry on a smaller row number too.
For instance, here is 6 represented using from 1 to 4 columns. Remember that we must choose a row for each of the N columns and as N decreases so must the row numbers chosen. The Binomial representation is then just the list of row numbers in order of the column numbers going down from N to 1.
All numbers in a Binomial base have their digits in decreasing order.

Notation

Once we have decided on the number of columns to use, it is the row number for each of the columns that forms the representation as a list of the row numbers.
To conform to the way other bases are shown, we list the row numbers from the largest down to 1.
Because the representation will change depending on how many columns we use, the Base is shown as Binom.
The two examples above are therefore
15 = Binomial(6,4) + Binomial(2,3) + Binomial(1,2) + Binomial(0,1)
= {6,2,1,0}Bin4

To represent 6
using 1 column
Binomial(6,1) is 6 so it's representation is {6}
using 2 columns
Binomial(4,2) + Binomial(0,1) = 6 + 0 = 6 which is {4,0}
using 3 columns
Binomial(4,3) + Binomial(2,2) + Binomial(1,1) = 4 + 1 + 1 = 6 which is {4,2,1}
using 4 columns
Binomial(5,4) + Binomial(3,3) + Binomial(1,2) + Binomial(0,1) = 5 + 1 + 0 + 0 = 6 which is {5,3,1,0}
Note that all the column numbers are used in order and that the row numbers decrease as the column numbers decrease.

To convert from Binomial representation to its base 10 value:

Insert the column numbers after each 'digit', numbering the columns from 1 on the right going right to left
and then interpret these are arguments to the Binomial function: for example:
{5,3,1,0}Binom has its numbers paired with columns in order, ending with column 1:
row n5310
col r4321
(n
r
)5100
We then interpret this as Binomial(5,4) + Binomial(3,3) + Binomial(1,2) + Binomial(0,1)
which evaluates to 5 + 1 + 0 + 0 = 6 = {5,3,1,0}Binom

Calculator for Binomial Representations

Binomial Representations C A L C U L A T O R
( n:
r:
)
10
up to: 10

with columns
{} Binom
R E S U L T S


 
calculator: Show maths Factorial Binomial Rational Multiples Base converter

You Do The Maths...

  1. What are these numbers in base 10:
    1. {1}Binom
    2. {2,1}Binom
    3. {3,2,1}Binom
    4. {4,3,2,1}Binom
    What is {N, N-1, ... , 3, 2, 1}Binom using N columns?
    {1} = Binomial[1,1] = 1
    {2,1} = Binomial[2,2] + Binomial[1,1] = 2
    {3,2,1} = Binomial[3,3] + Binomial[2,2] + Binomial[1,1] = 1+1+1 = 3
    {4,3,2,1} = 4
    {N, N-1, ... , 3, 2, 1}Binom = N
  2. What is {n, 0}Binom in base 10?
    {n,0} = Binomial[n,2] + Binomial[0,1] = n(n-1)/2
  3. Using your answer to Question 1 what is {N+2, N+1, N-2, N-3, ... , 2, 1, 0}Binom where the 'digits' from N-2 descend by 1 to end at 0?
    {N+2, N+1, N-2, N-3, ... , 2, 1, 0} = 2 N − 1
    1. Express 100=102 in base 9.
    2. What is 103 in base 9?
    3. How do your answers above relate to the Binomial Theorem? Think about 10 as 9+1 and raise it to a power.
    4. What is 103 in base 3 and how does it relate to its representation in base 9 above?
    102 = (9 + 1)2 = 1×92 + 2×9 + 1 = 1219 = 81+18+1 = 100
    103 = (9 + 1)3 = (9+1)3 = 13319
    In general, the 'digits' of (β+1)n are the binomial coefficients of row n provided they are no bigger than the base.
    Base β is related to base β2 by writing each base β2 'digit' as two base β digits:
    1000 = 13319 = 01 10 10 01 3 because 39 = 103

References

The Fibonacci Bases

The Fibonacci Numbers are 0, 1, 1, 2, 3, 5, 8, 13, 21, ... where each is the sum of the previous two numbers.
It has many interesting mathematical properties.
As a number base system, we again label the columns in order increasing to the left: 8, 5, 3, 2, 1 with just a single column labelled 1 so we only use Fibonacci numbers with index 2 or more because Fib(2)=1, Fib(3)= 2, ... .
Every whole number is a sum of different Fibonacci Numbers
that is: no Fibonacci number is used more than once.
We can use F to signify a representation using the Fibonacci numbers as column headers.

3 = 100F = 2 + 1 = 11F
Since Fib(1) is 1, we can simply use n Fib(1)'s to sum to any given number n.
But there is always another way to do this too for numbers bigger than 1.
If we write the number of times we need Fib(i) is column i, we have a number representation whose 'digits' are any whole number.

The minimal and maximal Fibonacci bases

However, if we restrict ourselves to using each Fibonacci number at most once then it is again possible to find a collection which is a set of Fibonacci numbers whose sum is any given integer.
Two in particular are of interest: since Fib(n-1) + Fib(n-2) = Fib(n), then any two consecutive 1s in a Fibonacci representation can be "carried" into the next column. It turns out it is always possible to find a set of Fibonacci numbers with any given sum that For example:
8 = Fib(6)
= 10000Fmin
= Fib(5) + Fib(4) = 1100F
= Fib(5) + Fib(3) + Fib(2) =1011Fmax
Here are all the Fibonacci representation for numbers 1 to 9:
Columns in the Fibonacci base as ..., 8, 5, 3, 2, 1
Columns in the Fibonacci base are ...8 5 3 2 1
n0123456789
nF0110100
11
1011000
110
1001
111
101010000
1100
1011
10001
1101

To convert to and from Fibonacci bases

There are usually many Fibonacci representations of a number using only the 'digits' 0 and 1 in every column and with the column headings (in reverse order): 1, 2, 3, 5, 8, 13, 21, ...
An easy method of converting a number to base F will give the Zeckendorf represenation which happens to have the least number of 1s of all the Fibonacci representations for the number.
  1. Write the Fibonacci column headings, in order from right to left, from 1,2,... until the column headings are larger than the value to be converted
  2. Start by writing the decimal number to be converted above the leftmost column header
  3. If the column heading is larger than the number, put a 0 in that column and keep the number to use on the next column on the right; Otherwise if the column header is not larger than the number, Subtract the column header value from the number and write a 1 in the column
  4. repeat the step above on the same or, if you did the subtraction, the reduced decimal number
  5. after the rightmost column you should have reached 0.
    If not, you've made a mistake so check your previous steps!
  6. Each column now has a 0 or 1 in it which is the Fibonacci (Zeckendorf) representation
For instance here we convert 52 to base Fibonacci:
A: number52 18 18 5 5 00 0
B: Fib col
heading
55 34 21 13  8  5  3  2  1
A-B
or A
18 18 5 5 0 0 0 0
F digit1 0 1 0 1 0 00
52 = 10101000F
To convert from base Fibonacci, write the column headings above the digits and add the column headings above each digit '1'.
Convert between base 10 and other integer bases in the Base Converter Calculator below.

References

Negatively indexed Fibonacci numbers

We can also express every number positive and negative using only the negative indices in the Fibonacci series:
i...-8-7-6-5-4-3-2-1012345678...
Fib(i)...-2113-85-32-1101123581321...
The advantage is that now both positive and negative whole numbers can be represented without an initial sign.
We have to choose between using the negative indices to the right of a radix point (as in integer-valued bases) but this has the disadvantage of having 0 digits at the front. Since the values are not fractional, listing them with indices running from -1:...-4, -3, -2, -1 makes more sense.
For instance:
i-5-4-3-2-1
F(i)5-32-11
00
11
2100
3101
410010
510000
610001
710100
810101
i-6-5-4-3-2-1
F(i)-85-32-11
00
-110
-21001
-31000
-41010
-5100101
-6100100
-7100001
-8100000
We add up the Fibonacci numbers for each column with 'digit' 1 in it to find the value represented.
Every integer value now has a representation using 0s and 1s only.
The above are minimal forms, using the least number of 1-digits. As for minimal Fibonacci representations (Zeckendorf representations) there will never be two 1's next to each other.
Here we call this the Fibonacci Negative Index representation or Fibnegi or just Fnegi.
2 = 100Fnegi
−3 = 1000Fnegi
Convert between base 10 and other integer bases in the Base Converter Calculator below.

How can we tell if a base Fibnegi value is negative?

The positive numbers start with an index of a positive Fibonacci number so are of even length. The negative numbers start with an index of a negative Fibonacci number so are of odd length.

References

Negative Bases

If we use base -10 (negadecimal) then the columns have the values
Column:...43210
Meaning...(-10)4(-10)3(-10)2(-10)1(-10)0
Value...10000–1000100–101
At first it may not be obvious that
Every whole number value, both positive and negative, has a unique representation in base -10 (negadecimal) using just the digits 0 to 9 without needing a sign
For instance, What about 1000? Try these practice problems:

You Do The Maths...

Check your answers with the calculator below
  1. Write these numbers in base −10:
    110, 119, 150, 199, 301, 399, 901, 999
    110 = 290-10
    119 = 299-10
    150 = 250-10
    199 = 219-10
    301 = 301-10
    399 = 419-10
    901 = 901-10
    999 = 19019-10
  2. How are these (base 10) values represented in base −10:
    −1, −2, −9, −10, −11, −100, −1000
    −1 = 1-10
    −2 = 2-10
    −9 = 9-10
    −10 = 190-10
    −11 = 191-10
    −100 = 100-10
    −1000 = 19000-10
  3. What are the first 5 column headers in base −2?
    Make a table of numbers from −12 to 12 in base −2.
    n10-12-11-10-9-8-7-6-5-4-3-2-10123456789101112
    n-21101001101011010101110001001111011111100110110110111011110010111010110111100011001111101111111100

Using Negative Digits

Instead of using powers of a negative number base to represent negative values, we could use negative 'digits'.
This is not as peculiar as it might at first sound. For example if the time is 2:50 then we can say "10 to 3" which is in effect using the hour 3 and the minutes -10.
We can also see times expressed with minutes from 0 to 59 (base 60) on timetables for example and, when saying the time the minutes go from -29 to +30 as in "29 minutes to 3" up to "3:30".
So for base β instead of using 'digits' 0 to β−1 we could just as effectively use 'digits' −β/2 to +β/2 if β is odd (rounding the halves). To make sure representations are unique when β is even we can decide to use either −β/2 or else β/2 but not both 'digits'.
Using two symbols for negative digits is inconvenient (is 3−5 two 'digits' or a subtraction?) so to use one single 'digit' per column and we can show negative 'digits' with the negative sign placed above the 'digit' so that −3 is written as 3.
Base −10 using the 10 'digits': 4, 3, 2, 1, 0, 1, 2, 3, 4, 5
0 .. 5: 0−10 .. 5−10
6 .. 15: 14−10 .. 15−10
16 .. 25: 24−20 .. 25−20
 
−4 .. −1: 4−10 .. 1−10
−16 .. −5: 24−10 .. 15−10
−26 .. −15: 34−10 .. 25−10
123 is 100 −20 −3 = 77
It is easy to see if a value represented using nega-digits is positive or negative. Show how
If the leftmost 'digit' is positive, the value is positive;
if the leftmost 'digit' is negative, the value is negative
Also, it is easy to negate a value in this system. Show how
Change each positive 'digit' into its negative form and vice-versa.

You Do The Maths...

Check your answers with the Calculator below
  1. How are -9 to 9 represented in base 3 using the 'digits' 1, 0 and 1?
    n0123456789
    n301111011111110111101100
    m0-1-2-3-4-5-6-7-8-9
    m31111011111110111101100

Rational Bases

Up to now we have only used whole numbers as our 'column headings' but always our 'digits' are whole numbers.
It is also possible to use column headings based on powers (a radix system) where the base is not a whole number but a fraction (a rational) such as 3/2.
In base β, an integer, when we add 1 to the units column and find a value equal to β, we take β away and add one (the carry) to the next column to the left.
In a rational base, say 3/2, when we find 3 in any column we carry 2 to the column to the left. This applies to any rational number β bigger than 1.

The 'digits' in rational base m/n are the digits of base m, namely 0 to m−1.

1 = 13/2 15/2 15/3
2 = 23/2 25/2 25/3
3 = 203/2 35/2 35/3
4 = 213/2 45/2 45/3
5 = 223/2 205/2 305/3
6 = 2103/2 215/2 315/3
7 = 2113/2 225/2 325/3
8 = 2123/2 235/2 335/3
9 = 21003/2 245/2 345/3
10 = 21013/2 405/2 3105/3
11 = 21023/2 415/2 3115/3
12 = 21203/2 425/2 3125/3
Here is how we convert 11 to base 3/2:
Divide 11 by 3 noting the remainder: 3 ) 11 3 rem =2
Multiply the quotient 3 by 2
and divide it by 3
3 ) 6 2 rem=0
Multiply the quotient 2 by 2
and divide it by 3
3 ) 4 1 rem=1
Multiply the quotient 1 by 2
and divide it by 3
3 ) 2 0 rem=2
STOP because
The number is now 0 and
all future 'digits' will be 0.
Read the remainders from bottom to top
21023/2

Rational bases less than 1

For rational base β less than 1 (but bigger than 0), reverse the digits of the representation in base β remembering to include the radix point. This is because (β)n is (1/β)−n. For example:
12 = 3125/3 = 2.133/5

Use the Base Converter below to convert to and from rational bases.

Rational base 2 is not the same as base 4/2

Since our algorithm to convert a number to a rational base produces 'digits' of the numerator as a base, then base 4/2 produces a number using 'digits from the list 0,1,2,3 whereas base 2/1 produces 'digits' from 0,1!
For example 23 in base 2 is 101112.
All conversions to base 2 are the same as to base 2/1.
But in base 4/2 we have 20234/2 = 2310.

In base p/q where p/q is in its lowest terms
every number has a unique representation.

But in a non-simple rational base p/q = b (a base that is a fraction p/q but p is a multiple of q and so the base simplifies to the number b)
every number from b upwards has more than one representation

In base 4/2, the column's values are 8=(4/2)3, 4=(4/2)2, 2=4/2 and 1
and the digits we can use are 0, 1, 2, 3. So 8 has 5 representations:
8 = 3×2 + 2×1 = 324/2
8 = 1×4 + 1×2 + 2×1 = 1124/2
8 = 1×4 + 2×2 = 1204/2
8 = 2×4 = 2004/2
8 = 1×8 = 10004/2

What we have found are the 5 partitions of 8 into bags (or multisets which are sets but with repetition allowed) using the parts 1,2,4 and 8 (columns of base 2, the base in simplest terms) up to a maximum of 3 times each (the remainders mod 4, the base numerator).

Rational Multiples Calculator

Rational Multi-Representation C A L C U L A T O R

representations of in base

R E S U L T S



 
calculator: Show maths Factorial Binomial Rational Multiples Base converter

Negative Rational Bases

We ensure that all remainders are non-negative when dividing a number by the numerator. The denominator of the base is always a positive number.
For example, 5 ÷ −4 is −1 with remainder 1
because 5 = −1×−4 + 1.
RULE: For all b>0 : a ÷ b = q remainder r
if and only if
a = b×q + r and 0≤r<b
Here is how we convert 5 to base −4/3 by repeatedly dividing by −4 (the numerator), noting the reminder, then multiplying the quotient by 3 (the denominator):
Divide 5 by −4 noting the remainder: −4 ) 5 −1 rem =1
Multiply the quotient −1 by 3
and divide it by −4
−4 ) −3 1 rem =1
Multiply the quotient 1 by 3
and divide it by −4
−4 ) 3 0 rem =3
STOP because
The number is now 0 and
all future remainders will be 0.
Read the remainders from bottom to top
311−4/3

You Do The Maths...

Check your answers with the Calculator in the next section
  1. Convert 23 to bases 6/3 and 8/4.
    Can you find a method of converting easily from one fractional base to another whose fraction simplifies to the same number?
  2. Make a table of the number of base 3/2 digits in numbers 0 to 30.
    Find a recursive formula for the digit count.
    Length is 1 for numbers 0,1,2,3
    Length is 2 = length(23/2)+1 for 3,4,5
    Length is 3 = length(43/2)+1 for 6,7,8
    Length of n is length(2 Floor(n/3))+1 in general.
    See A246435
    The number of digits in number n is one more than the number in 2 Floor(n/3) where for positive n, the Floor(n) is n if n is an integer otherwise it is the nearest integer below n. For positive n, the floor of n is just n without any digits after its decimal point.
  3. How many numbers have 1 digit in base 3/2? How many have exactly 2 digits? and length 3?
    3 have 1 digit,
    3 have 2,
    3 have 3 ....
    The list of the counts is 3, 3, 3, 6, 9, 12, 18, 27, 42, ... A081848
    1. Which is the smallest number to have exactly n base 3/2 digits?
      The smallest with 1 digit is 1 (or 0), with 2 digits is 3, with 3 digits in 6: 1, 3, 6, 9, 15, 24, 36, ... A070885
    2. Which is the smallest even number to have n digits in base 3/2?
      List them with their base 3/2 representations.
      What simple pattern connects the representations?
      The base 3/2 reps are
      n3/2n
      22
      214
      2106
      210110
      2101116
      21011024
      210110036
      2101100054
      They all are extensions of the previous one, adding one new digit on the right A303500. The infinite sequence of which each representation is an initial part is A304273 and the numbers themselves are in A305498
    3. What is the largest even number that can be written with n base 3/2 digits?
      What are their base 3/2 representations?
      nn3/2
      22
      421
      8212
      142122
      2221221
      34212211
      522122111
      8021221112
      The numbers are A305497
      their representations are A304272
      Each is a single digit extension of the previous one so that their representations are the initial part of the infinite sequence 21221112212112212121.... A304274

Irrational Bases

But what if our column headings were powers not of a whole number nor even a rational number but of an irrational number, such as Phi, the golden section number or √2? This also works with some conditions on the base and the 'digits'.
Let's look at an example: base Phi.

Base Phi (Φ)

Phi is 1.6180339... = (√5 + 1)/2, one of the golden section numbers (or golden mean or golden ratio).
Because Phi2 = Phi + 1 is a definition of the positive number Phi, we only need "digits" 0 and 1, called "phigits"!
The basic Phi rule is Phin+2 = Phin+1 + Phin
All natural numbers are representable in Base Phi using phigits 0 and 1 but we will need negative powers and so we need a "base point" to act like the decimal point in base 10.
Just as for Fibonacci base numbers above, we need never have two 1's next to each other since they combine to give a 1 in the next column to the left using the basic Phi rule.
Here Phi = 1·6180339... = phi–1
and its reciprocal (1/Phi) is also the same as Phi−1 which onthese pages is denotes by the lower case Phi:
phi = 0·6180339... = Phi – 1 = 1/Phi = Phi–1
Column Phi
power
phi
power
A + B Phi C + D phireal
value
5Phi5 phi-5 3 + 5 Phi 8 + 5 phi 11·0901699..
4Phi4 phi-4 2 + 3 Phi 5 + 3 phi 6·8541019..
5Phi3 phi-3 1 + 2 Phi 3 + 2 phi 4·2360679..
2Phi2 phi-2 1 + 1 Phi 2 + 1 phi 2·6180339..
1Phi1 phi-1 0 + 1 Phi 1 + 1 phi 1·6180339..
0Phi0 phi0 1 + 0 Phi 1 + 0 phi 1·0000000..
-1Phi-1 phi1 -1 + 1 Phi 0 + 1 phi 0·6180339..
-2Phi-2 phi2 2 - 1 Phi 1 - 1 phi 0·3819660..
-3Phi-3 phi3 -3 + 2 Phi -1 + 2 phi 0·2360679..
-4Phi-4 phi4 5 - 3 Phi 2 - 3 phi 0·1458980..
-5Phi-5 phi5 -8 + 5 Phi -3 + 5 phi 0·0901699..
For instance 2 is 10.01Phi = Phi + Phi−2.
1..13 in base Phi
11.Phi
210.01Phi
3100.01Phi
4101.01Phi
51000.1001Phi
61010.0001Phi
710000.0001Phi
810001.0001Phi
910010.0101Phi
1010100.0101Phi
1110101.0101Phi
12100000.101001Phi
13100010.001001Phi
phi = 1/Phi = Phi-1 = 0.6180339... = (√5-1)/2 and the basic phi rule is phin = phin+1 + phin+2.
Because Phi−n = phin it is easy to see that to get a base phi representation from a base Phi representation we merely reverse the order of the phigits including the base point:
4 = 101.01Phi = 10.101phi
6 = 1010.0001Phi = 1000.0101phi
Convert between base 10 and other integer bases in the Base Converter Calculator below.

More on base Phi

Other irrational bases: √n

If n is not a square number then
the even powers of √n are just the powers of n
so base √n includes all the representations of integers in base n.
Columns for base β=√n
...β6β5 β4β3 β2β1 β0
...n3n5/2n2n3/2nn1/21
...n3n2√nn2n√nn√n1
The table shows that
Because √n is itself not rational, the only values we can represent in base √n are A + B√n for whole numbers A and B.

If we represent A in base n then shifting all its digits one place left will effectively multiply the value by the base, √n.

So any value A √n will only have non-zero digits in the even-power places and always 0s in the odd-power places. We can then 'add' in the digits of B in base √n which similarly will have only 0s in the odd-power places and any non-zero digits will be in even-powered places.

Base Converter

About the Base Converter Calculator:

Base Converter Calculator

Base Converter C A L C U L A T O R
base
base 10
of numbers
from 10
to 10
converted to these bases:

R E S U L T S



 
calculator: Show maths Factorial Binomial Rational Multiples Base converter

Complex Number Bases

We can again extend our possible bases in a radix system to complex numbers and find a unique way to represent integers and other values.
...... more coming soon .....

The Chinese Remainder Theorem Number Base

This is a system which has many computational advantages, particularly for computers than can run parts of a computation in parallel since each digit is independent of the others.
.... coming soon .....

Monotonic numbers

For a given base, Monotonic numbers are those that have all their base digits already in order, from least to most, left to right, repetitions allowed. This is sometimes called non-decreasing order since any digit must be less than or equal to the digit in the next column to its right. In this section we investigate the monotonic numbers in base 10:
13 and 112 are monotonic but 103 and 192 are not monotonic.

You Do The Maths...

    1. How many monotonic numbers are there between 1 and 10?
    2. How many monotonic numbers are there between 10 and 100?
    3. How many monotonic numbers are there between 100 and 1000?
    1. 1 to 9: all are monotonic: 9 numbers
    2. 10 to 100:
      11 to 19, 22 to 29, ... 88-89, 99: 9+8+7+6+5+4+3+2+1 = 45 numbers
    3. 100 to 1000:
      100-200: 111, 112, ... 119, 122, 123, ... 129, ... 189, 199: 9+8+7+6+5+4+3+2+1 = 45 numbers
      200-300: 222, 223, ..., 229, 233...239, 244..249, ... 288-289, 299: 8+7+...+2+1 = 36 numbers
      then 28, 21, 15, 10, 6, 3 , 1
      Total is 45+36+28+21+15+10+6+3+1 = 165
    The sequence is 9, 45, 165, 495, 1287 ... A000581
  1. 13 and 112 are monotonic numbers. 13×112= 1456 is also monotonic
    33366667 and 333667 are monotonic and their product is 33366667×333667 = 11122255677889, also monotonic.
    Here is a method to find pairs of monotonic numbers whose product is also monotonic:
    1. Take two numbers each of which is
      either composed solely of 3s
      or has only 6s as its digits
      or with some 3s first followed by some 6s.
    2. Add 1 to both numbers
    3. Their product is monotonic
    For example: (33+1)×(366+1) = 34×367 = 12478 a monotonic number.
    Can you find any other examples that are not formed by the above method?
    All are of the form above: see the Blecksmith and Nicol Reference below
    1. 5, 35 and 335 are monotonic.
      Their squares are also monotonic: 52 = 25, 352 = 1225 and 3352 = 112225.
      What is the pattern here?
      5, 35, 335, 3335, 33335, ... any number of 3s followed by 5
      have a monotonic square.
      See the Blecksmith and Nicol Reference below
    2. There is another similar pattern starting with 17. What is it?
      17, 167, 1667, 16667, ... begin with 1, then any number of 6s and a final 7
      all have monotonic squares.
      This series and the one above are the only numbers with monotonic squares.
      See the Blecksmith and Nicol Reference below

References

Links and References


Valid HTML 4.01! © from 2017 Dr Ron Knott ronknott at mac dot com    Dr Knott's Maths Fibonacci and HOME page
created: 20 August 2017