How can Fourier division be used to divide large numbers without a calculator?

  • Context: Undergrad 
  • Thread starter Thread starter Stratosphere
  • Start date Start date
  • Tags Tags
    Large numbers Numbers
Join the discussion
Ask a follow-up here, or get your own question answered by working scientists, mathematicians and engineers — people, not an autocomplete.
Real named experts · corrections over time · the nuance an AI answer skips
7 replies · 29K views
Stratosphere
Messages
373
Reaction score
0
How would you Divide very large numbers without using a calculator?
EX. [tex]\frac{125000}{299000000}[/tex]
 
Mathematics news on Phys.org
Long ago, before calculators, logarithms were used and invented for this purpose. You'd divide by subtracting logarithms and antilog the result to get the answer.
 
One should usually first take out the obvious powers of ten, then factorize.

e.g.

[tex]\frac{125000}{299000000} = \frac{125}{299000}=\frac{5^3}{299\cdot 10^3} = \frac{5^3}{299\cdot (2\cdot 5)^3} = \frac{1}{299\cdot 2^3}[/tex]

And [tex]299\cdot 8 = 3 \cdot 10^2 \cdot 8 - 8 = 24 \cdot 10^2 - 8 = 2400 - 8 = 2392[/tex],

so that

[tex]\frac{125000}{299000000} = \frac{1}{2392}[/tex]

Which by hand is good enough for me.

(This might be wrong tho, it is kinda late here)
 
"How would you Divide very large numbers without using a calculator? "

Long division is a correct algorithm. Are you asking whether or not there exists a faster way?
 
csprof2000 said:
"How would you Divide very large numbers without using a calculator? "

Long division is a correct algorithm. Are you asking whether or not there exists a faster way?

Yes I am asking for a faster way.
 
Stratosphere said:
without using a calculator?

Slide rule?
 
You could use Newton-Raphson. Computing x = 1/y for given y amounts to solving the equation:

1/x - y = 0

Then, Newton-Raphson yields the following recursion for the nth approximation


x_{n+1} = x_n - (1/x_n - y)/(-1/x_n^2) =

x_n +x_n -y x_n^2 =

2 x_n - y x_n^2

The iteration doesn't involve any divisions, so it is a true division algorithm. The number of correct digits doubles after each iteration, while with long division you only get one decimal at a time, so it is much faster than long division.