Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Exact number to pi

  1. Nov 10, 2005 #1

    does anyone know the exact number to pi (yes the one over 1000 characters long:P) i need it a.s.a.p
  2. jcsd
  3. Nov 10, 2005 #2
    That is only an approximation. You can find certain amounts of digits on Google, but they aren't exact.
  4. Nov 10, 2005 #3


    User Avatar
    Homework Helper

    I dunno what's that for... But anyway, here it is:
    Pi to 1,000,000 places.
  5. Nov 10, 2005 #4


    User Avatar
    Science Advisor
    Homework Helper

    You can't write it fully in decimal form as it is irrational.
  6. Nov 10, 2005 #5
    You can calculate pi's decimal expansion using the algorithms here.
  7. Nov 10, 2005 #6
    Why do you need pi to that accuracy, and why have people got computers working out pi to that number of decimal places? What is the use, I don't get it...:confused:
  8. Nov 10, 2005 #7
    pi approximations

    :bugeye: There is a mathematician named Borwien who has come up with some neat algorithms. The best one in terms of speed of convergence vs. complexity
    converge quartically, i.e. the number of correct digit roughly quadruples every iteration, furthermore the algorithms only involve square roots and fourth roots, nothing as complicated as Ramujan's stuff. Using this algorithm a mere 15 iterations gives something like a billion digits. His derivation of it is an interesting mix of elliptic intergrals and the arithmetic-geometric mean of a number. What takes computers so long these days is just the mechanics of doing arithmatic on numbers with a billion digits. Look up Borwein if you are interested,
  9. Nov 11, 2005 #8

    matt grime

    User Avatar
    Science Advisor
    Homework Helper

    Conjecture: pi is normal. Normal roughly means that the digits of pi behave randomly in some sense. We can't prove this, but knowing pi to many places has provided strong supporting evidence for it to be true.

    Further, having a known (transcendetal) constant is useful for benchmarking alogrithms (ie how fast they converge)
  10. Nov 11, 2005 #9
    I remember trying to figure out a formula for calculating pi a few months ago and came up with: 2x(sin(90/x)) where a very big value of x gets a close approximation for pi.
    My calculator could only evaluate it to 10 digits though...the other millions of digits where given in standard form
  11. Nov 11, 2005 #10
    Taylor expand sin(90°/x) = sin([itex]\frac{\pi}{2x}[/itex]) around x=0, and see what you get. :)
    Last edited: Nov 11, 2005
  12. Nov 12, 2005 #11
    Can anybody tell me how to plot pi on anumber line?
  13. Nov 12, 2005 #12
    Gonna be a while before I can do that :redface:
  14. Nov 12, 2005 #13


    User Avatar
    Science Advisor

    Sure. Just place a dot at about 3.14 and then label it [tex]\pi[/tex].
  15. Nov 12, 2005 #14


    User Avatar
    Science Advisor

    What do you mean by "plot pi"?

    If you mean just mark a point on a line, pi is approximately 3.14. Marking that point should be sufficient. If you mean "construct a line segment, using compass and straight edge, having length pi, given a line segment of length 1", it is impossible. pi is transcendental and so not constructible.
  16. Nov 12, 2005 #15
  17. Nov 12, 2005 #16
    Ok, well basically, using calculus, you can show that

    [tex]\sin{x} = x - \frac{x^3}{3!} + \frac{x^5}{5!} - \frac{x^7}{7!} + ...[/tex]

    where x is in radians. (You can easily verify this for yourself on a graphing calculator, if you have one.) This infinite series converges for all x, but it is centered around zero, so it converges for low numbers "before" it converges for high ones. In fact, for numbers that are infinitely close to zero, you can just drop the higher order terms and say that [itex]\sin(x) = x[/itex]. You were taking x to be infinitely high, so [itex]\frac{\pi}{2x}[/itex] would be very low, and you can say that [itex]2x\sin(\frac{\pi}{2x}) = 2x(\frac{\pi}{2x}) = 2(\frac{\pi}{2}) = \pi[/itex]. Nice, huh?

    (Incidentally, I arrived at a formula very similar to yours when I was a sophomore in high school, and knew geometry but not calculus. Would I be correct in guessing that you arrived at it by calculating the circumference of a regular polygon with an infinite number of sides?)
  18. Nov 13, 2005 #17
    There is no exact number. It is Irrational. It's sum{Pi^2/2}=1/n^2 If I'm not mistaking...and imagine how many natural numbers there are...
  19. Nov 13, 2005 #18


    User Avatar
    Homework Helper

    I think you're driving at zeta (2).
    [tex]\zeta(2) = \Sigma_{i=1}^{\infty}\frac{1}{i^2} = \frac{\pi^2}{6}[/tex]
    The proof of the irrationality (or indeed transcendence) of pi does not depend on that equality, though.
  20. Nov 14, 2005 #19
    I wonder...why do you need it? are you trying to crack it? If you succeed let me know!
  21. Nov 18, 2005 #20
    By that, I meant that can u plot it by any other method. And I have also got the answer now. It is constructible by some other apparatus.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook