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

Wolframalpha it says the limit does not exist

  1. Aug 29, 2010 #1
    1. The problem statement, all variables and given/known data

    use definition for the limit of a function to show that
    [tex]\lim_{(x,y)\rightarrow (0,0)} xcos\frac{1}{y}[/tex]

    2. Relevant equations


    3. The attempt at a solution

    fisrt, i assumed it the limit is 0(i dont know if it is true), but i showed it, but something bugging me, i put in wolframalpha it says the limit does not exist, http://www.wolframalpha.com/input/?i=lim+xcos(1/y),(x,y)->(0,0)

    so who is wrong here?
    Last edited: Aug 29, 2010
  2. jcsd
  3. Aug 29, 2010 #2


    User Avatar
    Homework Helper

    Re: limit

    at a quicklook
    |cos(1/y)|<=1 for all y's,

    so you can choose x small enough to bring you close enough to zero. The function will oscillate with increasing frequency, but with decreasing magnitude as you approach the origin
  4. Aug 29, 2010 #3
    Re: limit

    so, it really approaches to 0, but how come wolfram-alpha say it does not exist. did i wrote anything wrong?
  5. Aug 29, 2010 #4
    Re: limit

    A limit of a function of several variables fails to exist if there are different limits from "different direction" i.e. from the "x" and "y" directions:

    Thus, you need to check if the limit as x -> 0 (treat y as a constant) of the function is the same as the limit as y -> 0 (treat x as a constant).

    Let's check limit from x direction:

    lim[x->0] x*cos(1/y) = 0*cos(const.) = 0

    But there is a problem with the limit of cos(1/y) when y->0.


    do you know why there is a problem with that y->0 limit? :)

    Good Luck.
  6. Aug 29, 2010 #5


    User Avatar
    Homework Helper

    Re: limit

    yeah its an interesting one...

    cos(1/y) is cleary undefined anywhere on the line y=0, so no limit on that line will exist for any function of cos(1/y)

    however, it does have the property that as you get close to x=0, |x||cos(1/y)|<=|x|which tends to zero

    also just to add, checking 2 startight line directions is suffcient to show a limit dne, but not to prove it exists
    Last edited: Aug 29, 2010
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook