I am trying to find

$$\lim\limits_{x \to \infty} n-\sqrt{n^2-4n} $$ using the definition of a limit.

I have tried to multiply top and bottom by $n+\sqrt{n^2-4n} $ giving $\frac{4n}{n+\sqrt{n^2-4n}} $ but this does not seem to help too much. Any hints would be much appreciated. Thank you.

  • $\begingroup$ Hint: $$0\leqslant\left(n-\sqrt{n^2-4n}\right)-2=\frac4{n-2+\sqrt{n^2-4n}}\leqslant{}{}{}{}\frac{4}{n-2}\leqslant\frac8{n}\quad (n\geqslant4)$$ But this approach will require to be able to prove that $8/n\to0$... $\endgroup$ – Did Feb 7 '16 at 7:34

You are right till there. Now divide both the numerator and denominator by 'n'. You get 4 / (1 + sqrt(1-4/n)) . as n tends to infinity, that gives 4/1+1 = 2;

  • $\begingroup$ I am trying to use the definition of the limit and not assume that 4/n tends to 0 as n goes to infinity. $\endgroup$ – IntegrateThis Feb 7 '16 at 1:36
  • $\begingroup$ @JamesDickens "I am trying to ... not assume that 4/n tends to 0" Wait, are you serious? $\endgroup$ – Did Feb 7 '16 at 7:14
  • $\begingroup$ @Did I think he wants to do an epsilon-delta proof. $\endgroup$ – Akiva Weinberger Feb 7 '16 at 7:23
  • $\begingroup$ @AkivaWeinberger ...Which, sooner or later, will need to prove or to admit that 1/n converges to 0. $\endgroup$ – Did Feb 7 '16 at 7:31
  • 1
    $\begingroup$ @Did Yeah. He doesn't want to assume it tends to zero, he wants to prove it tends to zero. (And he doesn't want to assume that $\lim f/g=\lim f/\lim g$ either, I bet.) $\endgroup$ – Akiva Weinberger Feb 7 '16 at 15:39

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.