Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Question: Give the set of points of discontinuity of $f(x)= \lfloor x \rfloor$.

Answer: When $x=0.9999, y=0$ but when $x=1, y=1$, so for a very small change in the value of $x$, the change in the value of $y$ is big and this happens at all the integers (positive as well as negative), therefore the set of all the integers is the answer. Am I correct?

share|cite|improve this question
Yes, you are correct. – Josué Tonelli-Cueto Oct 8 '11 at 12:29
Formally, you could use the definition of continuity, say, with epsilon = 1/2 – The Chaz 2.0 Oct 8 '11 at 12:33
Yes Its true for $x\in \mathbb{Z}$ – Ramana Venkata Oct 8 '11 at 12:42
Rigorously speaking, the values of f at x=0.9999 and at x=1 do not prove nor disprove the continuity of f at x=1. Hence I assume that 0.9999 stands for any value less than 1 and close to 1... in which case your idea is correct but cannot be called a proof. Yet. – Did Oct 8 '11 at 12:44
You have a clear geometric idea of what's going on. Depending on the type of course, you may be expected to write it up semi-formally (as $x$ approaches an integer $n$ from the left, $\lfloor x\rfloor$ does not approach $\lfloor n \rfloor$) or formally ($\epsilon$-$\delta$). Don't forget that you also need to show that your list of points of discontinuity is complete, that is, that if $a$ is not an integer, then the function $\lfloor x\rfloor$ is continuous at $a$. – André Nicolas Oct 8 '11 at 14:21

You can use limits/$\epsilon$-$\delta$ to show that if $a\in\mathbb{Z}$, then $f(x)=\lfloor x\rfloor$ is not continuous at $a$; and that if $a\notin\mathbb{Z}$, then $f(x)$ is continuous at $a$. So the points of discontinuity are exactly the integers.

Arguing first somewhat informally with limits:

Let $a\in\mathbb{Z}$. Then any $x\gt a$ with $0\lt x-a\lt 1$ will have $f(x)=f(a)=a$, so $\lim\limits_{x\to a^+}f(x) = f(a)=a$.

However, if $x\lt a$, $0\lt a-x\lt 1$, theN $f(x) = a-1$, since the largest integer less than or equal to $x$ is $a-1$. So $\lim\limits_{x\to a^-}f(x) = a-1$.

Since $\lim\limits_{x\to a^+}f(x) \neq \lim\limits_{x\to a^-}f(x)$, then $\lim\limits_{x\to a}f(x)$ does not exist, and a fortiori $f(x)$ is not continuous at $a$.

Same argument with $\epsilon$ and $\delta$:

Let $a\in\mathbb{Z}$, and let $\epsilon=\frac{1}{2}$. For any $\delta\gt 0$ take $x=\mathrm{max}(a-1, a-\frac{\delta}{2})$. Then $a-1\leq x \lt a$, so $\lfloor x\rfloor = a-1$. Therefore, $|f(x)-f(a)| = |(a-1)-a| = 1\gt \epsilon$. Thus, there exists $\epsilon\gt 0$ (namely, $\epsilon=\frac{1}{2}$) such that for all $\delta\gt 0$ there is an $x$ with $0\lt |x-a|\lt \delta$, and $|f(x)-f(a)|\gt \epsilon$. Hence, $f(x)$ is not continuous at $a$.

Showing continuity everywhere else somewhat informally using limits:

Now assume that $a\notin\mathbb{Z}$, and let $\delta$ be the distance from $a$ to the nearest integer (that is, $\delta=\min(a-\lfloor a\rfloor, \rceil a\lceil - a)$); since $a\notin\mathbb{Z}$, $\delta\gt 0$.

Then for any $x$ with $|x-a|\lt \delta$, we have $\lfloor a\rfloor \lt x\lt \lceil a\rceil$, so $f(x) = \lfloor a\rfloor = f(a)$. Hence, for all $x$ sufficiently close to $a$, we have $f(x)=f(a)$, so $\lim\limits_{x\to a}f(x) = f(a)$, hence $f(x)$ is continuous at $a$.

I'll leave it to you to take that "somewhat informal" argument and turn it into a forma $\epsilon$-$\delta$ proof of continuity at $a\notin \mathbb{Z}$.

share|cite|improve this answer

Yes, that is correct.

If you like epsilons and deltas, you can say that at any integer, if you set $ε=1/2$, then there is no $δ$ so small that if the distance from $x$ to the integer in question is $<δ$, then the distance from $f(x)$ to the value of the function at the integer in question is $<ϵ$. For some of those values of $x$, the distance will be 1. But at non-integers, no matter how small $ε$ is, it is enough to make $δ$ small enough so that the neighborhood of radius $δ$ contains no integers.

share|cite|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

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