Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

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

Is there a function $f:\mathbb N\to \mathbb R$ such that $\lim_{n\to\infty} f(n) = 0$ and for every $c\in\mathbb C$:

If $z_0=0$, $z_{n+1}=z_n^2+c$ and $|z_k|<2$, then there exists a point $c'$ in the Mandelbrot set satisfying $|c-c'|<f(k)$?

Such fact would be very useful for rendering the Mandelbrot set, as it would give the exact count of iterations needed for given resolution.

share|cite|improve this question
What is the motivation for the condition $|z_k| < 2$? Also, do you mean that for any $k$ with $|z_k| < 2$, $|c - c'| < f(k)$, or do you mean that if $|z_k| < 2$ for all $k$< $|c - c'| < f(k)$ for all $k$? – 6005 Jun 12 '13 at 14:48
@Goos: I think neither? What the OP wants is probably "If the Mandelbrot iteration starting from $c$ stays within the critical circle $|z| < 2$ for $k$ steps, then there exists $c'$ in the Mandelbrot set such that $|c - c'| < f(k)$." – Willie Wong Jun 12 '13 at 14:59
On the contrary, you have Koebe $\frac14$ disks completely disjoint to $M$ for every point that does escape. – Hagen von Eitzen Jun 12 '13 at 15:00
@Goos: It's well known that if $|z_k|>2$, then $(z_n)$ diverges, that's why. I mean that for any k with $|z_k|<2$ you can find $c'$ (depending on that k) such that $|c-c'|<f(k)$. I don't get the second part of this question (from "or do you mean"), why is "for all $k$" twice and what's with "$k<|c-c'|$"? – janek37 Jun 12 '13 at 15:00
@HagenvonEitzen: can you elaborate? Is the radius of the Koebe disk independent of $k$? – janek37 Jun 12 '13 at 15:41
up vote 0 down vote accepted

In principle, yes. The set of points whose orbits stay within $|z|< 4$ for the first $n$ iterations is a bounded open subset $M_n\subset\mathbb C$. We have $M_{n+1}\Subset M_n$ and $M=\bigcap _{n\in\mathbb N}M_n$. Let $f(n)=\max\{d(z,M)\mid z\in \overline{M_n}\}$. Then $z\in M_n$ implies there is a point $c\in M$ with $d(c,z)\le f(n)$. The sequence $(f(n))$ is nonincreasing, hence $a:=\lim_{n\to\infty} f(n)$ exists. Then for all $n$, we get a point $z_n\in M_n$ with $d(z_n,M)=a$. A subsequence of $(z_n)$ converges to some $\zeta$ with $d(\zeta,M)=a$, but we also have $\zeta\in\bigcap \overline{M_n}=M$. However, this mere existence of $f$ won't help you in practice.

share|cite|improve this answer
You're right. I hoped that there are some upper bounds for the $f$ you defined. Hopefully in the form $f(n)<Cq^n$ for some $C>0$, $0<q<1$. – janek37 Jun 12 '13 at 17:15
I've found… which helps a little, but still doesn't really hint on the number of iterations. – janek37 Jun 12 '13 at 17:29

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.