4
$\begingroup$

Denote by $A_n$ the number of prime numbers between $n$ and $n + \log n$.

Is it true that $A_n < const$?

UPD: Is it true that $A_n > \log \log n$ (or something another) for infinite number $n$?

$\endgroup$
4
  • 7
    $\begingroup$ No, thanks to the recent breakthroughs in bounded gaps between primes. $\endgroup$
    – Lucia
    Jul 1, 2015 at 21:06
  • $\begingroup$ @Lucia It's been over 2 years, can we still call it recent? :P $\endgroup$
    – Wojowu
    Jul 1, 2015 at 21:09
  • $\begingroup$ I don't understand: Are you mean results about twin primes? $\endgroup$ Jul 1, 2015 at 21:49
  • $\begingroup$ If the prime k-tuples conjecture holds, it may be that for infinitely many n that A_n is larger than (log n) /(log log n). Gerhard "Still Too Early To Tell" Paseman, 2015.07.01 $\endgroup$ Jul 1, 2015 at 22:06

2 Answers 2

11
$\begingroup$

Theorem 3.2 in Maynard shows that there are many values $x$ for which the interval $[x,x+\log x]$ contains $\gg \log \log x$ primes. This is a quantification of his earlier breakthrough work where he showed that there are intervals of length $\ll e^{(4+o(1))m}$ containing $m$ primes.

$\endgroup$
1
  • 1
    $\begingroup$ Excellent! For some reason I did not remember this refinement. $\endgroup$
    – GH from MO
    Jul 1, 2015 at 22:48
6
$\begingroup$

Based on the work of James Maynard and independently by Terry Tao, it is known that for any positive integer $k$, there exists a positive number $h(k)$ such that there exist infinitely many $k$-tuples of consecutive primes $p_1 < \cdots < p_k$ such that $p_k - p_1 < h(k)$. Applying this result to the current situation, for each positive integer $k$ there exist infinitely many $n$ such that between $n$ and $n + \log n$ there are more than $k$ primes (it is necessary of course to have $\log n > h(k)$, hence $A_n$ is not bounded.

See: http://annals.math.princeton.edu/2015/181-1/p07

and for the arxiv version:

http://arxiv.org/abs/1311.4600

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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