Cramér's conjecture#Cramér model

In number theory, Cramér's conjecture, formulated by the Swedish mathematician Harald Cramér in 1936,{{Citation |last=Cramér |first=Harald |title=On the order of magnitude of the difference between consecutive prime numbers |url=http://matwbn.icm.edu.pl/ksiazki/aa/aa2/aa212.pdf |journal=Acta Arithmetica |volume=2 |year=1936 |pages=23–46 |doi=10.4064/aa-2-1-23-46 |access-date=2012-03-12 |archive-url=https://web.archive.org/web/20180723035707/http://matwbn.icm.edu.pl/ksiazki/aa/aa2/aa212.pdf |archive-date=2018-07-23 |url-status=dead }} is an estimate for the size of gaps between consecutive prime numbers: intuitively, that gaps between consecutive primes are always small, and the conjecture quantifies asymptotically just how small they must be. It states that

: $p_{n+1}-p_n=O((\log p_n)^2),$

where p_n denotes the nth prime number, O is big O notation, and "log" is the natural logarithm. While this is the statement explicitly conjectured by Cramér, his heuristic actually supports the stronger statement

: $\limsup_{n\rightarrow\infty} \frac{p_{n+1}-p_n}{(\log p_n)^2} = 1,$

and sometimes this formulation is called Cramér's conjecture. However, this stronger version is not supported by more accurate heuristic models, which nevertheless support the first version of Cramér's conjecture.

The strongest form of all, which was never claimed by Cramér but is the one used in experimental verification computations and the plot in this article, is simply

: $p_{n+1}-p_n < (\log p_n)^2.$

None of the three forms has yet been proven or disproven.

Conditional proven results on prime gaps

Cramér gave a conditional proof of the much weaker statement that

: $p_{n+1}-p_n = O(\sqrt{p_n}\,\log p_n)$

on the assumption of the Riemann hypothesis. The best known unconditional bound is

: $p_{n+1}-p_n = O(p_n^{0.525})$

due to Baker, Harman, and Pintz.{{Citation | vauthors=((Baker, R. C.)), ((Harman, G.)), ((Pintz, J.)) | year=2001 | title=The Difference Between Consecutive Primes, II | journal=Proceedings of the London Mathematical Society | volume=83 | issue=3 | pages=532–562 | publisher=Wiley | doi=10.1112/plms/83.3.532}}

In the other direction, E. Westzynthius proved in 1931 that prime gaps grow more than logarithmically. That is,{{Citation |last=Westzynthius |first=E. |title=Über die Verteilung der Zahlen die zu den n ersten Primzahlen teilerfremd sind |language=de|journal=Commentationes Physico-Mathematicae Helsingsfors |volume=5 |issue=5 |year=1931 |pages=1–37 | zbl=0003.24601 | jfm=57.0186.02 }}.

: $\limsup_{n\to\infty}\frac{p_{n+1}-p_n}{\log p_n}=\infty.$

His result was improved by R. A. Rankin,{{cite journal |first=R. A. |last=Rankin |author-link=R. A. Rankin |title=The difference between consecutive prime numbers |journal=J. London Math. Soc. |volume=13 |issue=4 |date=December 1938 |pages=242–247 |doi=10.1017/S0013091500025633 |doi-access=free }} who proved that

: $\limsup_{n\to\infty}\frac{p_{n+1}-p_n}{\log p_n}\cdot\frac{\left(\log\log\log p_n\right)^2}{ \log\log p_n \log\log\log\log p_n} > 0.$

Paul Erdős conjectured that the left-hand side of the above formula is infinite, and this was proven in 2014 by Kevin Ford, Ben Green, Sergei Konyagin, and Terence Tao,{{cite journal | last1=Ford | first1=Kevin | last2=Green | first2=Ben | last3=Konyagin | first3=Sergei | last4=Tao | first4=Terence | title=Large gaps between consecutive prime numbers | journal=Annals of Mathematics | series=Second series | volume=183 | date=2016 | issue=3 | pages=935–974 | doi=10.4007/annals.2016.183.3.4 | doi-access=free| arxiv=1408.4505 }} and independently by James Maynard.{{cite journal | last1=Maynard | first1=James | title=Large gaps between primes | journal=Annals of Mathematics | series=Second series | volume=183 | date=2016 | issue=3 | pages=915–933 | doi=10.4007/annals.2016.183.3.3 | doi-access=free| arxiv=1408.5110 }} The two sets of authors eliminated one of the factors of $\log \log \log p_n$ later that year,{{cite journal | last1=Ford | first1=Kevin | last2=Green | first2=Ben | last3=Konyagin | first3=Sergei | last4=Maynard | first4=James | last5=Tao | first5=Terence | title=Long gaps between primes | journal=Journal of the American Mathematical Society | volume=31 | date=2018 | pages=65–105 | doi=10.1090/jams/876 | doi-access=free | arxiv=1412.5029 }} showing that, infinitely often,

: $\ {p_{n+1}-p_n} { > } \frac{c\cdot\log p_n \cdot\log\log p_n \cdot\log\log\log\log p_n}{\log\log\log p_n}$

where $c > 0$ is some constant.

{{anchor|Cramér model}} Heuristic justification

Cramér's conjecture is based on a probabilistic model—essentially a heuristic—in which the probability that a number of size x is prime is 1/log x. This is known as the Cramér random model or Cramér model of the primes.Terry Tao, [https://terrytao.wordpress.com/2015/01/04/254a-supplement-4-probabilistic-models-and-heuristics-for-the-primes-optional/#more-7956 254A, Supplement 4: Probabilistic models and heuristics for the primes (optional)], section on The Cramér random model, January 2015.

In the Cramér random model,

: $\limsup_{n\rightarrow\infty} \frac{p_{n+1}-p_n}{\log^2 p_n} = 1$

with probability one. However, as pointed out by Andrew Granville,{{Citation |last=Granville |first=A. |title=Harald Cramér and the distribution of prime numbers |journal=Scandinavian Actuarial Journal |volume=1 |year=1995 |pages=12–28 |url=http://www.dartmouth.edu/~chance/chance_news/for_chance_news/Riemann/cramer.pdf |doi=10.1080/03461238.1995.10413946 |access-date=2007-06-05 |archive-date=2015-09-23 |archive-url=https://web.archive.org/web/20150923212842/http://www.dartmouth.edu/~chance/chance_news/for_chance_news/Riemann/cramer.pdf |url-status=dead }}. Maier's theorem shows that the Cramér random model does not adequately describe the distribution of primes on short intervals, and a refinement of Cramér's model taking into account divisibility by small primes suggests that the limit should not be 1, but a constant $c \ge 2e^{-\gamma}\approx1.1229\ldots$ ({{OEIS2C|id=A125313}}), where $\gamma$ is the Euler–Mascheroni constant. János Pintz has suggested that the limit sup may be infinite,{{cite journal |first=János |last=Pintz |author-link=János Pintz |title=Very large gaps between consecutive primes |journal=Journal of Number Theory |volume=63 |issue=2 |date=April 1997 |pages=286–301 |doi=10.1006/jnth.1997.2081 |url=https://core.ac.uk/download/pdf/81196811.pdf}} and similarly Leonard Adleman and Kevin McCurley write

:As a result of the work of H. Maier on gaps between consecutive primes, the exact formulation of Cramér's conjecture has been called into question [...] It is still probably true that for every constant $c>2$ , there is a constant $d>0$ such that there is a prime between $x$ and $x+d(\log x)^c$ .{{cite book |first1=Leonard |last1=Adleman |author-link=Leonard Adleman |first2=Kevin |last2=McCurley |chapter=Open Problems in Number Theoretic Complexity, II |title=ANTS-I: Proceedings of the First International Symposium on Algorithmic Number Theory |location=Ithaca, NY |date=6 May 1994 |pages=291–322 |series=Lecture Notes in Computer Science |volume=877 |publisher=Springer |doi=10.1007/3-540-58691-1_70 |isbn=3-540-58691-1 |citeseerx=10.1.1.48.4877}}

Similarly, Robin Visser writes

:In fact, due to the work done by Granville, it is now widely believed that Cramér's conjecture is false. Indeed, there [are] some theorems concerning short intervals between primes, such as Maier's theorem, which contradict Cramér's model.Robin Visser, [https://warwick.ac.uk/fac/sci/maths/people/staff/visser/large_gaps_between_primes.pdf Large Gaps Between Primes], University of Cambridge (2020).

(internal references removed).

Related conjectures and heuristics<span id="Shanks conjecture"></span><span id="Cramér–Granville conjecture"></span>

File:Primegaps-new.png

Daniel Shanks conjectured the following asymptotic equality, stronger than Cramér's conjecture,{{Citation |first=Daniel |last=Shanks |title=On Maximal Gaps between Successive Primes |journal=Mathematics of Computation |volume=18 |issue=88 |year=1964 |pages=646–651 |doi=10.2307/2002951 |publisher=American Mathematical Society |jstor=2002951|zbl=0128.04203 |doi-access=free }}. for record gaps:

$G(x)\sim \log^2 x.$

J.H. Cadwell{{Citation |last=Cadwell |first= J. H. |title=Large Intervals Between Consecutive Primes |journal= Mathematics of Computation |volume= 25 |issue=116 |year=1971 |pages=909–913 |jstor=2004355 |doi=10.2307/2004355|doi-access=free }} has proposed the formula for the maximal gaps:

$G(x)\sim \log^2 x - \log x\log\log x,$

which is formally identical to the Shanks conjecture but suggests a lower-order term.

Marek Wolf{{Citation |last=Wolf |first=Marek |title=Nearest-neighbor-spacing distribution of prime numbers and quantum chaos |journal=Phys. Rev. E |volume=89 |year=2014 |issue=2 |pages=022922 |url=https://journals.aps.org/pre/abstract/10.1103/PhysRevE.89.022922 |doi=10.1103/physreve.89.022922|pmid=25353560 |arxiv=1212.3841 |bibcode=2014PhRvE..89b2922W |s2cid=25003349 }} has proposed the formula for the maximal gaps $G(x)$

expressed in terms of the prime-counting function

$\pi(x)$ :

: $G(x)\sim \frac{x}{\pi(x)}(2\log\pi(x)-\log x+c),$

where $c=\log(2C_2)=0.2778769...$ and $C_2=0.6601618...$ is the twin primes constant; see {{OEIS2C|A005597}}, {{OEIS link|A114907}}. This is again formally equivalent to the Shanks conjecture but suggests lower-order terms

: $G(x) \sim \log^2 x - 2\log x\log\log x - (1-c)\log x.$ .

Thomas Nicely has calculated many large prime gaps.{{Citation |last=Nicely |first=Thomas R. |doi=10.1090/S0025-5718-99-01065-0 |mr=1627813 |issue=227 |journal=Mathematics of Computation |pages=1311–1315 |title=New maximal prime gaps and first occurrences |volume=68 |year=1999 |doi-access=free |bibcode=1999MaCom..68.1311N }}. He measures the quality of fit to Cramér's conjecture by measuring the ratio

: $R = \frac{\log p_n}{\sqrt{p_{n+1}-p_n}}.$

He writes, "For the largest known maximal gaps, $R$ has remained near 1.13."

References

{{cite book |last=Guy | first=Richard K. | author-link=Richard K. Guy | title=Unsolved problems in number theory | publisher=Springer-Verlag |edition=3rd | year=2004 | isbn=978-0-387-20860-2 | zbl=1058.11001 | at=A8 }}
{{cite journal | author-link=János Pintz | last1=Pintz | first1=János | title=Cramér vs. Cramér. On Cramér's probabilistic model for primes | url=http://projecteuclid.org/euclid.facm/1229619660 | mr=2363833 | year=2007 | journal= Functiones et Approximatio Commentarii Mathematici | volume=37 | issue=2 | pages=361–376 | zbl=1226.11096 | issn=0208-6573 | doi=10.7169/facm/1229619660| doi-access=free }}
{{cite book | last=Soundararajan | first=K. | author-link=Kannan Soundararajan | chapter=The distribution of prime numbers | editor1-last=Granville | editor1-first=Andrew | editor1-link=Andrew Granville | editor2-last=Rudnick | editor2-first=Zeév | title=Equidistribution in number theory, an introduction. Proceedings of the NATO Advanced Study Institute on equidistribution in number theory, Montréal, Canada, July 11--22, 2005 | location=Dordrecht |publisher=Springer-Verlag | series=NATO Science Series II: Mathematics, Physics and Chemistry | volume=237 | pages=59–83 | year=2007 | isbn=978-1-4020-5403-7 | zbl=1141.11043 }}

External links

{{mathworld|title=Cramér Conjecture|urlname=CramerConjecture}}
{{mathworld|title=Cramér-Granville Conjecture|urlname=Cramer-GranvilleConjecture}}

Category:Analytic number theory

Category:Conjectures about prime numbers

Category:Unsolved problems in number theory