Open main menu
Home
Random
Recent changes
Special pages
Community portal
Preferences
About Wikipedia
Disclaimers
Incubator escapee wiki
Search
User menu
Talk
Dark mode
Contributions
Create account
Log in
Editing
Skewes's number
Warning:
You are not logged in. Your IP address will be publicly visible if you make any edits. If you
log in
or
create an account
, your edits will be attributed to your username, along with other benefits.
Anti-spam check. Do
not
fill this in!
{{unsolved|mathematics|What is the smallest Skewes's number?}} {{Short description|Large number used in number theory}} In [[number theory]], '''Skewes's number''' is the smallest [[natural number]] <math>x</math> for which the [[prime-counting function]] <math>\pi(x)</math> exceeds the [[logarithmic integral function]] <math>\operatorname{li}(x).</math> It is named for the [[South Africa]]n mathematician [[Stanley Skewes]] who first computed an [[upper bound]] on its value. The exact value of Skewes's number is still not known, but it is known that there is a crossing between <math>\pi(x) < \operatorname{li}(x)</math> and <math>\pi(x) > \operatorname{li}(x)</math> near <math>e^{727.95133} < 1.397 \times 10^{316}.</math> It is not known whether this is the smallest crossing. The name is sometimes also applied to either of the [[large number]] bounds which Skewes found. ==Skewes's bounds== Although nobody has ever found a value of <math>x</math> for which <math>\pi(x) > \operatorname{li}(x),</math> Skewes's research supervisor [[John Edensor Littlewood|J.E. Littlewood]] had proved in {{harvtxt|Littlewood|1914}} that there is such a number (and so, a first such number); and indeed found that the sign of the difference <math>\pi(x) - \operatorname{li}(x)</math> changes infinitely many times. Littlewood's proof did not, however, exhibit a concrete such number <math>x</math>, nor did it even give any bounds on the value. Skewes's task was to make Littlewood's existence proof [[Effective results in number theory|effective]]: exhibit some concrete upper bound for the first sign change. According to [[Georg Kreisel]], this was not considered obvious even in principle at the time.<ref>{{harvtxt|Kreisel|1951}} quotes A. E. Ingham (1932) and J. E. Littlewood (1948) as stating "that the proof was believed to be 'non-constructive', or to require 'new ideas' of proof to make it constructive."</ref> {{harvtxt|Skewes|1933}} proved that, assuming that the [[Riemann hypothesis]] is true, there exists a number <math>x</math> violating <math>\pi(x) < \operatorname{li}(x),</math> below :<math>e^{e^{e^{79}}}<10^{10^{10^{34}}}.</math> Without assuming the Riemann hypothesis, {{harvtxt|Skewes|1955}} later proved that there exists a value of <math>x</math> below :<math>e^{e^{e^{e^{7.705}}}}<10^{10^{10^{964}}}.</math> ==More recent bounds== These upper bounds have since been reduced considerably by using large-scale computer calculations of [[zero of a function|zeros]] of the [[Riemann zeta function]]. The first estimate for the actual value of a crossover point was given by {{harvtxt|Lehman|1966}}, who showed that somewhere between <math>1.53\times 10^{1165}</math> and <math>1.65\times 10^{1165}</math> there are more than <math>10^{500}</math> consecutive [[integer]]s <math>x</math> with <math>\pi(x) > \operatorname{li}(x)</math>. Without assuming the Riemann hypothesis, {{harvs|txt=yes|authorlink=Herman te Riele|first=H. J. J.|last= te Riele |year= 1987}} proved an upper bound of <math>7\times 10^{370}</math>. A better estimate was <math>1.39822\times 10^{316}</math> discovered by {{harvtxt|Bays|Hudson|2000}}, who showed there are at least <math>10^{153}</math> consecutive integers somewhere near this value where <math>\pi(x) > \operatorname{li}(x)</math>. Bays and Hudson found a few much smaller values of <math>x</math> where <math>\pi(x)</math> gets close to <math>\operatorname{li}(x)</math>; the possibility that there are crossover points near these values does not seem to have been definitely ruled out yet, though computer calculations suggest they are unlikely to exist. {{harvtxt|Chao|Plymen|2010}} gave a small improvement and correction to the result of Bays and Hudson. {{harvtxt|Saouter|Demichel|2010}} found a smaller interval for a crossing, which was slightly improved by {{harvtxt|Zegowitz|2010}}. The same source shows that there exists a number <math>x</math> violating <math>\pi(x) < \operatorname{li}(x),</math> below <math>e^{727.9513468}< 1.39718 \times 10^{316}</math>. This can be reduced to <math>e^{727.9513386}< 1.39717 \times 10^{316}</math> assuming the Riemann hypothesis. {{harvtxt|Stoll|Demichel|2011}} conducted an analysis with up to 2{{e|11}} complex zeros which gives computational evidence that a crossover may exist near <math>1.397162914 \times 10^{316}</math>. {|class="wikitable" style="text-align:left" |+ Interval of crossing !Year!! # of complex<br/> zeros used !! by !! Interval !! # of consecutive integers with <math>\pi(x) > \operatorname{li}(x)</math> given |- |2000|| {{10^|6}} || Bays and Hudson || [1.39821924{{e|316}}, 1.39821925{{e|316}}] || > 1{{e|153}} |- |2010|| {{10^|7}} || Chao and Plymen || [exp(727.951858), exp(727.952178)] || 1{{e|154}} |- |2010|| 2.2{{e|7}} || Saouter and Demichel || [exp(727.95132478), exp(727.95134682)] (without RH) <br/> [exp(727.95133239), exp(727.95133920)] (assuming RH) || 6.6587{{e|152}} (without RH) <br/> 1.2741{{e|151}} (assuming RH) |- |2010|| 2.2{{e|7}} || Zegowitz || [exp(727.951324783), exp(727.951346802)] (without RH) <br/> [exp(727.951332973), exp(727.951338612)] (assuming RH) || 6.695531258{{e|152}} (without RH) <br/> 1.15527413{{e|152}} (assuming RH) |- |} Rigorously, {{harvtxt|Rosser|Schoenfeld|1962}} proved that there are no crossover points below <math>x = 10^8</math>, improved by {{harvtxt|Brent|1975}} to <math>8\times 10^{10}</math>, by {{harvtxt|Kotnik|2008}} to <math>10^{14}</math>, by {{harvtxt|Platt|Trudgian|2014}} to <math>1.39\times 10^{17}</math>, and by {{harvtxt|Büthe|2015}} to <math>10^{19}</math>. There is no explicit value <math>x</math> known for certain to have the property <math>\pi(x) > \operatorname{li}(x),</math> though computer calculations suggest some explicit numbers that are quite likely to satisfy this. Even though the [[natural density]] of the positive integers for which <math>\pi(x) > \operatorname{li}(x)</math> does not exist, {{harvtxt|Wintner|1941}} showed that the [[Natural_density#Other_density_functions|logarithmic density]] of these positive integers does exist and is positive. {{harvtxt|Rubinstein|Sarnak|1994}} showed that this proportion is about {{val|2.6|e=-7}}, which is surprisingly large given how far one has to go to find the first example. ==Riemann's formula== [[Riemann]] gave an [[Explicit formulae (L-function)|explicit formula]] for <math>\pi(x)</math>, whose leading terms are (ignoring some subtle convergence questions) :<math>\pi(x) = \operatorname{li}(x) - \tfrac{1}{2}\operatorname{li}(\sqrt{x\,}) - \sum_{\rho} \operatorname{li}(x^\rho) + \text{smaller terms} </math> where the sum is over all <math>\rho</math> in the set of [[Riemann hypothesis|non-trivial zeros of the Riemann zeta function]]. The largest error term in the approximation <math>\pi(x) \approx \operatorname{li}(x)</math> (if the [[Riemann hypothesis]] is true) is negative <math>\tfrac{1}{2}\operatorname{li}(\sqrt{x\,})</math>, showing that <math>\operatorname{li}(x)</math> is usually larger than <math>\pi(x)</math>. The other terms above are somewhat smaller, and moreover tend to have different, seemingly random complex [[Argument (complex analysis)|arguments]], so mostly cancel out. Occasionally however, several of the larger ones might happen to have roughly the same complex argument, in which case they will reinforce each other instead of cancelling and will overwhelm the term <math>\tfrac{1}{2}\operatorname{li}(\sqrt{x\,})</math>. The reason why the Skewes number is so large is that these smaller terms are quite a ''lot'' smaller than the leading error term, mainly because the first [[complex number|complex]] zero of the zeta function has quite a large [[complex number|imaginary part]], so a large number (several hundred) of them need to have roughly the same argument in order to overwhelm the dominant term. The chance of <math>N</math> random complex numbers having roughly the same argument is about 1 in <math>2^N</math>. This explains why <math>\pi(x)</math> is sometimes larger than <math>\operatorname{li}(x),</math> and also why it is rare for this to happen. It also shows why finding places where this happens depends on large scale calculations of millions of high precision zeros of the Riemann zeta function. The argument above is not a proof, as it assumes the zeros of the Riemann zeta function are random, which is not true. Roughly speaking, Littlewood's proof consists of [[Dirichlet's approximation theorem]] to show that sometimes many terms have about the same argument. In the event that the Riemann hypothesis is false, the argument is much simpler, essentially because the terms <math>\operatorname{li}(x^{\rho})</math> for zeros violating the Riemann hypothesis (with [[complex number|real part]] greater than {{sfrac|1|2}}) are eventually larger than <math>\operatorname{li}(x^{1/2})</math>. The reason for the term <math>\tfrac{1}{2}\mathrm{li}(x^{1/2})</math> is that, roughly speaking, <math>\mathrm{li}(x)</math> actually counts powers of [[prime number|primes]], rather than the primes themselves, with <math>p^n</math> weighted by <math>\frac{1}{n}</math>. The term <math>\tfrac{1}{2}\mathrm{li}(x^{1/2})</math> is roughly analogous to a second-order correction accounting for [[square (algebra)|squares]] of primes. ==Equivalent for prime ''k''-tuples== An equivalent definition of Skewes's number exists for [[prime k-tuple|prime ''k''-tuples]] ({{harvtxt|Tóth|2019}}). Let <math>P = (p, p+i_1, p+i_2, ..., p+i_k)</math> denote a prime (''k'' + 1)-tuple, <math>\pi_P(x)</math> the number of primes <math>p</math> below <math>x</math> such that <math>p, p+i_1, p+i_2, ..., p+i_k</math> are all prime, let <math>\operatorname{li_P}(x) = \int_2^x \frac{dt}{(\ln t)^{k+1}}</math> and let <math>C_P</math> denote its Hardy–Littlewood constant (see [[First Hardy–Littlewood conjecture]]). Then the first prime <math>p</math> that violates the Hardy–Littlewood inequality for the (''k'' + 1)-tuple <math>P</math>, i.e., the first prime <math>p</math> such that : <math>\pi_P(p) > C_P \operatorname{li}_P(p), </math> (if such a prime exists) is the ''Skewes number for <math>P.</math>'' The table below shows the currently known Skewes numbers for prime ''k''-tuples: {|class="wikitable" style="text-align:left" !Prime ''k''-tuple!! Skewes number!! Found by |- |(''p'', ''p'' + 2)|| 1369391 || {{harvtxt|Wolf|2011}} |- |(''p'', ''p'' + 4)|| 5206837 || {{harvtxt|Tóth|2019}} |- |(''p'', ''p'' + 2, ''p'' + 6)|| 87613571 || Tóth (2019) |- |(''p'', ''p'' + 4, ''p'' + 6)|| 337867 || Tóth (2019) |- |(''p'', ''p'' + 2, ''p'' + 6, ''p'' + 8)|| 1172531 || Tóth (2019) |- |(''p'', ''p'' + 4, ''p'' +6 , ''p'' + 10)|| 827929093 || Tóth (2019) |- |(''p'', ''p'' + 2, ''p'' + 6, ''p'' + 8, ''p'' + 12)|| 21432401 || Tóth (2019) |- |(''p'', ''p'' +4 , ''p'' +6 , ''p'' + 10, ''p'' + 12)|| 216646267 || Tóth (2019) |- |(''p'', ''p'' + 4, ''p'' + 6, ''p'' + 10, ''p'' + 12, ''p'' + 16)|| 251331775687 || Tóth (2019) |- |(''p'', ''p''+2, ''p''+6, ''p''+8, ''p''+12, ''p''+18, ''p''+20)|| 7572964186421 || Pfoertner (2020) |- |(''p'', ''p''+2, ''p''+8, ''p''+12, ''p''+14, ''p''+18, ''p''+20)|| 214159878489239 || Pfoertner (2020) |- |(''p'', ''p''+2, ''p''+6, ''p''+8, ''p''+12, ''p''+18, ''p''+20, ''p''+26)|| 1203255673037261 || Pfoertner / Luhn (2021) |- |(''p'', ''p''+2, ''p''+6, ''p''+12, ''p''+14, ''p''+20, ''p''+24, ''p''+26)|| 523250002674163757 || Luhn / Pfoertner (2021) |- |(''p'', ''p''+6, ''p''+8, ''p''+14, ''p''+18, ''p''+20, ''p''+24, ''p''+26)|| 750247439134737983 || Pfoertner / Luhn (2021) |} The Skewes number (if it exists) for [[sexy prime]]s <math>(p, p+6)</math> is still unknown. It is also unknown whether all admissible ''k''-tuples have a corresponding Skewes number. == See also == * {{Slink|Mertens' theorems#Changes in sign}} ==References== {{Reflist}} {{Refbegin}} *{{citation|mr=1752093|first1=C.|last1= Bays |first2=R. H.|last2= Hudson |url=http://www.ams.org/mcom/2000-69-231/S0025-5718-99-01104-7/S0025-5718-99-01104-7.pdf |title=A new bound for the smallest <math>x</math> with <math>\pi(x) > \operatorname{li}(x)</math> |journal=[[Mathematics of Computation]]|volume=69|year=2000 |issue= 231|pages= 1285–1296 | zbl=1042.11001 | doi=10.1090/S0025-5718-99-01104-7 |doi-access=free }} *{{citation|mr=0369287|first=R. P.|last= Brent|authorlink=Richard P. Brent |title=Irregularities in the distribution of primes and twin primes |journal=[[Mathematics of Computation]]|volume=29|year=1975 |pages= 43–56|doi=10.2307/2005460| zbl=0295.10002 | jstor=2005460|issue=129|doi-access=free}} *{{citation|first=Jan|last=Büthe| title=An analytic method for bounding <math>\psi(x)</math>|year=2015|arxiv=1511.02032|bibcode=2015arXiv151102032B}} *{{citation | title=A new bound for the smallest <math>x</math> with <math>\pi(x) > \operatorname{li}(x)</math>|first1=Kuok Fai|last1= Chao|first2= Roger|last2= Plymen|year=2010| arxiv=math/0509312 |journal= [[International Journal of Number Theory]] |volume= 6|issue=3|pages= 681–690|mr=2652902 | zbl=1215.11084 | doi=10.1142/S1793042110003125 }} *{{citation|first= T.|last= Kotnik |doi=10.1007/s10444-007-9039-2 |title=The prime-counting function and its analytic approximations |journal=Advances in Computational Mathematics|volume=29|issue= 1|year=2008|pages= 55–70 | zbl=1149.11004 | mr = 2420864|s2cid= 18991347 }} *{{citation | last = Kreisel | first = G. | doi = 10.1017/S0022481200100581 | journal = The Journal of Symbolic Logic | jstor = 2267908 | mr = 49135 | pages = 241–267 | title = On the interpretation of non-finitist proofs. I | volume = 16 | year = 1951}} *{{citation|first= R. Sherman |last=Lehman|title= On the difference <math>\pi(x)-\operatorname{li}(x)</math>|journal= [[Acta Arithmetica]] |volume=11 |year=1966|pages= 397–410 | zbl=0151.04101 | mr=0202686 | url=https://eudml.org/doc/204773 | doi=10.4064/aa-11-4-397-410|doi-access=free }} * {{citation|first=J. E.|last= Littlewood | authorlink=J. E. Littlewood | title=Sur la distribution des nombres premiers|journal=[[Comptes Rendus]]|volume= 158 |year=1914|pages= 1869–1872 | jfm=45.0305.01}} * {{citation|first1=D. J.|last1=Platt|first2=T. S.|last2=Trudgian|author2-link=Timothy Trudgian|title=On the first sign change of <math>\theta(x)-x</math>|year=2014|arxiv=1407.1914|bibcode=2014arXiv1407.1914P}} *{{citation|mr=0866118|first= H. J. J. |last=te Riele|authorlink=Herman te Riele|title=On the sign of the difference <math>\pi(x)-\operatorname{li}(x)</math>|journal=[[Mathematics of Computation]]|volume=48|year=1987|pages= 323–328 |jstor=2007893 |issue=177 | doi = 10.1090/s0025-5718-1987-0866118-6 |doi-access=free}} *{{citation|mr=0137689|first1= J. B.|last1= Rosser|author1-link=J. Barkley Rosser |first2= L.|last2= Schoenfeld|author2-link=Lowell Schoenfeld |title=Approximate formulas for some functions of prime numbers |journal=Illinois Journal of Mathematics|volume=6|year=1962|pages= 64–94|doi= 10.1215/ijm/1255631807|doi-access=free}} *{{citation | last1 = Saouter | first1 = Yannick | last2 = Demichel | first2 = Patrick | doi = 10.1090/S0025-5718-10-02351-3 | mr = 2684372 | issue = 272 | journal = [[Mathematics of Computation]] | pages = 2395–2405 | title = A sharp region where <math>\pi(x)-\operatorname{li}(x)</math> is positive | volume = 79 | year = 2010| doi-access = free }} *{{citation|mr=1329368 |author2-link=Peter Sarnak |last1=Rubinstein|first1= M.|last2= Sarnak|first2= P. |title=Chebyshev's bias |journal=[[Experimental Mathematics (journal)|Experimental Mathematics]] |volume=3 |year=1994|issue= 3|pages= 173–197 |url= http://projecteuclid.org/euclid.em/1048515870 |doi=10.1080/10586458.1994.10504289}} *{{citation|first= S.|last= Skewes|authorlink= Stanley Skewes |title=On the difference <math>\pi(x)-\operatorname{li}(x)</math>|journal=[[Journal of the London Mathematical Society]]|volume=8|year=1933|pages= 277–283 | zbl=0007.34003 | jfm=59.0370.02 |doi=10.1112/jlms/s1-8.4.277}} *{{citation|mr=0067145| first= S.|last= Skewes|authorlink= Stanley Skewes |title=On the difference <math>\pi(x)-\operatorname{li}(x)</math> (II)|journal=[[Proceedings of the London Mathematical Society]]|volume= 5 |year=1955|pages= 48–70|doi=10.1112/plms/s3-5.1.48}} * {{citation|first1=Douglas|last1=Stoll| first2=Patrick|last2= Demichel|title= The impact of <math>\zeta(s)</math> complex zeros on <math>\pi(x)</math> for <math>x < 10^{10^{13}}</math>|year=2011|journal=[[Mathematics of Computation]]|volume=80|issue=276|pages=2381–2394|mr=2813366|doi=10.1090/S0025-5718-2011-02477-4|doi-access=free}} *{{citation|first=László|last=Tóth|title=On The Asymptotic Density Of Prime k-tuples and a Conjecture of Hardy and Littlewood|journal=Computational Methods in Science and Technology|volume=25|year=2019|issue=3|doi=10.12921/cmst.2019.0000033 |s2cid=203836016 |url=http://cmst.eu/wp-content/uploads/files/10.12921_cmst.2019.0000033_TOTH.pdf}}. *{{citation|mr=0004255|last= Wintner|first= A. |authorlink=Aurel Wintner |title=On the distribution function of the remainder term of the prime number theorem |journal= [[American Journal of Mathematics]]|volume= 63|year=1941|pages= 233–248|issue=2|doi=10.2307/2371519|jstor=2371519 }} *{{citation|first=Marek|last=Wolf|title=The Skewes number for twin primes: counting sign changes of π2(x) − C2Li2(x)|journal=Computational Methods in Science and Technology|volume=17|year=2011|pages=87–92 |doi=10.12921/cmst.2011.17.01.87-92 |s2cid=59578795 |url=http://cmst.eu/wp-content/uploads/files/10.12921_cmst.2011.17.01.87-92_Wolf_old.pdf}}. *{{citation|last=Zegowitz|first=Stefanie|title=On the positive region of <math>\pi(x)-\operatorname{li}(x)</math>|year=2010|url=http://eprints.ma.man.ac.uk/1547/|publisher=Manchester Institute for Mathematical Sciences, School of Mathematics, University of Manchester|series=Master's thesis|type=masters }} {{Refend}} ==External links== *{{cite web |url= http://demichel.net/patrick/li_crossover_pi.pdf |url-status= dead |archive-date= Sep 8, 2006 |archive-url= https://web.archive.org/web/20060908033007/http://demichel.net/patrick/li_crossover_pi.pdf |title= The prime counting function and related subjects |first= Patrick |last= Demichels |website= Demichel |access-date= 2009-09-29 }} * {{cite book |last= Asimov |first= I. |chapter= Skewered! |title= Of Matters Great and Small. |location= New York |publisher= Ace Books |date= 1976 |isbn= 978-0441610723 }} {{Large numbers}} [[Category:Large numbers]] [[Category:Number theory]] [[Category:Large integers]]
Edit summary
(Briefly describe your changes)
By publishing changes, you agree to the
Terms of Use
, and you irrevocably agree to release your contribution under the
CC BY-SA 4.0 License
and the
GFDL
. You agree that a hyperlink or URL is sufficient attribution under the Creative Commons license.
Cancel
Editing help
(opens in new window)
Pages transcluded onto the current version of this page
(
help
)
:
Template:10^
(
edit
)
Template:Citation
(
edit
)
Template:Cite book
(
edit
)
Template:Cite web
(
edit
)
Template:E
(
edit
)
Template:Harvs
(
edit
)
Template:Harvtxt
(
edit
)
Template:Large numbers
(
edit
)
Template:Refbegin
(
edit
)
Template:Refend
(
edit
)
Template:Reflist
(
edit
)
Template:Sfrac
(
edit
)
Template:Short description
(
edit
)
Template:Slink
(
edit
)
Template:Unsolved
(
edit
)
Template:Val
(
edit
)