2015
DOI: 10.1007/s00208-015-1290-0
|View full text |Cite
|
Sign up to set email alerts
|

Lower bounds for the maximum of the Riemann zeta function along vertical lines

Abstract: Abstract. Let α ∈ (1/2, 1) be fixed. We prove thatfor all sufficiently large T , where we can choose c α = 0.18(2α−1) 1−α . The same result has already been obtained by Montgomery, with a smaller value for c α . However, our proof, which uses a modified version of Soundararajan's "resonance method" together with ideas of Hilberdink, is completely different from Montgomery's. This new proof also allows us to obtain lower bounds for the measure of those t ∈ [0, T ] for which |ζ(α + it)| is of the order mentioned… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1

Citation Types

0
112
0
13

Year Published

2018
2018
2022
2022

Publication Types

Select...
6
2

Relationship

2
6

Authors

Journals

citations
Cited by 45 publications
(127 citation statements)
references
References 16 publications
(49 reference statements)
0
112
0
13
Order By: Relevance
“…There are different methods to prove the existence of large values of the Riemann zeta function along vertical lines in the critical strip. Montgomery [24] used a method based on Diophantine approximation to prove that (1) max 2T ] log |ζ(σ + it)| ≥ C(σ)(log T ) 1−σ (log 2 T ) −σ .…”
Section: Introduction and Statement Of Resultsmentioning
confidence: 99%
See 1 more Smart Citation
“…There are different methods to prove the existence of large values of the Riemann zeta function along vertical lines in the critical strip. Montgomery [24] used a method based on Diophantine approximation to prove that (1) max 2T ] log |ζ(σ + it)| ≥ C(σ)(log T ) 1−σ (log 2 T ) −σ .…”
Section: Introduction and Statement Of Resultsmentioning
confidence: 99%
“…Since the set Bad * σ (q) is relatively small, it is enough to bound individually R(χ) for χ ∈ G q . From the definition (31) and our choice of Y we immediately obtain (34) |R(χ)| 2 ≤ 2 2π(Y ) ≤ exp(2Y / log Y ) ≤ q a(σ) log 2+o (1) .…”
Section: Accordingly We Havementioning
confidence: 97%
“…Dans , ce résultat est étendu à c<1β ce qui représente une amélioration d'un facteur 2false(1βfalse) lorsque β12. Le Théorème améliore donc les exposants obtenus par Bondarenko et Seip d'un nouveau facteur 2 pour tout β[0,1[. (ii)Notre approche diffère sensiblement de celle de : à l'instar des méthodes développées dans [, ], nous relions le carré du maximum à la somme de Gál complète S(M). L'introduction du carré explique que le gain obtenu n'est que la moitié de celui qui a été obtenu pour la somme de Gál soit 1222=2.…”
Section: Introduction Et éNoncé Des Résultatsunclassified
“…Pour ce choix de l'ensemble M, nous avons doncV + 2 (q) q 3/2 L (q) 2+o(1) . )/V + 1 (q) L (q) 1+o(1) .Cela achève la démonstration du Théorème 1.5. Sommes Un lemme de localisation et coprimalitéLe résultat suivant est une conséquence facile de la construction employée pourétablir le Théorème 1.1.…”
unclassified
“…Let A be a finite set of natural numbers, and let f : A −→ C be any function. For α ∈ (0, 1], the so-called GCD sum a,b∈A (a, b) 2α (ab) α f (a)f (b) (1) has received particular interest ( [10,9,2,5,6,11,8]), owing to its connections to the resonance method for finding large values of ζ(α + it) (for instance in [1,8]) and to equidistribution problems (for instance in [4]).…”
Section: Introductionmentioning
confidence: 99%