Dusart's explicit exponential error bound for the Chebyshev theta function
OpenTaoFivePrimes.dusart_theta_exponential_errorexplicit-boundsnumber-theoryprime-number-theorem
Set and . For every positive real satisfying , the Chebyshev function satisfies
This is the theta-function part of Dusart's explicit zero-free-region estimate, specialized to Kadiri's admissible constant . The hypothesis implies and , meeting both thresholds of the source theorem. It provides a reusable analytic input for converting exponential decay into explicit inverse powers of , including the large-value argument in Dusart's 2018 Theorem 4.2.
This statement remains an analytic proof obligation; it is not a certificate that the zero-free-region argument has been formalized.
Preamble
import Mathlib.NumberTheory.Chebyshev
Formal statement
theorem TaoFivePrimes.dusart_theta_exponential_error
(x : ℝ) (hx : 0 < x)
(hlog : 70 * (569693 / 100000 : ℝ) ≤ Real.log x) :
|Chebyshev.theta x - x| <
x * Real.sqrt (8 / Real.pi) *
Real.sqrt (Real.sqrt (Real.log x / (569693 / 100000 : ℝ))) *
Real.exp (-Real.sqrt (Real.log x / (569693 / 100000 : ℝ))) := by sorrySource
P. Dusart, Estimates of ψ, θ for large values of x without the Riemann hypothesis, Math. Comp. 85 (2016), 875–888, Theorem 1.1, DOI 10.1090/S0025-5718-2015-03005-1. Primary author restatement: HDR, Théorème 45, printed p.37, and proof of Corollaire 46, printed p.46 (explicitly permits R=5.69693), https://www.unilim.fr/pages_perso/pierre.dusart/Documents/HDR_Dusart.pdf. Specialization uses log x ≥ 70R to imply X ≥ max(8.36,8/R).