Buchstab function
In-game article clicks load inline without leaving the challenge.

The Buchstab function (or Buchstab's function) is the unique continuous function ω : R ≥ 1 → R > 0 {\displaystyle \omega :\mathbb {R} _{\geq 1}\rightarrow \mathbb {R} _{>0}} defined by the delay differential equation
ω ( u ) = 1 u , 1 ≤ u ≤ 2 , {\displaystyle \omega (u)={\frac {1}{u}},\qquad \qquad \qquad 1\leq u\leq 2,}
d d u ( u ω ( u ) ) = ω ( u − 1 ) , u ≥ 2. {\displaystyle {\frac {d}{du}}(u\omega (u))=\omega (u-1),\qquad u\geq 2.}
In the second equation, the derivative at u = 2 should be taken as u approaches 2 from the right. It is named after Alexander Buchstab, who wrote about it in 1937.
Asymptotics
The Buchstab function approaches e − γ ≈ 0.561 {\displaystyle e^{-\gamma }\approx 0.561} rapidly as u → ∞ , {\displaystyle u\to \infty ,} where γ {\displaystyle \gamma } is the Euler–Mascheroni constant. In fact,
| ω ( u ) − e − γ | ≤ ρ ( u − 1 ) u , u ≥ 1 , {\displaystyle |\omega (u)-e^{-\gamma }|\leq {\frac {\rho (u-1)}{u}},\qquad u\geq 1,}
where ρ is the Dickman function. Also, ω ( u ) − e − γ {\displaystyle \omega (u)-e^{-\gamma }} oscillates in a regular way, alternating between extrema and zeroes; the extrema alternate between positive maxima and negative minima. The interval between consecutive extrema approaches 1 as u approaches infinity, as does the interval between consecutive zeroes.
Applications
The Buchstab function is used to count rough numbers. If Φ(x, y) is the number of positive integers less than or equal to x with no prime factor less than y, then for any fixed u > 1,
Φ ( x , x 1 / u ) ∼ ω ( u ) x log x 1 / u , x → ∞ . {\displaystyle \Phi (x,x^{1/u})\sim \omega (u){\frac {x}{\log x^{1/u}}},\qquad x\to \infty .}
Notes
- Бухштаб, А. А. (1937), [Asymptotic estimation of a general number-theoretic function], Matematicheskii Sbornik (in Russian), 2(44) (6): 1239–1246, Zbl
- , Wolfram MathWorld. Accessed on line Feb. 11, 2015.
- §IV.32, "On Φ(x,y) and Buchstab's function", Handbook of Number Theory I, József Sándor, Dragoslav S. Mitrinović, and Borislav Crstici, Springer, 2006, ISBN 978-1-4020-4215-7.
- , A. Y. Cheer and D. A. Goldston, Mathematics of Computation 55 (1990), pp. 129–141.
- "An improvement of Selberg’s sieve method", W. B. Jurkat and H.-E. Richert, Acta Arithmetica 11 (1965), pp. 217–240.
- Hildebrand, A. (2001) [1994], , Encyclopedia of Mathematics, EMS Press