cp's OEIS Frontend

This is a front-end for the Online Encyclopedia of Integer Sequences, made by Christian Perfect. The idea is to provide OEIS entries in non-ancient HTML, and then to think about how they're presented visually. The source code is on GitHub.

A076446 Differences of consecutive powerful numbers (definition 1).

Original entry on oeis.org

3, 4, 1, 7, 9, 2, 5, 4, 13, 15, 8, 9, 19, 8, 13, 4, 3, 16, 25, 27, 4, 16, 9, 18, 13, 32, 1, 35, 19, 18, 31, 8, 32, 9, 43, 16, 12, 17, 47, 49, 23, 27, 1, 53, 55, 16, 41, 23, 36, 61, 7, 4, 28, 24, 65, 36, 27, 4, 69, 71, 27, 8, 21, 17, 3, 72, 77, 47, 32, 81, 47, 36, 36, 49, 87, 8
Offset: 1

Views

Author

Jud McCranie, Oct 15 2002

Keywords

Comments

The term 1 appears infinitely often. Erdős conjectured that two consecutive 1's do not occur. (see Guy).

Examples

			The first two powerful numbers are 1 and 4, their difference is 3, so a(1)=3.
		

References

  • R. K. Guy, Unsolved Problems in Number Theory, B16

Crossrefs

Programs

  • Haskell
    a076446 n = a076446_list !! (n-1)
    a076446_list = zipWith (-) (tail a001694_list) a001694_list
    -- Reinhard Zumkeller, Nov 30 2012
    
  • Mathematica
    Differences[Join[{1},Select[Range[2000],Min[FactorInteger[#][[All, 2]]]>1&]]] (* Harvey P. Dale, Aug 27 2017 *)
  • Python
    from math import isqrt
    from sympy import mobius, integer_nthroot
    def A076446(n):
        def squarefreepi(n): return int(sum(mobius(k)*(n//k**2) for k in range(1, isqrt(n)+1)))
        def bisection(f,kmin=0,kmax=1):
            while f(kmax) > kmax: kmax <<= 1
            while kmax-kmin > 1:
                kmid = kmax+kmin>>1
                if f(kmid) <= kmid:
                    kmax = kmid
                else:
                    kmin = kmid
            return kmax
        def f(x):
            c, l = n+x, 0
            j = isqrt(x)
            while j>1:
                k2 = integer_nthroot(x//j**2,3)[0]+1
                w = squarefreepi(k2-1)
                c -= j*(w-l)
                l, j = w, isqrt(x//k2**3)
            c -= squarefreepi(integer_nthroot(x,3)[0])-l
            return c
        return -(a:=bisection(f,n,n))+bisection(lambda x:f(x)+1,a,a) # Chai Wah Wu, Sep 10 2024

Formula

a(n) = A001694(n+1) - A001694(n).