OFFSET
0,1
COMMENTS
4*n + 3 = (8*n + 6) / 2 is never a square, as 3 is not a quadratic residue modulo 4. Using this, we can show that each term has an even square part and an even squarefree part, neither part being a power of 2. (Less than 2% of integers have this property - see A339245.) - Peter Munn, Dec 14 2020
LINKS
Vincenzo Librandi, Table of n, a(n) for n = 0..10000
Eric Weisstein's World of Mathematics, Quadratic Residue.
Index entries for linear recurrences with constant coefficients, signature (4,-6,4,-1).
FORMULA
From R. J. Mathar, Mar 22 2010: (Start)
G.f.: 8*(27 + 235*x + 121*x^2 + x^3)/(x-1)^4.
a(n) = 8*A016839(n). (End)
a(0)=216, a(1)=2744, a(2)=10648, a(3)=27000, a(n) = 4*a(n-1) - 6*a(n-2) + 4*a(n-3) - a(n-4). - Harvey P. Dale, Dec 11 2012
Sum_{n>=0} 1/a(n) = 7*zeta(3)/128 - Pi^2/512. - Amiram Eldar, Apr 26 2023
MATHEMATICA
Table[(8*n+6)^3, {n, 0, 5!}] (* Vladimir Joseph Stephan Orlovsky, Mar 17 2010 *)
LinearRecurrence[{4, -6, 4, -1}, {216, 2744, 10648, 27000}, 30] (* Harvey P. Dale, Dec 11 2012 *)
PROG
(Magma) [(8*n+6)^3: n in [0..35]]; // Vincenzo Librandi, Jul 22 2011
CROSSREFS
KEYWORD
nonn,easy
AUTHOR
EXTENSIONS
More terms from Vladimir Joseph Stephan Orlovsky, Mar 17 2010
STATUS
approved