OFFSET
1,7
COMMENTS
On Jan 09 2009, Zhi-Wei Sun conjectured that a(n)>0 for every n=5,6,...; in other words, any integer n>4 can be written in the form p+F_s+F_{3t}/2 with p an odd prime and s,t>0. Sun verified this up to 5*10^6 and Qing-Hu Hou continued the verification (on Sun's request) up to 3*10^8. Note that 932633 cannot be written as p+F_s+F_{3t}/2 with p a prime and (F_s or F_{3t}/2) odd. If we set u_0=0, u_1=1 and u_{n+1}=4u_n+u_{n-1} for n=1,2,3,..., then F_{3t}/2=u_t is at least 4^{t-1} for each t=1,2,3,.... In a recent paper K. J. Wu and Z. W. Sun constructed a residue class which contains no integers of the form p+F_{3t}/2 with p a prime and t nonnegative.
REFERENCES
R. Crocker, On a sum of a prime and two powers of two, Pacific J. Math. 36(1971), 103-107.
LINKS
Zhi-Wei SUN, Table of n, a(n), n=1..50000.
D. S. McNeil, Sun's strong conjecture
Zhi-Wei Sun, A promising conjecture: n=p+F_s+F_t
Zhi-Wei Sun, A summary concerning my conjecture n=p+F_s+F_t
K. J. Wu and Z. W. Sun, Covers of the integers with odd moduli and their applications to the forms x^m-2^n and x^2-F_{3n}/2, Math. Comp. 78 (2009) 1853, [DOI], arXiv:math.NT/0702382
Terence Tao, A remark on primality testing and decimal expansions, Journal of the Australian Mathematical Society 91:3 (2011), pp. 405-413.
FORMULA
a(n) = |{<p,s,t>: p+F_s+F_{3t}/2=n with p an odd prime, s>1 and t>0}|.
EXAMPLE
For n=9 the a(9)=4 solutions are 3 + F_5 + F_3/2, 3 + F_3 + F_6/2, 5 + F_4 + F_3/2, 7 + F_2 + F_3/2.
MATHEMATICA
PQ[m_]:=m>2&&PrimeQ[m] RN[n_]:=Sum[If[PQ[n-Fibonacci[3x]/2-Fibonacci[y]], 1, 0], {x, 1, Log[2, n]+1}, {y, 2, 2*Log[2, Max[2, n-Fibonacci[3x]/2]]}] Do[Print[n, " ", RN[n]]; Continue, {n, 1, 50000}]
CROSSREFS
KEYWORD
nonn
AUTHOR
Zhi-Wei Sun, Jan 09 2009
STATUS
approved