OFFSET
1,2
COMMENTS
It is not yet known whether there is a finite number of "good numbers" A204142. This is equivalent to the question whether this sequence will eventually decrease to zero.
One has a(100)=408, a(316)=2179, a(1000)=13136, a(3162)=86603, a(10000)=609377.
LINKS
M. F. Hasler, Table of n, a(n) for n = 1..3000
PROG
(PARI) A204580(n)={ my(r=1, s); while( A204142[r]^2+2<=A204142[n], r++); s=r; sum(i=r, n, while(s>1 & A204142[s-1]*A204142[i]+2 > A204142[n], s--); sum(j=s, i, is/*pseudo*/prime(A204142[j]*A204142[i]+2)))} /* From n~3000 on, use of ispseudoprime is significantly faster. The vector A204142 must be already computed. */
CROSSREFS
KEYWORD
nonn
AUTHOR
M. F. Hasler, Feb 11 2012
STATUS
approved
