7 Mar
2011
7 Mar
'11
4:43 p.m.
The absence of any really big terms from Neil's 458M means that, at around m=10^900000000, the sums 2 Tan[n Pi ] Sum[------------, {n, -m,m}] == n (1 + n Pi) 1 Cot[- - k] 2 Sum[-(----------), {k, -m,m}] ~ 9.71628672863 1 2 -(-) + k 4 (take the limit Pi^2 for the n=0 term) will agree to > 600M digits. However, no one has managed to prove that these sums even converge, let alone to 9.7162867286... --rwg