Is there a complex version of this, where the quadratic in q is split into the product of two factors 1 +- q e^(i theta)? If so, there might be other polynomials available as numerators. Rich ------------- Quoting rwg@sdf.lonestar.org:
prod((q^(2*2^-n)+2*cos(theta/2^n)*q^2^-n+1)/4,n,1,inf) = (q^2-2*cos(theta)*q+1)/(log(q)^2+theta^2)
- n - n 2 2 theta 2 inf q + 2 cos(-----) q + 1 /===\ n 2 | | 2 q - 2 cos(theta) q + 1 | | -------------------------------- = ----------------------- | | 4 2 2 n = 1 log (q) + theta
--rwg
_______________________________________________ math-fun mailing list math-fun@mailman.xmission.com http://mailman.xmission.com/cgi-bin/mailman/listinfo/math-fun