Possible Duplicate:
Can Reduce *really* not solve for x here?
Why can't Mathematica solve this equation?
f[x_] := Exp[Cos[3*x]]
g[x_] := (1/3)*x^3 - x^2 + 2
Solve[f[x] == g[x], x]
(* Solve::nsmet: This system cannot be solved with the methods available to Solve. >> *)
Solve[f[x] == g[x] && Abs[x] < 10, x, Reals]. For more detailed discussion of transcendental equations see e.g. http://mathematica.stackexchange.com/questions/4694/can-reduce-really-not-solve-for-x-here/4697#4697, a general issue ofRootsee e.g. http://mathematica.stackexchange.com/questions/13767/how-do-i-work-with-root-objects/13768#13768 – Artes Dec 15 '12 at 21:17xto be real, but then the system might be unable to prove that there were all solutions found. AssumingRealsone gets5solutions, while assumingComplexone gets385solutions. – Artes Dec 15 '12 at 21:23