Two methods are discussed for solving nonsmooth equations. The first method, a parametrized Newton method, uses a damping parameter for the Newton step and a regularization parameter for the chosen member of the generalized Jacobian, and, therefore, is well-defined even when the generalized Jacobian is singular. The second method is a Broyden-like method based on a so-called point-based smooth approximation function, which generalizes the technique of splitting the nonsmooth function into a smooth and a nonsmooth part. For both methods local linear and superlinear convergence results are proven. Numerical examples are given for four nonlinear complementarity problems from literature. The numerical results are compared with other methods for solving nonsmooth equations.