Isn't it simpler (and more efficient) to build a interpolating polynomial approximation for sin(x) for the range [0,pi/8) (using Chebichev iso Lagrange interpol

Isn't it simpler (and more efficient) to build a interpolating polynomial approximation for sin(x) for the range [0,pi/8) (using Chebichev iso Lagrange interpolation fe)