I thought the function f below, when compiled, will be optimized and will be equivalent to the function g below.
// 4 multiplications and 2 additions. // 2 mu