Implicit conversion from float -1.9090909090909065 to int loses precision