Implicit conversion from float -21.042944785276077 to int loses precision