Implicit conversion from float -101.48148148148147 to int loses precision