Move from FP register to integer register : Copies the raw 32-bit bit-pattern stored in FP register $f1 into integer register $t1. Useful for reading a single-precision float's bits or the result of 'floor.w.s', 'cvt.w.s', etc.