Convert the initial value to a binary number, then perform operations on each bit of the binary number (including the sign bit): change 0 to 1 and 1 to 0. The result is the two's complement of the final value. To convert it back to the original code, take the two's complement again, and you get the computed result.