32 bits.

Here is an addition problem using 4-bit operands:

1111 011110010000 Overflow happened

Two four-bit numbers are added,
but the sum does not fit in four bits.
If we were using five bits the sum would be
`1 0000.`

But with four bits there is no room
for the left-most "1".
Because the carry out from the most significant
column of the sum is "1",
the 4-bit result is not valid.
The column is called the *most significant* column
because it corresponds to the highest power of two.
The bits in the leftmost columns are called the *most significant bits*
or the *high-order bits*.

The electronic circuits of a processor can easily detect overflow of unsigned binary addition by checking if the carry-out of the leftmost column is a zero or a one. A program might branch to an error handling routine when overflow is detected.

Add these unsigned numbers, represented in eight bits. Determine if overflow occurs.

0010 11000101 0101