Analyzing Incorrect Results in Binary Subtraction

[btbam91]

I am asked to come up with two 4-bit binary numbers, that when subtracted, provide the wrong result for both the unsigned and signed cases, inspecting only the first 4 bits of the output.

I have come up with numerous combination where the unsigned is correct, the signed is wrong, and vice versa. But I can't get both of them to be wrong!

Thank you for the help!

 

[Simon Bridge]

Can you express the conditions that the operation gets the wrong result, in general terms, for each case?