Subtract Two 8-bit Numbers

A number sits at 2000H and another at 2001H. Subtract the second from the first and leave the difference at 2002H.

Addition was symmetric: a + b and b + a give the same answer, so getting the operands the wrong way round did not matter. Subtraction is not symmetric, and that changes three things:

1. Order matters. SUB always computes accumulator minus operand. The number you are subtracting from has to be the one in A.
2. The carry flag means something different. In an addition, carry set means the answer was too big. In a subtraction it means the answer went below zero — a borrow. There is no separate borrow flag; the 8085 reuses carry for both jobs.
3. The result wraps. 25H - 4AH is negative, and eight bits cannot hold a sign. The accumulator keeps the low eight bits of the true answer, and the borrow is the only record that anything was lost.

You will do the subtraction first, then make the program react to the borrow instead of merely producing it.