why doesnt the interpreter just do the second thing internally? floating points are weird
If you need actual decimal math you'll use BCD. Floats are for quickly approximating.
Because this isnt solvable for any n_1 + n_2
>>107716422proof