With test tones its pretty noticeable most of the time. There are really cheap o scopes out there that are than good enough for the task, even still i understand its not in everyone budget but that doesn't make it ok to use a DMM. As said before when a sub is being excited its impedance changes ,which is why its called impedance and not Resistance. Also you would have to do this at every frequency you plan to use the amp and then i guess you have to find some sort of equilibrium between all the test points to set the gain. That ohms law formula that everyone uses is completely useless for this application unless you are using an x value resistor that can handle the power, but then that value is still useless because it would mean nothing in the world.Right and wrong here.
Yes its always a good idea to use an oscope to detect clipping. Unfortunately this is not feasible with most people's budget. To get around this you just set the gain at a safe volume around 80-90% of what your headunit will go to. This will set your gain at a conservative level where you won't have to worry about clipping. One thing to consider here is that clipping is not always audible.
I guess the point i'm trying to make is that people should really stop telling other people that this is the "best/right/only" way to set your gain, because in reality all it dose is create a false sense of security that could lead to a greater likelihoods of blowing something up.