Tater -- You'll get a 'resonant frequency' at even intervals of some integer n times 1/2 the wavelength. The value at 1/2 the wavelength is a nice base to quote so that's what's called resonant frequency. The wavelength of the sound is a function of how fast it travels in the medium that it's giving a pressure wave to.
In air that's the speed-o-sound which we all know at around (I can't remember exactly so help me here) 550mph. Crud. Metric system would really help here but I learned part in one system, part in another. The speed is a function of the density of the medium. It's linearly proportional to that, in fact.
Anyway, obviously if you know the speed the wave travels and its frequency (frequency = how long it takes to make one full sine wave pattern) then you can divide speed/freq = (m/sec) / sec to get the length of the wave in meters.
Am I making any sense here? Am I just telling you something you know already?
You're floor's fixed around its edges (we hope). You want to find the minimum wavelength that makes the middle of the floor go up & down the most. That's the resonant frequency of the floor. At that time when the subwoofer is making the floor move the most, you'll be creating the biggest pressure fluctuation in your room. That also means that the microphone's transducer will be shoved the hardest, and you're hearing the loudest sound for a given amount of power input (power set by volume control). The computer is probably looking for where that maximimum volume is for the very lowest frequency, thus pinpointing your 1/2 wavelength value. That's your resonant frequency, and that's why you need a microphone to gather the feedback.
I hope this helps somewhat. Good luck!
Now if you can help me write a big report on SAN vs NAS...