The problem with the new Trinnov room correction software is that it doesn't use a single mic. Up until now the setup/correction software has just been concerned with frequency and time. Only the distance to the speaker was calculated. So they setup software would send out a click to each speaker and time how long it takes for the front edge of the sound pulse to reach the mic. That first blip would be considered the shortest distance from the speaker to the mic. Any additional sound would be from reflections.
Trinnov uses a 4-mic capsule. It doesn't just measure the time it takes for the rising edge of the click to reach one mic, it also uses the delay between each mic in the cluster. The mics are arranged in a 3D pattern that when knowing the slight delay between each one, you can compute the distance, angle around, and elevation above the measuring position. Trinnov uses this to virtually remap the speakers position in space (by mixing content into neighboring speakers with phase and delay adjustment). The problem comes when a speaker has more than one high frequency driver, and even more so when they are pointed in different directions. The mic cluster gets two rising edges, with a very small delay between them, and even worse each tweeter may be "heard" first by different mics. So the calculations are completely thrown off. The Trinnov processor either can't locate the speaker in space, or guesses its location completely wrong.
