Sure.
The crossovers that I design are not just optimized for on axis response like many designers will do.
I allow for a couple of other fixed variables. There will be a floor and a fixed distance to the floor. The floor is also more often than not carpeted. There is a semi fixed distance to the ceiling and it almost never has any treatment on it at all.
Listeners never listen below tweeter axis (head on the floor). However, they are often standing in the room or moving around.
A dip in the response due to an out of phase relationship of the drivers (distance offset) seen by moving the ear or mic to the floor has less effect on the overall room response than a dip in the vertical response (above the speaker). A dip in the vertical off axis is more present and more easily seen in the reflections from the ceiling and causes more of a disruption in the over all in room response.
All that being said....
I obviously design to obtain an even in room response. So I try to maintain good phase relationships not only on axis but also above. Doing this will result in out of phase cancellations as you drop below the tweeter level.
So if you keep the same crossover, but flip the speaker over, the on axis response is the same but the room response changes.
Make sense?
