Yesterday I wrote this post on another thread:
https://www.diyaudio.com/community/...l-range-line-array.353658/page-4#post-7042144
This explains why an array is such a different animal once we use it in a room. It takes work, as all parallel planes need some help to get the best out of it.
It also lays bare the inherent flaws of Stereo above say 700 Hz. Yes, that cross talk. If we obscure the cross talk, using mid/side EQ and an ambience back-drop,
we get something wonderful. More clear than it should be (if we watch an array vs a single driver I think we'd all expect the opposite) and with that ambience
we can paint a more ideal picture of a space that we are in. Why? Because robbing that energy off of all parallel planes is the reason we notice the Stereo flaws.
Bringing back some late lateral energy hides that flaw, especially when using the mid/side EQ.
Maybe I should explain using the arrays well below their natural cut-off too. As that was another cause of confusion. "It's simply not good practice, right?"
Well, it wouldn't be if you didn't have the capacity available to do so. But due to the help of the room, the arrays can reach quite low and do so easily with a bit
of EQ (which is dropped by 9 dB from earlier renditions). Throwing away that capacity is a waste of energy that's almost free(*). As it (again) helps to share the load
with the subwoofers to fill in, or average out the room, just like the array does at higher frequencies. (distributed bass)
I think I'll have to get deeper into this subject somewhere soon. As I doubt the idea comes across that easy with words alone.
(*)= and everywhere there is a dip due to room dimensions, I place a PEQ to drop the amount of power going there, and the subwoofer will make up for that loss.
So in the end, nothing gets over stressed and we're only using the strengths and not the weaknesses. The end result is better and measurably cleaner.
(and a lot of work to do so

)