I've got a DCX2496, some OB speakers, a mic and measurement setup, and LOTs of measurements. I've basically optimized my speakers using a few different crossover points, and a few different crossover slopes (and separate EQ of response for each of these). What's odd is that from a measurement standpoint, my LR12, LR24, Butt48, and LR48 all sound different from each other, despite the similar FR measurements.
After a long time of listening and tweaking of each, I find that I like the LR12 settings, the LR24 settings are OK, and the LR48 is terrible sounding. I figured that I just didn't like the sound of steep filters. So I stuck with shallow slopes for a long time. But I've recently played around with the Butterworth 48 slopes on a lark. And they seem to give the clarity and cleanliness of the LR48 slopes, while keeping the musicality of the LR12 slopes. Has anyone else noted this type of difference in sound quality when using an active crossover?