Just curious, HAL. I have always heard that, if possible, a first-order x-over was the way to go. Did you play with a first-order (6dB/octave) x-over before you settled on a sixth-order? I called BG several years ago and was told that the NEO-8's, especially in multiple arrays, could take a shallower high-pass slope without a problem. Looking at the curves of the cone midranges, I conclude that the cones could be operated higher, such that a first-order x-over could be used as a low-pass filter as well. I'm already running my cones without attenuation at the low end.
So, it seems that a first-order x-over at a higher x-over point is at least feasible, and - - theoretically, at least - - desirable. I'm chicken to try it, but I don't mind your blowing up your own speakers or lab in the interests of science.
Not that I am at all dissatisfied with what I have. My Danny Ritchie 9's are the best speakers I have ever had (many, many really good ones, including KLH-9s and Maggie 20.1s) or heard (a veteran of many shows), but I can't help wanting to mess around with them. "They're coming to take me away, away . . ."