I'm obviously a fan of Be tweeters--I've done several designs for Jim using the SB Acoustics version, including the 9.5. It measures flat as a pancake, can be crossed very low even without a wave guide, and has super low distortion. it would be my tweeter of choice in a 2-way. I haven't heard the 9.5 in stereo, but my single prototype is clear as a bell. On the downside, its horizontal dispersion isn't great at higher frequencies. You can see that if you compare the Spinorama plots for the Revel and the BMR. Of course, part of that difference is due to the Revel's wave guide. Speaking of which, I'm still not convinced wave guides are the best way to go. They almost inevitably narrow dispersion and often cause response irregularities on axis. These usually are smoothed out across the listening window, and that's certainly an important characteristic. However, those plots don't tell you how the brain processes the differences in arrival times and frequency response of direct vs. reflected sound. My experience has been that significant peaks and dips in the direct sound are audible even if later arrivals are smoother.
My best guess is that if you compared stereo pairs of the BMR and Revel, you would notice the latter's on-axis peak at 5 kHz, particularly if they were toed in (I don't remember what Harman's advice is on positioning). I'm also pretty sure there wouldn't be much agreement about which speaker sounded better. A modest peak in that region can add sparkle on some music, and wouldn't be noticed on other program material. The other difference that should show up is in sound staging. I'm sure some people would appreciate the very wide and deep sound stage created by the BMR's broader dispersion, particularly on non-studio recordings in good venues. Others might prefer what I suspect is a more focused from the Revel. I could turn conjecture into a more informed opinion if Harman would ship me off a pair of Revels free of charge. I'll get my contact info out to them immediately.
