Music in the Round #99: Revel & exaSound

There is necessity as well as comfort in having a long-term reference recordings and, system. The necessity derives from the familiarity with the reference that allows for comparisons and contrasts with the equipment being tested. The comfort that comes from the familiarity lets me relax and enjoy recreational music, relieved from the need to focus my attention intently on the sound. I do relish getting my hands on lots of interesting audio equipment and getting to play it in my own home, but it's like a two-month one-night stand: The new stuff usually goes back even if I am impressed. I don't change my audio equipment often.

It is not until something truly disruptive comes along that I think about making changes. This occurred with the Benchmark AHB2 stereo power amplifier. After the initial review, I returned the amp and tried to move on but could not. Memories of what it did with the midrange of my Bowers & Wilkins 800 Diamond loudspeakers haunted me, and I had fantasies of what a trio of monoblock AHB2s would sound like.


I had already moved on to the B&W 802 D3 Diamond speakers and was in the process of auditioning other speakers when the three AHB2s arrived. I found that their midrange transparency and absence of noise was addictive on every recording I tried. I knew I had to keep them, so I bought them.

The next disruption was less an event and more of a process. Since my adoption, in early 2016, of the B&W 802 D3, I have reviewed about a dozen loudspeakers in the same room, but recently some of them have been so unusual that the mere experience of them has provoked pointed comparisons and considerations. Things seem to be changing.

The first of these speakers was Bang & Olufsen's technology tour de force, the BeoLab 90, a huge, floorstanding, DSP-enabled, powered loudspeaker. These had it all: balanced tonality, extended and detailed bass, notable independence from room influences, and the ability to play with aplomb at any demanded volume. They made the 802 D3s (and every other speaker) sound ordinary, but even if I could afford a trio of these, three would not fit in my room. Still, they demonstrated a challenge in terms of how a speaker interacts with room acoustics.

Next up was a real disrupter, the Kii Audio Three, a stand-mounted, DSP-enabled, powered loudspeaker. To an amazing degree, the small Kii Three was as successful as the BeoLab 90 in freeing the speaker's performance from the influence of room effects. In fact, because of its size, there was greater freedom in placing it to optimize imaging and soundstaging, while each B&O could go in only one spot in my room. Both models advanced the growing argument in my mind that conventional speakers will be facing fierce competition if the DSP/powered speaker paradigm becomes generally accepted in the audiophile market.

After that came the comparatively conventional Paradigm Persona 5F loudspeaker. I was impressed with its excellence on wide–dynamic-range recordings and, particularly, its low-frequency performance. In reviewing it for Stereophile, I drew a parallel between the between the Persona 5F's performance and "speakers using DSP to minimize boundary interactions, such as the Kii Audio Three and the Bang & Olufsen BeoLab 90. The Paradigms accomplished much the same thing but without the bells and whistles." However, its rising high-frequency response, distinct even to these aging ears, demanded tinkering with the toe-in to find an off-axis response that would sound flatter.

And then I visited Harman, where I enjoyed extended discussions with designer Kevin Voecks and other Revelers, as well as with author and Harman consultant Floyd Toole, whose book, Sound Reproduction: The Acoustics and Psychoacoustics of Loudspeakers and Rooms, Third Edition (Focal Press, 2017), should be required reading for anyone with a serious interest in audio. Their premise, backed up by blind listening tests, is that a speaker system that has a flat and smooth on-axis frequency curve and has off-axis output that is similarly smooth (measured in an anechoic chamber) will be minimally affected by the boundary reflections in a real listening room (footnote 1). This, of course, is one of the raisons d'être of DSP speakers, whose radiation pattern is determined through the use of dynamic signal processing and not just by the physical configuration of the drivers and enclosure.

I returned to New York to begin my audition of the Revel Performa F228Be, the latest product of the Revel philosophy. It was, pardon me, a revelation as a full-range passive loudspeaker. Placement was uncritical, as was toe-in, as long as the speakers were not very close to side walls. Audible resolution was strikingly detailed but without noticeable highlighting. In a real room—my room—they seemed to approach what the DSP speakers do.

What now?
Cards on the table: I was facing a critical choice. Do I continue to listen to traditional loudspeakers, or do I make the shift to DSP? The Kii Three is really appealing and affordable, but because of the uncertainty of its suitability for multichannel and its somewhat limited bass output power, I could not get myself to pull the trigger. Kii has addressed the latter issue with the release of their BXT bass module, which turns the stand-mounted Three into a floorstander, although at significant additional cost. The Dutch & Dutch 8c active loudspeaker ($12,500/ pair; see the review in the August 2019 Stereophile) is even more appealing in price and performance, but my wife and I dislike bulky boxes perched on stands. Imagine 5 of them!

Beyond that, I have a concern about service and maintenance for complex integrated products. With conventional speakers and amps, I can find backups if a device requires service. With these integrated products, would small part failures necessitate returning the entire speaker to the factory Will the firmware supplied with the speaker support changes in music formats and delivery systems And given the reliance of such products on proprietary operating firmware, how can they survive if the original manufacturer does not

Or maybe I'm just chicken. I try to convince myself that going to a DSP-based integrated system would complicate my reviewing responsibilities, and there's a germ of truth in that. More critical for me is that managing multichannel playback is complicated enough when using devices and software that directly support multichannel; I'd rather not rely on devices from companies that are not yet addressing multichannel.

I filtered the wide range of options by insisting that any new speaker of mine must be a three-way design that's up to 50" tall; floorstanding; full-range; and visually pleasing in our living/listening room. The latter encompasses personal preferences for smooth contours, a slim shape, and a high-quality wood finish. I also insist on the availability of horizontal and vertical dispersion data from reviews (including John Atkinson's reports) or from the manufacturer. It is surprising how few manufacturers will provide this data even though any serious loudspeaker company should have it.


If you could start all over . . .
After months of auditioning and soul-searching, I bought three Revel Ultima Studio2 loudspeakers ($15,998/pair) for my front L/C/R trio and a pair of Revel's Performa3 F206 speakers ($3500/pair) for the surround channels. In the case of the Studio2s, it's baffling to think that I purchased a speaker model I had reviewed more than a decade ago, but they were clearly my best option: Nothing else suited so well my auditory experience, aesthetic preferences, and my listening room, even at a significantly higher price. As for the F206, which is the F208's little brother, it fit my room setup better than anything else I found. Besides, Revel designs all their speakers with the same goals—and the measured data supports this.

Footnote 1: This can be visualized in John Atkinson's graphs of horizontal (and vertical) off-axis frequency response and in Harman's Spinorama graphs.

Bogolu Haranath's picture

May be KR could review the new McIntosh M-1254 Class-D, 4-channel amp ($4,500), with the new multi-channel set-up :-) .........

Kal Rubinson's picture

The MI254 does not look like something appealing to me. It has 4 channels while I can see uses for 3,5,7........ Looks like a CI device.

Bogolu Haranath's picture

Their cost seem to be reasonable for their specified power output ........ You could consider two of those for a 7 channel set-up, including height channels (or, side channels) ...... Of course, in that 7 channel set-up one of the channels will not be used :-)

Kal Rubinson's picture

Certainly, it has its uses but I am sticking with 5.1 because there's no significant music repertoire in formats with more channels at this time.

Sebastiaan's picture

Dear Kal, Just like you I am amazed by the sonic qualities of the Benchmark AHB2 (They drive my Magnepan MG20.7's in monoblock). May I ask which gain setting you to use? I was floored by the boost in performance when I set it to the lowest gain-setting. It does require a pre-amp processor be able to provide a sufficient voltage swing. My Emotiva XMC1 with 11Vrms voltage swing can drive it. The Gain switch is worthwhile to explore. Once you think you get this fantastic performance, it even gets better.

Kal Rubinson's picture

Yes, I have written about this option and how well it suits me to use the lowest gain setting with my Audio Research MP1 preamp. The MP1 has oodles of output (>14Vrms) but not so great SNR. OTOH, aside from the improved noise performance, I do not perceive any other sonic changes.

Axiom05's picture

Hi Kal. Are you using the Studio2's with or without the grills? I haven't done any sound comparisons but the grills make the speakers visually neutral which complements their neutral SQ. Cheers!

Kal Rubinson's picture

So far, I keep them on.

Ortofan's picture

... Sescom, which use Switchcraft mini-XLR and Neutrik XLR connectors?

Kal Rubinson's picture

I did not buy them from Sescom but the ones I did buy had the Switchcraft mini-XLRs that failed. The specific fault was inadequate internal strain relief. It may not be entirely the fault of the Switchcraft connectors as much as the assembler's choice of cable.

Now using new cables with Rean mini-XLRs (and Neutrik XLRs), heftier cables and excellent strain relief.