Although my parents were musicians, they were totally uninterested in the technology of music reproduction. But even as a child I was obsessed with audio, and I would not take no for an answer over the question of our family getting a “stereo.” This involved replacing our one-driver radio/record player with a two-driver stereo rig—one speaker driver for each channel. Still I longed for the multidriver Jensens, Electro-Voices, or ARs that to my mind stood for “hi-fi. ”
Looking back I realize that, while the single-driver speakers were colored and had no deep bass and no real top-end, they did offer a kind of realism that the (usually) incoherent multi-driver speakers lacked. And did those one-driver stereo speakers ever image. Even then I knew that was special. Just pull out the retractable wires from that “suitcase” Motorola—with detachable speakers, of course—and a vast sonic picture unfolded. What I didn’t know then was that the convincing stereo imaging was not in spite of, but because my parents’ parsimony forced those single-driver speakers on me. When the time came for me to buy my first “serious” speakers, I ended up with the Spendor BC-1. It had more than one driver, but with the crossover point at 3kHz the BC-1’s bass/midrange driver covered most of the frequency range. Much of the stereo magic was back, and everything else was of course vastly better than the old suitcase Motorola. Even my parents liked the BC-1, and yet the magic was not quite complete.
After these alarms and excursions I arrive at the heart of the matter—stereo works ideally with point- or line-source speakers (or planar radiators so large that one is hearing a plane wave at all frequencies, which seldom happens). Multi-driver speakers just never quite do it. At some level, the ear/brain is always whispering “not quite real” to anything else. It is not only not nice to fool Mother Nature, it is basically impossible.
This long preamble is a way for me to explain why the single-driver Rethm speaker is interesting. In some obvious ways it is not really competitive with multi-driver speakers. There is no deep bass, and response is already weakening at 150Hz. There is little upper treble, and what’s there is quite irregular. The midrange is far from uncolored. And yet, and yet…at some level, the Rethm sounds real, more nearly real than multidriver speakers tend to.
The $7500 2nd Rethm is built around one of the Lowther drivers, which, though they’ve been around forever, remain rare and peculiar. The underlying idea is that in order to make a single driver that covers the whole frequency range with high sensitivity— Lowther started when all amps were lowpowered tube types—what you need is a huge magnet driving a lightweight diaphragm. Fair enough. Except for high voltage in place of high magnetism, which is an electrostatic, it’s hard to think of how else to do it. But of course the diaphragm can’t remain rigid. Above a certain frequency it is essentially operating as a non-pistonic bending-wave driver. And—say it with a British accent, please—it is hard to make a bendingwave driver without resonances, don’t you know? And the Lowthers are peaky. Rethm’s designer, Jacob George, has worked hard with modified “whizzer” cones to smooth things out. And he has smoothed them a good bit. Without this whizzer assembly—it’s removable—the driver is much peakier. Even with it, we are not listening to a Spendor or Harbeth, although off-axis the Rethms are a lot smoother than on. I found the speaker to sound much better when EQ’d to pretty flat—still no deep bass and rolled off in the high treble, but pretty flat in between. It may seem weird to EQ a speaker where signal purity— there’s no crossover—is of the essence, but whatever works.
Equalized or not, you’re getting something unique here. Try some Blumlein stereo recordings from Opus 3, Sheffield, or Water Lily Acoustics. The stereo image is so good it’s scary. This is not a matter of “soundstage.” Rather, it is that single instruments and voices have an enhanced sense of reality because they are in fact exactly somewhere, or are perceived so, just like real sound sources. Maybe this seems unimportant from a musical standpoint, but I think that would be a misconception. Sound that is almost but not quite located is fatiguing in my experience. It’s as if, by instinct, one is constantly trying to figure out where things are, but not quite managing to do it. There is a good reason why many people find mono with one speaker more relaxing than (most) stereo. The location search is turned off.
People who admire Lowther-based speakers often speak of their dynamic liveliness. I find it hard to separate this effect from some combination of their peakiness on the minus side and their complete coherence on the plus side. I can’t really explain it. After all, every broadband speaker has a short “rise time.” Yet transients do seem very lively on the Rethms. And the super-high sensitivity (over 100dB/1W/1m) Rethms were flying with the 2.5Wpc SET that designer Jack Speares made for me. But curiously, the Rethms worked best with my Bryston 14B solid-state amp. At 500Wpc the amplifier must have hardly noticed that the Rethms were even hooked up, but apparently the first fraction of a watt of the Bryston is just as superb as the later ones.
Another aspect of the Rethms singlepoint imaging characteristic is the stability of imaging as the listener changes height. Stand up and the stereo image is still right there. Small changes in tonal balance occur, but the image remains rock-solid and immovable. This makes things seem that much more realistic.
But even more important than the stability and solidity of images—compelling as they are—is that the speakers allow for extremely wide lateral placement. The single-point nature gives centerfill without loss of coherence when the speakers are placed 90-degrees apart (the theoretically ideal Blumlein positioning), or even the 120-degrees that John Dunlavy used to use in his CES demos. (Dunlavys with their symmetric driver arrangement were exceptionally coherent and acted as a quasi-point source remarkably well). High-enders like to talk about outside-the-speaker images—it’s a sort of audiophile obsession—but the most convincing stereo to my mind comes when stage width is provided by speaker separation. Only when images are between the speakers can all the audible cues to location, including amplitude, agree. The Rethms are superb in this regard, every bit as good as the Dunlavys used to be. With a good recording, a remarkable impression of the actual spatial character of an orchestra can be obtained.
One naturally expects the Blumlein theoretic speaker position to work with Blumlein stereo recordings, but the Rethms will do this with well-done multimiked recordings, too. The Audio Plus Rachmaninoff Symphonic Dances (Mata conducting Dallas) sounds amazingly authentic, spatially. While the Rethms do not conceal the microphone patterns, neither do they fling them in your face. This is a great recording, and it sounded like it here. The Rethms also brought to life older recordings, jazz monos from the ’50s and the like, recordings from an era when the recording process was more direct than it has been most of the time since. Duke Ellington’s band has seldom sounded so directly in touch with the present.
All told, the 2nd Rethms present something of a paradox. They are quite large and visually conspicuous and yet they have little bass. They are moderately expensive and yet they lack the high treble extension of, say, Infinity speakers that cost less than $500 the pair. At first sight, one might be simply tempted to dismiss them. And yet, they do things that very few other speakers will do anywhere near as well. If you were to judiciously add a subwoofer (and for me, a bit of equalizing), and could give up the extreme top end—well, then you’d end up with a system with a full, deep sound and with the startling stereo of a true single pointsource with no crossover (aside from the deep bass crossover to the subwoofer).
No one will accuse me of underrating the importance of flat frequency response. If anything, by stressing the fundamental importance of this I’m a bit far-out by high-end standards. But times have changed. Flat response is easy in the DSP world. It may sound peculiar to combine DSP correction with a driver technology older than most of the people reading this, but progress can make strange companions. In a world where DSP makes flat response easy, the character of a driver as a radiation source can be more important than the details of its broadband response. It may seem ironic, but the DSP revolution could bring new life to the old driver technology.
This is not to suggest that the 2nd Rethm cannot exhibit its virtues without some sort of response corrections. As noted, it is much smoother off-axis than directly on, and properly positioned may satisfy all but the BBC-neutrality maniacs of the world. (In this case, it takes one to know one.)