Reprinted with permission from Stereophile Magazine, Vol. 19 No. 4.
Equipment Report
Audio Artistry Dvorak Loudspeaker System
By Shannon Dickson
I first heard Audio Artistry's Dvorak loudspeaker during the 1994 WCES. I was leaning against the wall in a corridor of the Sahara Hotel's bi-level complex, trying to avoid being run over by swarming hordes while shootin' the breeze with Corey Greenberg. I was thankful for the respite from what seemed to be an endless succession of rooms playing Eric Clapton's "Tears from Heaven." As happens at most CE Shows, a particular tune emerges as the "official" demo track and gets saturation coverage. "Tears" was definitely the one for the '94 Hi-Fi lovefest. It's a great song -- but after hearing it 30 times in a row, you do start to feel a little tight around the collar.
Chatting with Corey, on the other hand, is always fun, so when a nearby door opened up, filling the hallway with the very same song, you'd have though I'd hardly notice. On the contrary, I immediately lost my train of though, said "Aloha" to Corey, and drifted into the room like a Stepford wife heeding a subliminal message. By the last refrain of Clapton's poignant ballad, I felt like I was hearing the song for the first time. During the next hour's demo of classical, jazz, and rock tunes, I was genuinely enthralled by the way these speakers conveyed the essence of whatever music they were reproducing.
The impression of that first experience was so strong and persistent that I just had to see for myself if this speaker really did possess a unique and compelling communicative skill.
artistry
Audio Artistry is located in a suburb of Raleigh, NC and was formed almost three years ago by president Marshall Kay, a test and measurement applications engineer for Hewlett-Packard specializing in the application of HP's wide variety of electronic test equipment. Marshall teamed up with Kurt Pasquale (an expert in computer-aided design) and Tom Hoffman (owner of a local high-end retail store) to build, refine, and market a line of speakers designed by co-partner Siegfried Linkwitz. Mr. Linkwitz is a senior design engineer for Hewlett-Packard [see the interview elsewhere in this reprint], and is the same Linkwitz of the widely used Linkwitz-Riley crossover topology.
the challenge
I'll paraphrase the introductory abstract found in Linkwitz's 1992 AES paper, in which he describes the prototype for the Dvorak in great detail: "A relatively small-sized dipole loudspeaker system has been developed using conventional cone-type drivers to obtain sufficiently large volume displacements. The 3-way system has dipole directional characteristics over the 20Hz to 1.7kHz frequency range for reduced interaction with the listening room. Effects of baffle shape upon the radiation pattern have been investigated. Active crossovers and dipole specific equalization have been used to obtain a flat frequency response." (My emphasis.) This rather curt, matter-of-fact description contains a wealth of information about the Dvorak's unusual characteristics.
The inspiration for the Dvorak arose from two principal factors. First, about nine years ago, the designer's reference speaker consisted of two small satellites and a central subwoofer. This system was capable of good dynamics, image specificity, and fine detail, as well as an even midrange/treble tonal balance and decent reproduction of instrumental timbres. But the design suffered from limited image height and a soundstage width restricted to the area between the two satellites. Additionally, no matter how small and dead the enclosure was made, some boxy coloration was inevitable due to resonant energy storage and its subsequent delayed release. Most important, all traditional box speakers radiate low frequencies in an omnidirectional pattern, something that Linkwitz felt produced unacceptable masking and colorations.
Faced with these shortcomings, Siegfried received his second impetus for the Dvorak design in 1986, when he designed a sound-reinforcement speaker for audio-video presentation in a large, highly reverberant gymnasium. In order to improve speech and music intelligibility in this environment, he built a long, vertical line-source dipole consisting of twelve 6" cone drivers with the drive-signals electrically tapered to concentrate high frequencies in the two center drivers of the array. In spite of concrete walls and a large parquet wooden floor, this system produced excellent clarity and speech recognition even in the rear of the gym.
As an experiment, Siegfried then broke these dipole columns into a stereo pair and set them up in his living room. Despite the speaker's skewed frequency response, the sound was surprisingly open and produced outstanding image height and width, as well as a marked reduction in room interactions through the midrange and bass. These findings inspired a major revision in his thinking -- away from the semi-point-source monopole designs popular for home playback and toward a moving-coil speaker that would maintain dipole directivity from the midrange through the lowest frequencies. Numerous conversations with acoustician Dr. Brian Elliott, and subsequent work by Elliott on the challenge of building an effective dynamic dipole subwoofer, dovetailed nicely with Siegfried's efforts from the midband upward. The full-range Dvorak began to take shape.
dipoles & monopoles
In order to better understand the Dvorak, we must examine the basic relationships between panel dipoles, moving-coil box speakers, and the Dvorak's successful combination of the two principles -- with a particular emphasis on room/speaker interactions. (Note: The attributes and limitations detailed below are relative comparisons between the various major speaker types I'm familiar with, rather than judgments rendered against an absolute yardstick.)
We are still a long way from the perfect transducer. All sorts of complex intermodulation and harmonic distortions still exist which -- along with limitations in driver technology, diffraction problems, crossover-related dispersion anomalies, and speaker-room interactions -- plague even our most advanced designs. Not to mention the drawbacks of two-channel stereo. Please keep this in mind when interpreting the following comments.
Although a thorough examination of room acoustics is beyond the scope of this article, I'll be referring often to three fundamental interactions that each play a large role in what we actually hear from every speaker design: 1) Room resonances typically affect the region below approximately 200Hz and result from standing-wave excitation of room modes. 2) Initial reflections from room boundaries often negatively affect imaging and the resolution of transient detail. 3) Reverberation refers to longer-duration sounds resulting from cumulative reflections, and affects timbre principally above 200Hz. Let's look first at how these phenomena impact dipoles.