MartinLogan Prodigy loudspeaker

Larry Greenhill, July, 2001

Electrostatic speakers are my passion. Why else have I put up with their high prices, unreliability, low power handling, tendency to arc, high-frequency beaming, limited bass response, and widely fluctuating impedances?

Take the Quad ESL-57 (footnote 1). Configured as a gnarly, stubby panel, it was often mistaken for an oversized space heater. Left on, it sucked up dust and grime as its panels filtered the air. Far worse, it arced—those flickering blue lights in the panel signaled diaphragm perforation and eventual destruction.

But the sound! When I sat in their tiny sweet spot, my ESL-57s had gobs of air, liquid transparency, neutrality, pinpoint imaging, lightning-fast transient response, wide and deep soundstaging, and low distortion, all with no listener fatigue. Their midrange—far more liquid and transparent than that of Quad's 1981 ESL-63, in my opinion—was the most natural I have heard from any loudspeaker, and justified the post-traumatic stress disorder my '57s caused me.

So I was both excited and uneasy when a review pair of MartinLogan's new hybrid Prodigys arrived in my listening room. I needn't have worried—Gayle Sanders, MartinLogan's president and chief designer, has spent his career making his electrostatic designs safe and reliable for audiophiles like me (see Wes Phillips' May 1997 interview with Gayle).

The Prodigy is imposing: 5' 7" tall, 16.5" wide, and more than 2' deep. Its two sections—the electrostatic panel and the chunky, back-busting bass module—couple to form a 133-lb unit. The Prodigy's technology—trickled down from the company's flagship, the $70,000/pair Statement Evolution2—includes a proprietary electrostatic transducer and a woofer array, called "ForceForward" by MartinLogan, that is intended to reduce bass room interactions and resonances. MartinLogan uses a proprietary manufacturing technique to curve the Prodigy's perforated, metal-framed panels into a "curvilinear line source" (CLS) transducer. This frees the loudspeaker from the narrow dispersion pattern common to most electrostatics.

MartinLogan's CLS screens are arc-proof. Sanders and his engineers determined that panel arcing was due to local buildup of charge from uneven coatings on the 0.0005"-thick polyester diaphragm. While other manufacturers hand-paint diaphragms with a conductive material, MartinLogan vapor-deposits the conductor as a 1-angstrom-thick (0.0000000001m) layer onto the diaphragm's surface. This process produces an optically transparent conducting layer with uniform characteristics of surface resistivity that prevent charge buildup and migration, and arcing. According to Sanders, this layer has the electrical surface conductivity of human skin, allowing it to hold a static charge of from 1500V to 3000V.

The Prodigy's diaphragm is sandwiched between perforated steel plates, called stators, that carry the varying audio-signal voltage while providing insulation to protect the user. A nylon-based powder insulation is deposited at very high temperatures on the perforated steel stators to provide a safety tolerance of 12kV. Fourteen evenly spaced ClearSpars, or horizontal Lexan insulators/spacers, subdivide the panel and create a uniform gap between the stators and the diaphragm. When the amplifier's audio signals are stepped up to high voltages by a transformer and applied to the stators, an electrostatic field of opposing voltages produces forward and backward movement of the diaphragm, which in turn generates sound waves in air.

One problem with electrostatic speakers has been the tendency of the charged membrane to collect dust. However, the Prodigy's diaphragm accumulates dust only slowly because it automatically shuts down its electrical field when no music is playing. When dust has accumulated, the manual advises the owner to leave the speaker unplugged for 12 hours, then vacuum the panels.

The Prodigy's second-order crossover switches the signal from the bass module to the electrostatic panel at 250Hz. It uses point-to-point wiring and physically separated high- and low-pass crossover circuits to reduce magnetic interaction. The panel driver's dual transformers are said to result in lower distortion and increased power handling.

---

Footnote 1: I found that the original Quad ESL-57 ("57" refers to the year of their design) sounded their best when driven by Mark Levinson ML-2 monoblocks, which cost $8000/pair in 1980. This classic solid-state amp was one of the few designs that remained in true class-A operation over its entire operational range into any load—including the Quad ESL-57's impedance of less than 1 ohm in the treble frequencies. Such inefficiency meant that the ML-2s put out 25W as audio and 300W as heat. Subjected to such constant cooking, early ML-2s suffered frequent output-transistor failure, pushing their owners further out on audio's bleeding edge.

The bass module is a sealed dual-woofer design. The forward-facing 10" ScanSpeak aluminum-cone woofer carries the bass audio signal. An independently enclosed, 10" ScanSpeak fiber-cone cone faces to the rear and is driven by a circuit that processes level, phase, and frequency range to create a partial cancellation of the front woofer's output in the lower frequencies (footnote 2). This ForceForward technology specifically reduces the common 50Hz null and 100Hz peak found in many rooms. For those requiring more forward bass, the Prodigy has a rear-panel switch for increasing the below-50Hz response by 3dB.

The wing nuts on the Prodigy's speaker binding posts can be easily tightened without pliers or a wrench. Four binding posts are supplied on each speaker, allowing for biwiring. The excellent owner's manual gives instructions for horizontal passive biamplification (use a tube amplifier for the electrostatic panels, solid-state for the bass modules) and vertical passive amplification (two identical stereo amplifiers, one dedicated to each speaker). Spiked feet—called Energy Transfer Coupler (ETC) Spikes—are supplied with jam nuts to allow for easy leveling.

Setup
I placed the Prodigys where the Revel Ultima Salon loudspeakers had sounded best: 63" from the rear wall and 36" from the side walls, sitting out in the room on a circular area rug. The speakers faced the long axis of my narrow listening room, which is 26' long, 13' wide, and 12' high, with a semi-cathedral ceiling. One long wall is covered with bookshelves, the other has a bay window. At the opposite end, the room opens into a 25' by 15' kitchen through an 8' by 4' doorway.

An electrostatic speaker needs to be plugged into an electrical outlet. When playing, the speaker displays a blue light—in the shape of the MartinLogan logo—through its electrostatic screen. Unlike the blue arcing in my old Quads, this light can be switched on and off from a pushbutton in the bass module.

Because the Prodigy's rated voltage sensitivity of 91dB/2.83V/m lies well above the average range of B-weighted speaker sensitivities reported in Stereophile, I found that I needed to set my Krell KBL preamplifier's volume control significantly lower than usual.

I first drove the Prodigys with the high-powered Bryston 7B-ST monoblocks—tested at 954W into 4 ohms—because of the 7Bs' power reserves and easily visible front-panel clipping indicators. Configured in serial bridged mode, the Brystons' bass response was very tight, deep, and fast, their soundstage huge—the bass lines on "Unfinished Sympathy," from Massive Attack's Blue Lines (Circa WBRX2), were solid and full. However, I found the soundstage depth shallow and the transients harsh. I switched the 7B-ST into their parallel bridge mode—tested at 595W into 2 ohms—which enables the amplifier to drive low-impedance loads. Although the bass became less prominent, the soundstage assumed a more natural depth and width and the edginess disappeared. Even so, I preferred the less powerful Mark Levinson No.334 to the Bryston for its sweeter, more transparent sound.

Final adjustments included comparative nearfield (8') and farfield (16') listening, low-frequency signal-generator sweeps, phase checks, pink noise, and fine-tuning of listening position for optimal soundstaging and imaging. The Prodigy's deep-bass output fell off smoothly between 41Hz and 35Hz in my listening room, with no doubling. Playing Stereophile's Test CD 3 for channel checks and phasing, I carefully positioned my chair in the Prodigys' nearfield until I could hear the in-phase pink-noise signal as a centrally focused sonic image. Soundstaging was optimized when speakers and chair described a 7' equilateral triangle (measured from the panels' approximate centers).

The Prodigy's electrostatic panel extends from 31.5" to 60" above the floor. This covers an area below my ear height (38") when I'm seated in my listening chair and extends above my ear height when I'm standing. This explained why the speaker's tonal balance didn't change when I stood while playing pink noise.

I broke the Prodigys in by playing music from an FM tuner for 12 hours, followed by 12 hours of the "Special Burn-In Noise" from Stereophile's Test CD 3.

Sound
I first put on some warmup music—David Hudson's Didgeridoo Spirit (Indigenous Australia IA2003 D), Fleetwood Mac's The Dance (Reprise 46702-2), and David Bowie singing "Putting Out Fire," from the Cat People soundtrack (MCA MCAD-1498). The Prodigys' transparency and deep, wide soundstage quickly involved me in the music. But at first, the bass response seemed understated, shy, and reticent.

After a month of listening, things had changed: The louder I played the Prodigy, the better it sounded. [Stereophile webmaster Jon Iverson also reports that the Prodigy requires a significant amount of break-in.—Ed.] Its bass blossomed when driven hard. At a level where my usually unflappable Mark Levinson No.334 began to show signs of compression, the Prodigys opened up. The bass became forceful and taut, and the instruments separated and became more distinct.

---

Footnote 2: In his August 2000 review of the MartinLogan Prodigy in Hi-Fi News & Record Review, Martin Colloms pointed out, "If [the two woofers] were both equal and always out of phase you would get dipole upper bass while the low bass would cancel out." MartinLogan gives an example of what happens at 50Hz: "The acoustic difference between the drivers is about 2.3'. At 50Hz, the phase shift caused by the traveling wave is about 37 degrees. When added to the 127 degrees phase shift of the passive crossover, the front waves meet with a relative phase shift of 127 degrees - 37 degrees, which equals 90 degrees. The back waves meet with a relative phase shift of 127 degrees + 37 degrees, which equals 164 degrees. The front waves meeting add at 90 degrees, causing an addition resulting in an amplitude gain of 3dB. The back waves meeting at 164 degrees causes a subtraction, leading to an amplitude reduction of 11dB. The total difference between the two waves is then 3dB + 11dB = 14dB."

At these levels, the Prodigy's bass response sounded detailed and taut, and showed the advantages of the ForceForward design. Solid, clean bass extended down to 31.5Hz in my room when playing a 1/3-octave warble tone at -20dB (Test CD, Stereophile STPH 002-2). The bass drum in Owen Reed's La Fiesta Mexicana was tuneful, solid, and powerful (Fiesta, Reference Recordings RR-38CD). The final organ chords of Part 1 of Elgar's The Dream of Gerontius (Test CD 2, Stereophile STPH 004-2) and the repetitive bass-drum beat in "Cosmos Old Friend," from the Sneakers soundtrack (Columbia CK 53146), were clean, but I had to listen carefully because their reproduction via the Prodigys was subtle. Similarly, Michael Arnopol's plucked double bass on Patricia Barber's "Use Me," from Companion (Premonition/Blue Note 5-22963-2), pulsed, throbbed, and burned with no sign of bloat.

The Prodigy's bass had excellent pitch definition, the woofer's sonic characteristics to mesh well with those of the electrostatic panels. It captured the pounding tom-toms on Richard Thompson's "I Misunderstood," from Rumor and Sigh (Capitol CDP 7 95713 2); and the 32Hz bass notes from the beginning of Strauss's Also sprach Zarathustra, from "Ascent" on Time Warp (Telarc CD-80106), were clean and tight.

The Prodigy's midrange was transparent, timbrally accurate, and free of congestion and distortion. Vocal/clarinet/piano selections came alive as the speakers created a wide, seamless soundstage that gave no clue of their positions in the room. Vocals were reproduced with a transparency I'd missed ever since I'd let the Quad ESL-57s escape from my listening room. Suzanne Vega's startling a cappella vocal on "Tom's Diner" (Solitude Standing, A&M CD 5136) floated in the room with a lifelike fullness. The tonality of the saxophone and guitar were startling on the title track of the L.A. Four's Going Home (Ai Music Co. 3 2JD-10043). The same rich but totally natural timbre was heard in Buddy Miller's mando-guitar accompaniment to Emmylou Harris's "Prayer in Open D" on Spyboy (Eminent EM-25001-2). The guitar work on that track was crystalline and airy, with silken tonality.

The female voice was rendered faithfully, with natural timbre and low distortion. Patti Austin's rendition of Armando Manzanero's "Only You" (Hothouse, N2K 10023) was etched and holographic. Kim Cattrall reading "Little Dog's Day" sounded see-through clear and bell-like on Mark Levinson's Live Recordings at Red Rose Music, Volume 1 (Red Rose Music RRM 1). And I was transfixed by the a cappella choral blend on "Calling My Children Home," from Spyboy. Emmylou Harris's voice was effortless, ethereal, clear, and translucent.

Male vocalists fared just as well, with no midrange anomalies, suckouts, or other colorations. Willie Nelson sounded clear, clean, and totally free of honk or hollowness on "Getting Over You" and "Don't Give Up," from Across the Borderline (Columbia CK 52752). Harry Connick, Jr.'s tenor on "Don't Get Around Much Anymore," from the When Harry Met Sally... soundtrack (Columbia CK 45319), had none of the darkness and over-richness I routinely hear from dynamic speaker systems.

The Prodigy's treble spectrum was smooth and beguiling, with no brightness, steeliness, or metallic edge. Bells heard over this speaker had a magical sheen—as on Dvorák's Carnival, from Nature's Realm (Water Lily Acoustics WLA-WS-66-CD). The Japanese and Korean temple bells that back up Shane Cattrall's reading of Psalm 23 on Live Recordings at Red Rose Music were reproduced with stunning realism, transparency, and detail.

The Prodigys were able to maintain image stability at high volumes. On Going Home, the L.A. Four was precisely positioned on a wide soundstage, playing duets and solos on guitar, double bass, drums, flute, and saxophone. Images snapped into focus on the Prodigys' wide, well-defined sweet spot, whether José Carreras' holographic tenor singing the opening Kyrie of the Misa Criolla (Philips 420 955-2), or Richard Thompson's guitar as heard just outside the right speaker in the instrumental close of "Why Must I Plead," from Rumor and Sigh.

The sharpest and most precise imaging was heard from Sacred Feast, Sony's multichannel Super Audio CD recording of the choral group Gaudeamus (Sony SACD-9). Using a new Sony SCD-C555ES carousel SACD player, I selected a multichannel playback mode that fed full-range signals to the two front loudspeakers with no center channel. The result was a pure, airy, well-defined chorus rich in natural timbres.

Of all the Prodigy's sonic characteristics, the most impressive was its dynamic range. In my large listening room, it was one of the few loudspeakers that did not limit and crunch on the choral peaks of Elgar's Dream of Gerontius. Playing Spyboy's apocalyptic "Deeper Well" at top volume, I could still follow Emmylou Harris's birdlike voice and hear her lyrics clearly, despite the throbbing, churning bass synthesizer and distorted electric guitar.

The Prodigy handled both ends of the dynamic range beautifully, but just to be sure, I decided to push it hard. (footnote 3 I reconnected the Bryston 7B-STs and cranked up the volume until the amps' clipping lights flashed when playing snare-drum rimshots. In this configuration, the Prodigy bettered all loudspeakers heard in my living room. Not only did it reproduce the powerful wall of synthesizer and electric guitar and the explosive rimshots in the drum solo on "The Maker," the showstopper from Emmylou Harris's Spyboy, but it conveyed inner detail that most speakers miss. I heard a transparent, multi-layered aural portrait: a wall-to-wall tapestry of voices, synthesizer, drums, guitar, and crowd noise; the distorted but musical electric guitar solo; Harris's delicate but strained voice; the layered effect of the male backup singers; and those explosive rimshots.

Another revelation came when I played my favorite jazz selection, "The Mooche" from the Jerome Harris Quintet's Rendezvous (Stereophile STPH013-2). The Prodigy reproduced—better than I'd ever heard them—the honky timbre of the saxophone, the blattiness of the trombone, and the luminous, shimmering, see-through clarity of Steve Nelson's vibraphone.

Conclusions
The MartinLogan Prodigy reaffirmed my passion for electrostatic loudspeakers. Sure, I was swayed by the usual electrostatic attributes—low distortion, timbral accuracy, and deep, wide soundstaging that took my breath away—but there was more.

The epiphany came after I'd living with the Prodigys for two months. I don't know why it took that long—perhaps the woofers needed the time to fully break in and reveal their full dynamic range and power-handling capabilities—but music I'd always loved then came alive in a brand-new way. Instrumental timbres and colors became much more vivid, intense, and startlingly realistic. The speakers' dynamic range expanded, allowing them to play louder, with greater depth, three-dimensionality, and detail.

For these reasons, I strongly recommend you audition a pair of well-broken-in Prodigys with your favorite source material. Crank up the volume and listen. I promise you—the MartinLogan Prodigy will be a revelation, and the best cure for electrostatophobia.

---

Footnote 3: Other critics have crowed about the Prodigy's dynamic range. See Myles Astor's review in Ultimate Audio (Winter 2001, Vol.4 No.4, pp.60-64), in which he praised the Prodigy's "ability to reproduce the speed and reduce the smearing of plucked instruments." UA named the Prodigy its Audio Product of the Year. In the December 2000 Home Theater (pp.113-120), Jerry Kindela lauded the Prodigy for its "marriage of micro- and macrosounds."

Sidebar 1: Specifications

Description: Hybrid electrostatic/sealed-box moving-coil loudspeaker. Drive-units: 48" by 16" curvilinear electrostatic HF/midrange panel; 10" long-excursion, aluminum-cone woofer; 10" high-rigidity, high-excursion fiber-cone woofer with cast basket. Bass-control switch: +3dB at 50Hz. Crossover frequency: 250Hz. Crossover type: quasi-second-order, 12dB/octave. Frequency response: 28Hz-22kHz, ±3dB. Dispersion: horizontal, 30 degrees; vertical, 4' line source. Sensitivity: 91dB/2.83V/m. Nominal impedance: 4 ohms. Minimal impedance: 1 ohm at 20kHz. Recommended amplification: 80-300W.
Dimensions: 67" (1702mm) H by 161/2" (419mm) W by 28" (711mm) D. Weight: 133 lbs (29.5kg).
Finishes: light or black oak, maple, standard; natural cherry, dark cherry, dark oak, walnut, mahogany, piano black, aniline dyes (red, orange, yellow, green, blue, purple) available for $300-$500/pair.
Serial numbers of units reviewed: PEMAO16/17.
Price: $10,000/pair. Approximate number of dealers: 180. Warranty: 5 years with return of registration card, nontransferable.
Manufacturer: MartinLogan, Ltd., 2101 Delaware Street, Lawrence, KS 66046. Tel: (785)-749-0133. Fax: (785)-749-5320. Web: www.martinlogan.com.

Sidebar 1: Associated Equipment

Analog sources: Linn Sondek LP12 turntable with Lingo power supply, Ittok tonearm, Spectral moving-coil cartridge.
Digital sources: Krell MD-1 CD transport, Adcom GDA-700 D/A processor, Sony SCD-C555ES SACD player.
FM tuners: Day-Sequerra FM Reference Classic, Rotel RH-10, Magnum Dynalab MD-102 and MD-205 Sleuth RF amplifier.
Preamplification: Krell KBL preamplifier, Mark Levinson ML-7A preamplifier with L-2 phono section, Margulis and Duntech MX-10 moving-coil preamplifiers.
Power amplifiers: Mark Levinson No.334, Bryston 7B-ST monoblocks.
Loudspeakers: Dynaudio Contour 3.0, Revel Salon, B&W Nautilus 805, Quad ESL-63, Velodyne HGS-18 subwoofer.
Cables: Digital: Silver Starlight and Ultralink 75 ohm coax. Balanced Interconnects: Bryston, Krell Cogelco Yellow, PSC Pristine R30 silver-alloy. Singled-Ended Interconnects: Randall Research, Mark Levinson HFC (with Camac connectors), Totem Acoustic Sinew, Coincident CST Interface, Ultralink Performance Audio. Speaker: Levinson HFC-10, PSC Pristine R-50 biwired double ribbons, Ultralink Excelsior 6N OFHC, Coincident Speaker Technology CST 1.—Larry Greenhill

Sidebar 3: Measurements

As is my current practice, the MartinLogan Prodigy was measured out of doors, in my enclosed backyard. I usually raise the speaker being measured on a high stand so it is well away from the ground. In the case of the Prodigy, though, its 133-lb weight and awkward bulk meant that I couldn't get it off the ground. (Apparently, it is impossible to acquire suitable skyhooks in my neck of the Brooklyn woods.) I indicate in the text where this practical difficulty impacted specific measurements.

The Prodigy's voltage sensitivity was a little below specification but still good for a panel speaker, at an estimated 88dB(B)/2.83V/m, and this will be aided in-room by the Prodigy's line-source behavior. (The interaction of the tall source and the floor and ceiling results in an in-room sound that does not fall off with distance as quickly as with a conventional point source.) However, the plot of the MartinLogan's impedance magnitude against frequency (fig.1, solid trace) indicates it to be, as specified, basically a 4 ohm load through much of the audioband. This will mean the speaker will actually draw 2W from the amplifier at that voltage level, not 1W.

Fig.1 MartinLogan Prodigy, electrical impedance (solid) and phase (dashed). (2 ohms/vertical div.)

As with other electrostatics, the Prodigy's impedance drops to a very low value above the audioband: to 1 ohm at 25kHz, which I suspect is the series impedance of the internal and external wiring. Fortunately, even with wideband media like SACD and DVD-Audio, there is very little energy in this frequency region, so the partnering amplifier will not be required to drive much power into this low impedance. But look at what happens around 8kHz, a frequency where there is significant musical energy: an impedance of 3.55 ohms coincides with a capacitive phase angle of -58 degrees, which results in a much lower effective impedance in this region. A powerful amplifier capable of handling 2 ohm loads will work best with the Prodigy.

The two traces in fig.2 show the individual responses of the Prodigy's panel and its dynamic woofer section. The former was taken on-axis halfway up the panel, on its centerline, at a distance of 50"; the latter is actually the complex sum of the front- and rear-woofer nearfield responses, taking into account the differences in distance between the two drive-units and a nominal farfield listening position. (The two woofers are wired out of phase; but because the front and rear woofers face in opposite directions, an electrical pulse results in both cones moving in the same direction.)

Looking first at the panel, a strong reflection of its output from the ground between it and the microphone leads to a narrow but deep suckout at 1.6kHz, and to a rather "furry"-looking trace at higher frequencies. Ignore those characteristics, which are specific to the measuring situation, and the response trend is basically smooth and flat up to 10kHz, above which the speaker's output shelves down by 3-4dB, due to the proximity effect you get with a physically large speaker like the Prodigy. (This effect was described in my measurements accompanying our May 1997 review of the MartinLogan SL3—see Gayle Sanders' "Manufacturer's Comment" on that review, also included in the website reprint.) Moving the microphone out to 100" gave a flatter top-octave balance (not shown).

Fig.2 indicates that the panel rolls off with an approximate 18dB/octave slope below 300Hz, the woofer section coming in smoothly below that frequency. The woofers peak at between 50 and 70Hz, with a typical sealed-box rollout at 12dB/octave reaching -6dB in the low 30s—a little higher than I was expecting from the specification. However, checking my notes revealed that the bass control had not been switched to the "+3dB" position for this measurement. In addition, the close proximity of the woofers to a listening room's floor will tend to boost the low bass. LG did note in his review that, judged with the 1/3-octave warble tones on Stereophile's original Test CD, the Prodigy's bass extended down to 31.5Hz in his room—what I would have expected.

Fig.2 MartinLogan Prodigy, anechoic response on axis midway up panel at 50", averaged across 30 degrees horizontal window and corrected for microphone response, with complex sum of woofer responses plotted below 300Hz.

The presence of the early ground reflection makes the MartinLogan's lateral dispersion plot (fig.3) look a little hashy in the high frequencies. Nevertheless, the panel's dipole behavior can be clearly discerned from this graph. Interestingly, the top two audio octaves feature wider dispersion than the region between 500Hz and 2kHz, presumably due to the panel's curvilinear shape. In normal-sized listening rooms, this will tend to compensate for any top-octave depression resulting from the proximity effect. Though the vertical dispersion plot (fig.4) also looks rather hashy due to the unavoidable early reflection, the speaker's response doesn't change significantly over the height of the panel, meaning that it will be relatively insensitive to changes in listener ear height.

Fig.3 MartinLogan Prodigy, horizontal response family at 50", normalized to on-axis response, from back to front: differences in response 90 degrees-5 degrees off-axis; reference response; differences in response 5 degrees-90 degrees off-axis.

Fig.4 MartinLogan Prodigy, vertical response family at 50", normalized to on-axis response, from back to front: differences in response 15 degrees-5 degrees above reference axis; reference response; differences in response 5 degrees-10 degrees below reference axis.

In fig.5, you can easily see the ground reflection in the Prodigy's impulse response after the 5ms mark. But the impulse response itself, just before the 4ms line, has a good, time-coherent shape marred only by some low-level high-frequency ringing.

Fig.5 MartinLogan Prodigy, impulse response at 50" (5ms time window, 30kHz bandwidth).

The ground reflection can also be seen in the step response (fig.6), but note what a perfect right-triangular shape the panel's initial step has. The setback of the bass cabinet results in the positive-going step of the summed woofer outputs coinciding in time with the panel step's recrossing of the time axis. This can be seen in fig.7, which shows the individual step responses (the panel is in red, the woofers in blue), and implies a seamless transition between the two sections in the frequency domain.

Fig.6 MartinLogan Prodigy, step response at 50" (5ms time window, 30kHz bandwidth).

Fig.7 MartinLogan Prodigy, step response of panel (red) and woofers (blue) at 50" (5ms time window, 30kHz bandwidth).

Finally, the presence of that early ground reflection interferes with the Prodigy's cumulative spectral-decay plot. (The presence of reflections in this kind of waterfall plot is indistinguishable from delayed energy due to resonances.) But windowing the impulse response just before the reflection gave the plot shown in fig.8. The dotted region in this graph indicates where the data are invalid, due to the necessary truncation of the impulse response. But the dropoff of the initial sound is fairly clean up to the 1.5ms mark, broken only by the presence of some delayed energy at 12.6kHz, this correlating with the frequency of the ringing seen in the impulse response (fig.5). This may well be too high in frequency and too low in level to have an effect on the sound, given that LG commented so favorably on the Prodigy's high frequencies.

Fig.8 MartinLogan Prodigy, cumulative spectral-decay plot at 50" (0.15ms risetime).

As I have mentioned before, measuring large panel speakers is an exercise fraught with practical difficulties and, sometimes, frustration. Nevertheless, the Prodigy is the best-measuring electrostatic speaker I have examined. I am not surprised LG was impressed by its sound quality.—John Atkinson

MartinLogan Prodigy: Manufacturer's Comment

Editor: Greetings and thanks for the comprehensive review.

In the late 1970s, when Ron Logan Sutherland and I began electrostatic transducer research, it seemed impossible to resolve the myriad problems associated with this fascinating technology. Since that day, innovation after innovation has been required to accomplish the dream and create a reliable state-of-the-art electrostatic system. Today, our entire team is very proud of the Prodigy, and we thank Mr. Greenhill for telling it like it is—in an ultimate design, electrostats not only are superior in how they reveal information and render detail, but also are capable of explosive uncompressed dynamics.

As we focused on dynamic driver design to interface with our electrostat, the challenge was to design a high-resolution dynamic transducer that exhibited the same linearity and low distortion as our electrostatic panel (less than 0.07% THD). To solve this issue, significant attention was paid to reducing distortion-inducing components created by the voice-coil's inductive field, as well as pushing the linear travel to the maximum. Although expensive, the results were substantial. Looking at fig.1, you can see that the typical THD distortion component of the Prodigy's electrodynamic driver at 200Hz at 94dB is less than 0.07%, which is close to that of our electrostatic panel. Looking at a typical electrodynamic driver (fig.2) at those same levels, you can see that the THD is typically in the 0.5$n1% range—seven to 14 times the distortion of the Prodigy woofer.

Fig.1 MartinLogan Prodigy woofer, distortion spectrum, 200Hz at 94dB.

Fig.2 Typical electrodynamic driver, distortion spectrum, 200Hz at 94dB.

Another interesting side note develops from our unique dispersion characteristics. The enclosed waterfall plot (in which we have opened the time window to 20 milliseconds) shows what happens to the acoustic information once the transducer generates energy into a room. The energy field from a typical-quality point source is demonstrated in fig.3. Notice the "chatter" that invariably results from the speaker/room interaction. Notice in fig.4 that the Prodigy's wave-launch room interaction causes significantly less resonance. It has been our experience that the reduction of "chatter" or nearfield reflections results in an greater ability to achieve image, focus, and ambient retrieval in a variety of rooms, helping the Prodigy to achieve high levels of performance in multiple-room configurations.

Fig.3 Typical-quality point source loudspeaker, in-room waterfall plot (20ms time window).

Fig.4 MartinLogan Prodigy, in-room waterfall plot (20ms time window).

Our engineering team developed ForceForward bass alignment to give the Prodigy more uniform bass performance in a myriad of room configurations. The positive consumer response has let us know that we have taken a giant step forward toward resolving room/speaker interface problems. By the way, yes, the Prodigy does require a lengthy break-in period in order for the system to blossom fully.

Thanks also to John Atkinson for modifying his testing formats to reveal more fully the acoustic parameters of our unique technology, and thanks again to Stereophile for your commitment to music and high-performance listening.—Gayle Martin Sanders, President, MartinLogan