Déjà Vu All Over Again 

By Tom Nousaine Contributing Editor

At the Home Entertainment 2005 show in New York last spring (successor to the Stereophile hi-fi show), in a seminar titled “The Great Debate—Subjectivism on Trial,” John Atkinson “invited” Arny Krueger to get out from behind his PC and “confront in person those he criticized.” At least that’s Mr. ’Kinson’s spin on things—that Arny Krueger (and people like me and Peter Aczel) have been hiding behind their keyboards, afraid to discuss the issue face to face or in public. Quotes following are from Mr. Atkinson’s column “As We See It” in the July 2005 Stereophile, including some of his quotes of what Krueger had said.

Mr. Atkinson attempts to build a self-serving argument out of Mr. Krueger’s criticism of Stereophile, namely that it “willfully ignores much that is known about reliably evaluating audio products,” frequently reaching conclusions and making recommendations “that are improbable if not completely wrong” and not taking “enough pains to ensure that it is publishing correct information.” According to Mr. Atkinson, Krueger is basing his assertions “on the one criticism he has repeatedly made over the years that is correct: to wit, that Stereophile’s reviewers do not perform their listening evaluations under blind conditions.” Interestingly, Arny Krueger never mentioned blind tests in his criticisms. This straw-man argument is entirely Mr. Atkinson’s personal invention.

Mr. Atkinson has long pointed out that no reviews currently published in Stereophile use blind tests. This, of course, is true, but the literal criticism in Krueger’s words is not that blind tests aren’t used in listening evaluations but that what “is known about reliably evaluating audio products” has been ignored by Mr. Atkinson and Stereophile.

Mr. Atkinson then goes on to relate an anecdote about a 1978 blind listening test conducted by Martin Colloms, which showed that subjects were unable to distinguish reliably between three amplifiers, two solid-state and one tubed, in a blind comparison. Following that comparison, Mr. Atkinson claims to have made a solid-state buying decision based on those results but later he “began to realize that even though the sound of my system with the [solid-state] Quad was the same as it ever had been, the magic was gone…until I replaced the 405 with an M&A tube amplifier two years later.”

He then concludes that “normal listening had revealed what the blind test had missed” and that the original comparison failed to prove that the amplifiers sounded the same—therefore any criticism of sighted listening must be wrong. Furthermore, he makes an interesting observation about psychological bias—that “the nonaudio [i.e., psychological] factors were all working in favor of my not hearing any problem with the amplifier,” as if anybody would have conscious awareness of bias held subconsciously. He conveniently ignores that those “nonaudio factors” gathered over a two-year period are more likely to have influenced his opinion about the sound of the amplifiers. “Nonaudio factors” certainly could not have changed the physical sound the devices reproduced.

Even so, if the amplifiers truly did sound different, why wasn’t Mr. Atkinson able to prove that (figuratively) blindfolded in the first place? Mr. Atkinson must be arguing that the techniques used to control listener bias, by themselves, interfere with auditory perception. While this is literally true in the psychological perspective, there is no way that a figurative blindfold could have any effect on the acoustical sound being reproduced by a playback device or system.

There had been more than a couple of dozen bias-controlled listening tests published in the popular press by the early 1990s, all of which showed that any competent amplifier device, with flat frequency response into a loudspeaker load in an acoustic space and not overdriven, is capable of delivering a sonically transparent representation of the input signal into the load and into the listener’s ear. Thus it is patently clear that the Krueger criticism hits the nail right on the head. John Atkinson and his staff regularly publish reviews and recommendations about the sound of equipment such as amplifiers, cables, bricks, stands, and assorted “tweaks” that have never been shown to have a sonically audible effect in any way unless they have demonstrably and measurably worse performance.

One doesn’t have to use blind tests to recognize and acknowledge the existing evidence. Indeed, Mr. Atkinson’s entire argument and the main focus of Stereophile is to encourage readers (and industry personnel) to ignore existing evidence and rely on Stereophile’s reviews and other uncontrolled listening that simply promote existing high-end mythology. Indeed all the arguments about the problems of bias-controlled listening tests revolve around the fact that the controlled listening tests have never verified nonsonic artifacts.

In the final analysis, The Great Debate, like paranormal psychology, Nessie, Bigfoot, and Alien Visits, only keeps resurfacing among high-enders because they aren’t able to bring any real evidence to the table. Amp-sound guys are all café racers—the rides look good parked outside the coffee house, but don’t expect to see any of them at the proving grounds or the race track.

It’s also interesting that John Atkinson pointed out that he, personally, had participated in more blind listening tests than most “objectivists.” While this is probably true, it could also be pointed out that he apparently also has more practice in ignoring, misrepresenting, and misinterpreting controlled listening test results than anyone else.

Editor’s Note: About 12 years ago, I published in The Audio Critic the following comments on The Great Debate. “…the unending back-and-forth on the constantly shifting ground of this controversy has left many audiophiles with the impression that a fundamental clash of philosophies was taking place, the audio world’s equivalent of capitalism vs. socialism, religion vs. atheism, Republicans vs. Democrats, protectionism vs. free trade, batting pitchers vs. designated hitters, etc., etc. That is an enormous misperception. The way the lines are drawn today, the debate doesn’t have two arguable sides. It’s more like laetrile vs. the AMA or the Ku Klux Klan vs. civil rights.” A lot has changed in audio in the last 12 years but apparently John Atkinson hasn’t.

Table of Contents

Postscript to the Linkwitz Lab "Orion" Review 

Listening every day to the unique Orion loudspeakers (see the review posted February 6, 2005) I have a recurrent chilling thought. I think of a moneyed audiophile in a high-end audio salon reaching for his checkbook, about to shell out a five-figure or perhaps even six-figure sum for one of the astronomically priced loudspeaker systems from Wilson Audio or Von Schweikert or Burmester or Sonus Faber or some other iconic brand. And I want to cry out, “No, you fool, don’t do it—it won’t sound as good!” The trouble is, there’s no way an audiophile can listen to the Orions because they aren’t set up in any store, high-end or not. It’s extremely frustrating.

It’s a fact that, in most cases, the sonic superiority of one audio component to another is subtle. You have to listen intently and for more than just a few minutes before you are able to decide which one sounds better. Not so with the Orions. Their superiority to conventional loudspeakers regardless of price is as obvious as the nose on your face. Maybe I didn’t make that quite clear enough in the review. You don’t have to listen for more than 15 seconds to realize that the soundstage is deeper, more layered, more detailed, more palpably lifelike than you’ve ever heard out of any pair of speakers. The effect isn’t subtle; it’s overwhelming. But who will believe me sight unseen, sound unheard?

There is a limited number of audiophiles who will take my word on audio issues at face value. They haven’t been disappointed in the past. For the vast majority of audiophiles there is no solution here. You are most unlikely to hear the Orions until you own them yourself. Maybe one of the very few current owners will invite you for a listening session. Siegfried Linkwitz in his Northern California vacation home is one of them (see http://www.linkwitzlab.com/orion_searanch.htm). I certainly won’t invite you; I’ve stopped bringing audiophiles home to play with me in my electronic sandbox ;-).

Table of Contents

Once Again, CDs 

Please refer to the February 7, 2005 posting of CD/DVD reviews for some remarks on my approach to music reviewing and my perception of the current music scene. Also note, in the reviews below, that the year in parentheses after the CD number is the year of recording, not the year of release.

CDs from Harmonia Mundi

W. A. Mozart: Le Nozze di Figaro. Simon Keenlyside, Il Conte; Véronique Gens, La Contessa; Patrizia Ciofi, Susanna; Lorenzo Regazzo, Figaro; Angelika Kirchschlager, Cherubino. Concerto Köln, René Jacobs, conductor. HMC 801818-20 (3 SACDs, 2003).

The discography of Le Nozze is huge and distinguished; yet another recording with something new and different to offer is unlikely. This is that unlikely one. If I could keep only one of my numerous versions of this amazing masterwork, this could very well be the one I’d choose. Because it’s “period practice”? No, but the crisp attacks, insistently defined rhythms, and transparency of texture that are concomitant to period practice result in a unique sparkle here, which is irresistible. Not that it’s Baroque practice; that would be wrong, as René Jacobs explains it in the accompanying booklet, just as wrong as the postromantic style of so many performances in the 20th century. He calls his approach neoclassical, meaning an updated version of late-18th-century practice. I lack the necessary scholarship to evaluate the validity of his approach; all I know is that everything is musical and in perfect taste, nothing is static. None of singers possess astonishing voices but they are all first-rate musicians and sing beautifully, without any awkward glitches. There is a rightness about the whole production and an upbeat spontaneity that make most other performances sound stiff and studied by comparison. Jacobs is a fabulous musician. Of course, one must keep returning to Mozart as the real hero of this opera. I believe that, even though Don Giovanni and Così fan tutte may contain some pages of even greater music, it was in Le Nozze di Figaro that Mozart realized for the first time that he was the world’s greatest composer of opera, and that realization caused unprecedented exuberance. It’s in the music and it’s unique. As for the audio, I have nothing but praise for both the stereo and the SACD layers, although my enthusiasm for surround sound is now tempered by the incredible soundstaging of the Linkwitz Lab “Orion” speakers in stereo. (See my review posted on February 6, 2005.) The recorded quality matches the performance style—crisp, fine-textured, unforced, transparent.

Maurice Ravel: Intégrale de l’œuvre pour piano. Alexandre Tharaud, piano. HMC 901811-12 (2 CDs, 2003).

All of Ravel’s works for solo piano, every one of them? Played by a young Frenchman who does it better than just about anyone else? That’s right! Unless we go back to Gieseking, Cortot, and a few others of the shellac era, I can’t think of any pianist of recent times who plays this rather specialized music more elegantly, accurately, transparently, with more subtle color than Tharaud. It’s pianism of the highest order. I don’t agree with those who consider any of this sublime music, but Ravel certainly exploits the resources of the piano as few others, and his style is instantly recognizable after just a few bars—the sign of a major composer. It’s a delight to listen to Tharaud wring out the keyboard in Scarbo and Alborada del gracioso. What virtuosity, what cool! He makes these fiendishly difficult pieces sound easy and natural. The recording of the piano has exactly the right amount of resonance, and the dynamic range is exceptionally wide. I don’t see much point in other Ravel solo piano recordings from now on!

CDs from Hyperion

Franz Liszt: Six grandes études de Paganini, S141; Franz Schuberts Märsche für das Pianoforte übertragen, S426. Marc-André Hamelin, piano. CDA67370 (2002).

Marc-André Hamelin is a special case. He isn’t exactly a poet of the piano, although he is a wonderful musician. His specialty is prestidigitation, in the original, root sense of the word—presto digits. You think Horowitz had fast and accurate fingers? That was a mid-20th-century standard. This Canadian pianist is a 21st-century superman. His speed and accuracy are simply unbelievable. When you listen to him play these showoff pieces by Liszt you don’t know whether to break into uproarious applause or to giggle in bewilderment, it’s so improbable. Actually, speed is the least of it; it’s the utter transparency of his superfast playing that is astonishing—he never, never “schmears,” every note is perfectly distinct. The Paganini and Schubert transcriptions of Liszt are great fun but not great music; it’s the process of the virtually unplayable made gorgeously playable that is the attraction here. The recording of the Steinway in the great Henry Wood Hall in London could be just a wee bit drier for my taste, but the dynamics and overall presence are just about perfect.

Sergei Rachmaninov: Piano Concerto No. 1 in F-sharp Minor, Op. 1; Piano Concerto No. 2 in C Minor, Op. 18; Piano Concerto No. 3 in D Minor, Op. 30; Piano Concerto No. 4 in G Minor, Op. 40; Rhapsody on a Theme of Paganini, Op. 43. Stephen Hough, piano; Dallas Symphony Orchestra, Andrew Litton, conductor. SACDA67501/2 (2 CDs, 2003–04).

This is an important release, a significant addition to the Rachmaninov discography. It isn’t every day that all four piano concerti of the composer, including the rarely played Fourth, are recorded over a period of only three weeks (the Paganini variations were recorded about 10 months earlier) by the same world-class pianist and orchestra, resulting in a completely unified, consistent approach—and in state-of-the-art sound. Stephen Hough is perhaps not quite a household name among casual CD buyers but he is technically and musically the equal of just about any pianist in the world today, and the Dallas orchestra under Litton is not far below the Big Five in quality. They play this anachronistically Romantic music (composed between 1891 and 1934, not 50 years earlier!) romantically, not sentimentally, with close attention to the original intentions of the composer and without any of the exaggerated accretions it has accumulated over the years. I cannot recall better performances, perhaps not even Rachmaninov’s (primitively recorded as the latter are). Hough is absolutely brilliant in the bravura passages and warmly lyrical in the songful moments; Litton is utterly meticulous in his phrasing of the orchestral parts and exercises very precise control over the Dallas players. The whole production is a delight. The audio is as good as it gets today; the recording engineer was Jeff Mee, who used to be the great John Eargle’s sidekick at Delos, and the sound has the characteristic panoramic quality that was Eargle’s signature. I have revised my opinion of the difference between stereo and SACD surround sound (see my Linkwitz Lab “Orion” review, posted on February 6, 2005), but in this case both the stereo and the SACD layers of the discs are superb, and the differences are mainly playback-equipment related. If you are a Rachmaninov fan—and who isn’t, at least a little—check out this Hyperion set.

CD from the London Symphony Orchestra

Hector Berlioz: Harold en Italie, Op. 16; Ballet Music from Les Troyens. London Symphony Orchestra, Sir Colin Davis, conductor; Tabea Zimmermann, viola. LSO0040 (2000 & 2003).

I like Harold perhaps even more than the Symphonie fantastique; both are the products of Berlioz’s incredibly original, imaginative early period. Try to predict the next two bars of the music from the preceding two bars—you just can’t, as you can even in Mozart, and then the next two bars sound inevitable! Berlioz and Sir Colin Davis go together like oysters and Chablis, an unsurpassable combination; the LSO is a great orchestra; the net result is the finest performance of Harold I have ever heard. Tabea Zimmermann is a magical violist, and the whole performance sparkles and dances like no other. On top of it, the recorded sound is simply superb; the Barbican in London is a highly “audiogenic” venue, and Tony Faulkner, the sound engineer, has made the most of it. The soundstage is exceptionally wide and deep, and the string sound in the lower frequencies is particularly rich. A great CD.

CDs from the Philadelphia Orchestra

Robert Schumann: Symphony No. 1 in B-flat Major, Op. 38 (“Spring”); Symphony No. 2 in C Major, Op. 61; Symphony No. 3 in E-flat Major, Op. 97 (“Rhenish”); Symphony No. 4 in D Minor, Op. 120; Overture to Manfred, Op. 115; Violin Concerto in D Minor, WoO 23; Andante and Variations, WoO 10. The Philadelphia Orchestra, Wolfgang Sawallisch, conductor; Leonidas Kavakos, violin. Clara Schumann: Five Songs, Op. 12, Nos. 4 & 11 and Op. 13, Nos. 1, 2, & 3. Thomas Hampson, baritone; Wolfgang Sawallisch, piano. POA2003 (3 CDs, 2002–03).

Wolfgang Sawallisch is arguably the world’s greatest Schumann specialist. The Philadelphia Orchestra is arguably the world’s greatest orchestra (and unarguably one of the ten greatest). The beautiful new Verizon Hall in Philadelphia’s Kimmel Center for the Performing Arts is arguably one of the acoustically greatest modern concert halls. Get the picture? This is a monumental set of CDs, perpetuating for posterity Sawallisch’s farewell series of Schumann performances in 2002 and 2003. (He occasionally guest-conducts the orchestra to this day, but a complete Schumann series will not happen again; he is in his 82nd year and not in the best of health.) Schumann’s music flows from Sawallisch’s baton in a completely natural, uncomplicated way, as if he were telling us a familiar tale close to his heart. It is musicianship of the highest order. The textures are crystal clear, contradicting Schumann’s stereotyping as a muddy orchestrator, and every bar is brimming with life. If all of this music were as beautiful as, for example, the Adagio movement of the Second Symphony, the whole set would be pure heaven from beginning to end, but that’s not quite the case. There are boring moments in Schumann’s music, passages of mere note spinning. Some will undoubtedly disagree with that opinion. The audio is stereo only, all of it recorded live, and it is a very beautiful sound, big and rich, doing full justice to the famous Philadelphia strings. In the fortissimo moments the bass gets a little tubby, probably because the hall was so new at the time of the recordings that they hadn’t figured out yet the tunable acoustical chambers, which can significantly change the hall’s signature. That a production of this importance could not find a commercial label but had to be released, by default, under the orchestra’s own name is a sad commentary on the classical recording scene.

CDs from Telarc

Hector Berlioz: Requiem, Op. 5. Atlanta Symphony Orchestra & Chorus, Robert Spano, conductor; Frank Lopardo, tenor. SACD-60627 (2003).

Nineteen years after the famous Robert Shaw recording of this music with the same Atlanta orchestra, Telarc’s Jack Renner went back to Symphony Hall to record it once again, this time with Robert Spano and the Direct Stream Digital recording system. One could debate which performance is better, but there can be no argument about the sound. Recording techniques have definitely advanced since 1984, good as they were even then. This time, the 5.1-channel SACD layer of the disc is clearly preferable to the stereo layer, regardless of the stereo speakers. The brass outbursts in the Tuba mirum, if nothing else, require directional cues from the four points of the compass, and only a surround system can reproduce them as intended. The whole vast choral and orchestral perspective works better in surround sound. It’s demo-quality 5.1. The performance itself could perhaps be more dramatic and spontaneous but it is musically impeccable—very polished, meticulous, and faithful to Berlioz. In the Sanctus, tenor Frank Lopardo is a little bit too distant and his singing is better interpretively than vocally. Other than that it’s basically a very good performance. The Requiem is a complex, multifaceted work, involving huge forces, and perfection from beginning to end is virtually impossible.

Jennifer Higdon: Concerto for Orchestra; City Scape. Atlanta Symphony Orchestra, Robert Spano, conductor. SACD-60620 (2003).

All new music, whether good or bad, is strange and baffling when heard for the first time, some more than others. Jennifer Higdon (a fellow Philadelphian) is that rarity among modern composers whose music has some immediate charm on first listening. One reason is the unmistakable melodic content, another is the gorgeous orchestration. The Concerto is a brilliant orchestral showpiece, very entertaining from the first bar to the last, although parts of it sound perhaps too much like a movie soundtrack (but then, again, so does Tristan und Isolde, which is not Wagner’s fault but the movies’). Spano and the Atlanta orchestra play the music with the utmost virtuosity, fully on a par with the Philadelphia Orchestra under Sawallisch, who gave the world premiere of the piece, which I happen to have attended. The City Scape sounds like three additional movements of the Concerto, possibly because it was composed around the same time. No matter; I am just happy that one can hear 21st-century music which is of obviously high quality and not impossible to understand. The recording by Jack Renner is as high-definition as one could wish; I tend to prefer the stereo layer because of the extraordinary soundstaging of my Linkwitz Lab “Orion” speakers, but the 5.1-channel SACD layer is also excellent and will be preferred by most listeners with 5.1 systems.

Igor Stravinsky: Le Sacre du printemps. Carl Nielsen: Symphony No. 5, Op. 50. Cincinnati Symphony Orchestra, Paavo Järvi, conductor. SACD-60615 (2004).

There is really only one way to play the Sacre, considered by many to be the greatest musical masterpiece of the 20th century, and that is to play every bar and every note of it accurately, exactly as written. No “interpretation” is necessary; Stravinsky himself believed that music doesn’t “express” anything, i.e., that “expression” is bogus. Paavo Järvi (Neeme Järvi’s son) comes reasonably close to playing it that way; it is a superbly organized performance, and the Cincinnati orchestra executes his input perfectly. The music was composed as a ballet and when played this way it sounds danceable. The Nielsen symphony is another matter; it is in its own quirky way as strong and self-assured as a Beethoven symphony and doesn’t mind being interpreted. Järvi interprets it beautifully. The juxtaposition of the two works on the same disc is a bit of a programming mystery; if I wanted to be cynical about it I’d say the main reason was that the timing adds up to 73 minutes, a generous length for a CD. The audio, this time by Michael Bishop, is quite magnificent; the sound of the 5.1-channel SACD layer has exceptional breadth and dimensionality, although I am inclined to favor the stereo layer as soundstaged by the Linkwitz Lab “Orion” speakers. You’re way ahead with either sound.

Table of Contents

2-Channel Power Amplifier (Z Series)
Parasound Zamp v.3

Parasound Products, Inc., 950 Battery Street, San Francisco, CA 94111. Voice: (415) 397-7100. Fax: (415) 397- 0144. E-mail: sales@parasound.com. Web: www.parasound.com. Zamp v.3 Zone Amplifier, $299.00. Tested sample on loan from manufacturer.

The three most salient facts about the Parasound Zamp v.3 are: (1) it’s small, only half the width of a conventional audio component, about the size and weight of Merriam-Webster’s Collegiate Dictionary; (2) it’s cheap, one dollar under $300 and about 1/30th the price of a pair of similarly powered single-ended triode monoblocks by one of the high-end voodooist manufacturers; (3) it’s very good, not a whit inferior to those triode jobs. Welcome to the delirious world of comparative consumer audio.

The Design

The Zamp v.3 is a 45/45-watt stereo amplifier. That rating is into 8Ω; into 4Ω it’s 60/60 watts. Meaning: the power supply is not too hefty, but for many applications the power is entirely adequate. The power supply isn’t switch-mode, as you might have inferred from the size of the amplifier; everything in the Zamp v.3 is analog, and the relatively large toroidal power transformer is cleverly shoehorned into the tight-fitting chassis. Even the speaker terminals (spaced for dual banana plugs, thanks goodness) are slanted instead of vertically placed, to keep the chassis height to a minimum. The terminals can be bridged for mono operation; the rated output into 8Ω is then 90 watts. Input impedance is 33 kΩ; output impedance is virtually zero.

In the back (no, not in front) there are two tiny level controls for the two channels; the maximum gain is 22 dB. Theoretically, you could use the Zamp v.3 to play line-level program material without a preamp, but it would be rather inconvenient to reach behind the chassis every time you wanted to adjust the volume. The level controls are more for “gain staging” in an installation to achieve the lowest possible overall background noise. Other features are loop jacks next to the input jacks, for passing on the input signal to another audio component; a ground-lift switch to interrupt ground loops (shades of Bryston!); and an automatic signal-sensitive on/off system, including also a 12V external trigger. Lots of bells and whistles for only $299.

I did not get a circuit schematic, so we’ll have to do without one of Dr. David Rich’s circuit critiques you may remember from the old print days.

The Measurements

I usually start with the frequency response at 1 watt into 8Ω; in the case of the Zamp v.3 it is so flat that I’ll skip the graph. The response is 0.1 dB down at 20 Hz and 0.25 dB down at 20 kHz; at 50 kHz it is down only 1.4 dB. Flat enough for me.

Much more interesting is the THD+N versus power at various frequencies. Fig. 1 shows the 20 Hz, 1 kHz, and 20 kHz distortion into 8Ω; Fig. 2 shows the same into 4Ω. Most publications show only the 1 kHz curve because it’s usually better than the 20 Hz and nearly always much better than the 20 kHz. (The latter condition is known as dynamic distortion.) That’s the case here, too, but even the 20 kHz distortion is more than “good enough for government work;” after all, –76 to –77 dB is only about 0.015% and that’s still excellent, especially since the measurement bandwidth was expanded to 80 kHz to include the second, third, and fourth harmonic. The distortion curves are essentially noise-dominated, and clipping at each frequency is at approximately the rated power level into both load impedances. Fig. 3 shows the FFT spectrum of a 1 kHz tone at full rated power into 8Ω. The second and third harmonics are at –92 and –94 dB; the higher harmonics line up from –102 dB down; all totally insignificant. This is a clean amplifier.

Fig. 1: THD+N vs. power of one channel into 8Ω at 20 Hz (green), 1 kHz (cyan), and 20 kHz (red).

Fig. 2: THD+N vs. power of one channel into 4Ω at 20 Hz (green), 1 kHz (cyan), and 20 kHz (red).

Fig. 3: Spectrum of a 1 kHz tone in one channel at 45 watts into 8Ω.

I also ran a test that I usually don’t bother with, namely THD+N versus frequency at 1 watt into 8Ω. Noise becomes more of a factor at such a low level, especially since I opened up the measurement bandwidth to <10 Hz–80 kHz. The result is shown in Fig. 4; as you can see, all distortion components are below –82 dB from 20 Hz to 3 kHz, rising to a maximum of –69 dB at 20 kHz. There is absolutely nothing to worry about here; this is all an order of magnitude or more below the threshold of audibility, although some megabuck amplifiers yield better numbers (so what?). The most remarkable thing is that exactly the same test at 40 watts results in virtually identical curves, as shown in Fig. 5.

Fig. 4: THD+N vs. frequency of both channels at 1 watt into 8Ω.

Fig. 5: THD+N vs. frequency of both channels at 40 watts into 8Ω.

The single-point noise measurement, unweighted, with reference to rated maximum output, is –106.5 dB. That, folks, is low noise.

Channel separation is shown in Fig. 6. At higher frequencies the numbers are typical of the best amplifiers; at lower frequencies the maximum separation of 66 dB is a little bit on the scant side but still guaranteed inaudible.

Fig. 6: Channel separation at 1 watt into 8Ω.

The ability of an amplifier to drive widely fluctuating load impedances, such as presented by certain loudspeakers, is measured by the PowerCube test. As far as I know, The Audio Critic is the only American audio journal to publish PowerCube measurements. The instrument for the test is made in Sweden; it produces repeated 1 kHz tone bursts of 20 ms duration into 20 different complex load impedances across the amplifier (magnitudes of 8Ω, 4Ω, 2Ω, and 1Ω, phase angles of –60°, –30°, 0°, +30°, and +60°). The graphic output of the instrument shows the 20 data point connected to form a more or less cubelike polyhedron. The test really separates the men from the boys when it comes to real-world loads rather than just resistances.

The Zamp v.3’s PowerCube is shown in Fig. 7. Look at the top edge of the polyhedron. That’s the output into 8Ω at 1% distortion. It’s perfect at all phase angles and it comfortably exceeds the low-distortion rated power because the output in the 22-to-23-volt range translates to watts in the mid-60s. Into 4Ω the power doesn’t double (that would be a level surface at the top) but it’s still quite substantial considering the less-than-monstrous power supply and it holds up well at all phase angles. Into 2Ω and 1Ω the output declines precipitously but it’s still far from negligible, and there is no sagging at any phase angle. All-in-all, this is a very respectable PowerCube for a tiny amplifier.

Fig. 7: PowerCube of the Zamp v.3. The three axes are output in volts, impedance in ohms, and phase angles in degrees.

The Sound

As I used to state over and over again in the print version of The Audio Critic, all amplifiers having high input impedance, low output impedance, flat frequency response, low distortion, and low noise floor sound exactly the same when operated at matched levels and not clipped. This has been proven so many times in double-blind listening tests that opinions to the contrary by the tweako/weirdo element of the audiophile community and by the subjectivist audio press can be totally disregarded. They just don’t get it.

The Parasound Zamp v.3 is no exception to the rule. Just because it’s small and cute and cuddly, just because it costs only $299, it sounds no different from $9000 amplifiers that have comparable power outputs. The possible exception to that statement would be the handling of difficult loads; the PowerCube test indicates no problems in that area (except, of course, limited power into exceptionally low impedances). I have no difficult-to-drive loudspeakers on the premises at the moment.

So, let me conclude with the considered opinion that the Parasound Zamp v.3 is one helluva 45-watt amplifier, regardless of price tag and chassis size. That the price and the size are remarkably small is an aspect of comparison shopping, not of audio performance.

Table of Contents

 Paradoxes and Ironies of the Audio World:

The Doctor Zaius Syndrome

By Peter Aczel Editor
(Abbreviated and edited version of a still timely article first published ten years ago.)

When the truth is so terrible that admitting it would surely make the whole system crumble, ape logic demands denial and coverup.

Have you ever seen that marvelous 1967 science-fiction movie The Planet of the Apes? If you have, you will recall that it depicts a planet of the future where English-speaking anthropoid apes are the rulers and humans are speechless beasts of burden, enslaved by the apes and despised as a totally inferior species. The apes have horses and guns but no real technology. Doctor Zaius, the subtle and highly articulate orangutan who is this society’s “Minister of Science and Defender of the Faith” (he is played by the great Maurice Evans), knows something the other apes do not: that humans in a past era possessed not only speech but superior technology, flying machines, powerful weapons, and so forth, all of which served only to bring about their eventual downfall and reduce them to their present condition. Doctor Zaius fervently believes that any knowledge of this truth about humans would totally destabilize the society of apes and result in the end of their world. The ape dogma he fanatically protects, even though he knows better, is a blatant denial and coverup of the actual history of the vanished human civilization and a paean to the eternal superiority of the ape.

I won’t give away the rest of the plot to those of our readers who haven’t seen the movie and may want to, but doesn’t Doctor Zaius resemble certain key figures in the high-end audio community? He knows the truth but it’s bad for the establishment. The system would come crashing down if the truth were revealed. To pick an obvious example, consider John Atkinson, the subtle and highly articulate editor of Stereophile. Don’t you think he knows? Of course he knows. But if he admitted that $3000-a-pair speaker cable is a shameless rip-off or that a $7000 amplifier sounds no different from a $1400 one, the edifice of high-end audio would begin to totter—or so he thinks (and may quite possibly be right). Consequently, he spouts convoluted scriptural arguments and epistemological sophistries, just like Doctor Zaius, in order to pervert the obvious, uncomplicated, devastating truth.

There is a perfect illustration of this process in the August 1994 issue of Stereophile, where Zaius-Atkinson once again bashes blind listening tests in an “As We See It” editorial. Such tests are of course considered extremely threatening by a publication that reports night-and-day differences in sound which absolutely nobody can hear when the levels are matched and the brand names concealed. He brings up all kinds of intricate flaws and drawbacks that may very well exist in some blind tests but turns his back on the large number of blind tests in which all of his objections have been anticipated and eliminated and which nevertheless yield a no-difference result every time. He knows very well, for example, that no one has ever, ever proved a consistently audible difference between two amplifiers having high input impedance, low output impedance, and low distortion, when operated at matched levels and not clipped—but like Doctor Zaius he conceals that knowledge. He’d rather collect rare case histories of screwed-up blind tests than deal with the vast body of correctly managed blind tests that undermine the Stereophile agenda. (Just for the record, I’ll state for the nth time that there are only two unbreakable rules in blind testing: matched levels and no peeking at the nameplates. To eliminate “stress,” take a week or a month for each test, send everybody else out of the room, operate the switch yourself at all times, switch only twice a day—whatever. The results will still be the same.)

Our columnist Tom Nousaine, in a recent conversation with me, stated his belief that any longtime audio reviewer who has tested hundreds of different audio components over the years knows exactly what the truth is about soundalikes because it is utterly impossible to escape that truth after so much hands-on experience. It asserts itself loud and clear, again and again. Therefore, he argued, the audio journalists who invariably report important sonic differences are most likely a bunch of hypocrites, i.e., exhibit the Doctor Zaius Syndrome. I was strongly inclined to agree with him…

Why do I even bother to tell you all this?

All of our readers who have been with us for more than just one or two issues are aware of my enormous frustration on the subject of scientific truth in audio. The very idea of a Doctor Zaius Syndrome, even it’s only a parody, suggests the existence of antiscience in audio as a tradition, not just a momentary aberration—and a tradition it is, going back to the early 1970s, at the very least. In the late ’40s and throughout the ’50s and ’60s, whatever the most highly qualified and experienced engineers said about audio was the accepted truth. Then came postmodern irrationalism, post-Watergate anomie, fortune tellers in high places, pyramid power, Jesus-haired record-store clerks as self-proclaimed audio experts, untutored high-end journals, pooh-poohing of engineering societies, derision of degreed academics—the B.S. era of audio (and I don’t mean Bachelor of Science). Today, the melancholy truth is that tweako cultism has become mainstream audio, at least above a certain price range, and engineering facts are regarded as disturbingly radical or at least eccentric. The scientific audio community has been marginalized.

I despair at this point of a journalistic solution. Even if The Audio Critic increased its circulation by a factor of 50 overnight—I’m being deliberately absurd— it might still be too late for the message. The cultists have been too deeply indoctrinated and too long. The pimply-faced kid in the Bon Jovi T-shirt who tried to sell you AudioQuest Sorbothane Feet (the bigger kind) in your local audio salon is not going to change his belief system. Not in this antirationalist age and culture.

I can think of only one effective remedy. Many years ago, long before our younger readers became interested in audio, the Federal Trade Commission put an end to fraudulent power-output claims in amplifiers. Today, the power-output specification must take the form of “200 watts continuous power into 8 ohms from 20 Hz to 20 kHz at less than 0.25% total harmonic distortion.” Before then, the same amplifier could have claimed 800 watts because it could produce that for 2 milliseconds at 1 kHz into 2 ohms with 10% distortion. What if the FTC suddenly became interested in audio cable advertising, for example? That chattering sound you hear comes from the teeth of cable vendors at the mere mention of the possibility. And that low, rumbling sound you hear is Doctor Zaius growling, “That's heresy!”

Anyone out there whose nephew or brother-in-law is a young, crusading, Ralph-Nader-like employee of the FTC? Get him interested!

Table of Contents

Center-Channel Speakers:

What the Industry (Even the "White Hats") Often Refuses to Tell You

By Tom Nousaine Contributing Editor

What’s wrong with two-channel stereo? The main problem: it’s a sweet-spot deal. Good spatial rendition requires that listeners sit on the centerline. Off-axis listeners get nearside bias—the stage and images follow them to the near speaker. Farther phantom sources, even for centered listeners, have a midband response-error characteristic (the Holman dip) and tend to produce soft, foggy images. While the latter characteristic is often preferred by artists because it helps mask vocal or performance deficiencies, it lessens accuracy of reproduction in home systems.

The biggest potential advantage of multichannel audio—and I’m not limiting myself to music only—is a stable soundstage with excellent placement and tracking of moving sound sources, a soundstage that doesn’t shift with listener position, perverting timbre, location, and direction, but provides an improved sense of ambience and envelopment. Two-channel playback is similar to looking into the performance space as opposed to being in it. In other words, with a high-performance multichannel recording everybody, even off-axis listeners, get a soundfield that doesn’t have to be dependent on listening position. It’s much easier to “go to the performance” with multichannel.

Because center stage is the primary focus of program content of practically all recorded sound, the center-channel speaker then becomes the single most important speaker in a system. It needs to have a flat response characteristic (natural timbre) and, because it services off-axis listeners and provides another sidewall first reflection into the listening area, it needs to have smooth off-axis response as well.

My observation about sound systems is that very few of them, including two-channel systems, use dead center as a listening position, except for occasional monitoring or demonstration purposes. Solitary listeners will often sit 10 to 15 or more degrees off axis. Auxiliary listening seats may be as far as 30 to 45 degrees off center. Few people, even those that are hardcore two-channel listeners, use a sweet-spot-only strategy. This further emphasizes the importance of center speakers for modern systems.

Looking at the market, I’ve found that nearly all center-channel speakers have a low, wide profile using an MTM—often called D’Appolito—array, with a pair of 4", 5.25", or 6" woofers flanking a dome tweeter. This style is decor-friendly, has a better partner acceptance quotient (PAQ?), and is physically well suited for perching atop a large-screen rear-projection display. While the driver layout of these speakers is sometimes called a D’Appolito array, few of them fit the criteria of the originator as to driver spacing, crossover frequency, and final response.

These speakers, even the better ones, have a universal characteristic: strong, sometimes severe, lobing in the horizontal plane, which will be launched into the listening area either through reflected sound or direct radiation to off-axis listeners. Fig. 1 displays an example of a typical, currently sold, MTM center channel speaker. Lobing begins at 7.5 degrees and by 22.5 degrees is quite strong. This is a sweet-spot device!

Fig. 2 and Fig. 3 are examples of a “better” MTM sold as a left-center-right (LCR) model for use at all channel locations. Used vertically, it has smooth, even response. Used horizontally, lobing is quite evident and, in my opinion, prevents this speaker from delivering an acceptable level of performance for use in a high-quality playback system that aims to accommodate anything but a single-listener sweet spot.

A few manufacturers address this condition. Fig. 4 and Fig. 5 show an MTM speaker that uses an offset tweeter. The tweeter is located above the line that connects the centers of the woofers. This particular speaker is also billed as an LCR model; the layout is the same for left, center, and right, with the center channel used horizontally. Fig. 6, also a two-way speaker, has a differing layout of horizontally arrayed drivers. Performance as a center channel is quite good, although the speaker is about the same “height” as a small vertically arrayed two-way speaker. Because of the unusual driver layout, directivity is somewhat different to the left and right sides of the speaker.

A few manufacturers make a coaxial center-channel speaker. Coaxial speakers usually have response anomalies caused by mounting the tweeter on a post that extends from the pole piece in the location where the dust cap would normally reside. Some of them use a tweeter that is suspended from a clip or assembly attached to the frame at the edge. This also causes acoustic shadowing, although it often works better than a post mount. One manufacturer uses a coaxial driver with the tweeter located where the dust cap is typically found. This is a logical extension of the coaxial idea, but every one of these devices I’ve used suffers from a response deviation at higher frequencies, which seems to be related to reflections from the tweeter radiating up the cone. A number of years ago DCM (the old DCM, not the Mitek company that currently markets the brand) had a coaxial speaker in which the tweeter was offset and mounted on a mesh screen in front of the woofer. As I recall, it worked pretty well. I wonder whatever happened to that idea?

Occasionally I run across a 2.5-way MTM system where one of the woofers is a passive radiator. These systems also display lobing, as will a 6.5" two-way when you lay it on its side, but performance varies to the left and right sides of the speaker. The basic problem remains: drivers operating side by side will show driver interference (off-axis lobing) in the crossover area where they are both operating. When the flanking drivers cover differing frequency ranges, directivity will vary left to right.

One method enthusiasts use to address a problem is simply to throw money at it. Fig. 7 shows a heavy, much larger, high-cost solution using exotic drivers. If you are rich enough, you too can buy a poor-performing channel. Just get out your checkbook. As a rule of thumb, hurling money at an audio problem or dilemma is nearly always a fast track to worse performance. In this case, the speaker has fairly uniform directivity—response is equally bad at all radiating angles.

I’ve had the opportunity to test hundreds of different loudspeaker systems over the past decade. From 80 to 100 Hz upward, the best systems—and I mean best in an absolute sense—have been 6.5" two-way designs. The speaker graphed in Fig. 8 is among the finest I’ve encountered. As you can see, it has tightly controlled directivity in the horizontal plane out to 60 degrees. This means that all listeners get the same basic sound quality, and reflected sound will be returned with the same timbre. There is a relatively small amount of lobing in the vertical plane, which means that this speaker sounds best vertically placed, and that’s how I use it. This cabinet is only 14" tall, looking almost graceful perched on a 50" rear projector, but hidden behind a 115" perforated screen even that doesn’t matter. What’s the downside? It doesn’t have adequate dynamic capability below 80 Hz. So what? If this were a full-range speaker, in the truest sense, I’d cross it over at 80 Hz anyway because center-channel placement always compromises bass performance.

Two-way speakers by manufacturers such as Paradigm, PSB, Polk, NHT, JBL, Boston Acoustics, and Infinity make excellent center-channel choices, often at remarkably low cost. The optimal center-channel speaker choice will usually be a good-quality, vertically deployed 6.5" two-way speaker. Good sound, even greater value.

One problem with this idea is that dealers are generally reluctant to sell these speakers in quantities of one. You generally have to buy a pair. But at these prices, why not? Better yet, can you dig out one of those 5.25" or 6.5" two-ways you have gathering dust in the basement?

Dedicated center-channel loudspeakers are designed and marketed as horizontally arrayed, low-profile, gracefully styled accessories for large-screen television. While the inclusion of a pair of woofers improves low-frequency capability, few of them will deliver an honest 40 Hz at the listening position and, frankly, there is no need for them to do so. They offer no other performance advantage, and most lack acceptable sound quality for more than one listener.

I’m amazed at how few people, even “the good guys in the white hats,” ever mention the disadvantages of the typical low-rider center channel. Even the best companies with the smartest designers seldom talk about this issue. They seem to be happy with performance that fits into a relatively narrow listening window and happily sell thousands of low-riders. Major magazines don’t seem to be interested in an exposé of the issue (“too technical”). High-end publications stubbornly cling to stereo. Still others subscribe to the unfortunate scheme of five full-range speakers with a single central listener, which is killing DVD-A and SACD.

The film people have always dragged the audio crowd kicking and screaming into the future. Where did most people first hear reproduced sound? Radio, of course, but that’s the last time that audio was a frontrunner. Stereo? At the movies. Dolby surround? At the cinema. True multichannel sound? THX-certified theaters. Dolby Digital, followed by DTS, is still the de facto multichannel standard.

Furthermore, film soundtracks are purposefully designed to accommodate an audience, a large audience. By contrast, many DVD-A and SACD recordings are mixed without a center channel, so we still get a stage that moves with the listener. Many are mixed according to the five-speaker format with the mixer centered in the sound field, so the end user is expected to use the same playback scheme.

Personally I love surround sound. In fact, I can’t live without surround sound for either movies or music. I also love having girls over for movies and music. So I’m just amazed that the DVD-A and SACD folks haven’t figured out that most customers no longer use the “motorcycle” paradigm for entertainment. (You can get more than one person on a motorcycle but only one gets to “ride” it.)

I’m a big fan. I hope this discussion will help you to make choices about the most important speaker in your multichannel system.

Graphs:

The solid traces represent on-axis (0°) response. The dotted traces are horizontally off axis at 15° intervals.

Fig. 1: A modern center-channel speaker. Note the lobing that starts dead center and gets more severe as radiating angle increases.

Fig. 2: Speaker is used vertically at all channel positions except center. Response is quite good—flat spectrum, well-controlled directivity.

Fig. 3: Same speaker as in Fig. 2 but used horizontally. Note the lobing near the crossover, which begins quickly and becomes relatively more severe at wider radiating angles.

Fig. 4: Used vertically, this LCR model with offset tweeter has a good spectrum and reasonably well-controlled directivity.

Fig. 5: Used horizontally, the same 3-way LCR with an offset tweeter actually works better than when vertically placed.

Fig. 6: This speaker uses an unusual driver layout. It has good performance but is still about a foot high in the horizontal position. Response is similar for both center-channel and other-channel use. However, the unusual layout means the speaker has somewhat different response to the left and the right .

Fig. 7: This expensive system with exotic drivers has somewhat uniform directivity but little else to recommend it.

Fig. 8: This vertically arrayed model has a flat spectral character and tightly controlled directivity to 60° off axis. Because it has a single 6.5" woofer, SPL capability below 80 Hz is diminished.

Table of Contents

2-Channel Power Amplifier with Processor

Soaring Audio SLC-A300

Soaring Audio, Inc., 2900 Fantasy Lane, Sparks, NV 89436. Voice: (866) 640-SOAR or (775) 425-8000. Fax: (775) 425-3000. E-mail: sales@soaringaudio.com. Web: www.soaringaudio.com. SLC-A3000 power amplifier with “Signal Loss Compensator,” $3400.00. Tested sample on loan from manufacturer.

Tweako/weirdo amplifier design is alive and well and living in Reno. The Soaring Audio SLC-A300 soared briefly at the January 2004 CES, after which I would have expected it to glide listlessly to earth, but judging from the SA website it is apparently still aloft and winging it. The theory (or shall I call it justification?) on which the design is based is so vague, so ambiguously presented, so devoid of scientific plainspokenness that I must be careful lest my well-known fondness for bashing technobabble get out of hand.

The SA party line is that an inventor, one Jan Coyle, got together with an EE professor, namely Dr. Bill Avery, and a Golden Ear, Daniel Kolbet, who can hear the difference between resistors. The happy triangulation of these talents produced, after years of experimentation, the patented Signal Loss Compensator circuit, which ends the agony of frustration with deficient sound. Hidden details emerge, the perceived dynamic range expands, and “digital grit” due to the 44.1 kHz or 48 kHz sampling rate (what—you can’t hear it?) is eliminated. The explanation of both the problems and the solution, on the website and in the literature, is so secretively allusive, noncommittal, and devious that it’s not worth repeating. Dr. Bill Avery is the head of the company; his home telephone number is also the company’s telephone number (doesn’t that tell you something?); and he is supposed to be a professor of electrical engineering at the University of Nevada, Reno, even though there is no trace of him on the university’s website, which lists every EE faculty member. So much for background.

The Coyle Signal Loss Compensator circuit is built into the power amplifier and cannot be disabled. I would have liked to test the plain power amp without enhancement but couldn’t. It simply isn’t available as a neutral conduit. There is a left-channel SLC control, a right-channel SLC control, and a volume control, all three of which interact crazily with the input signal level to determine the output level. Turning the SLC knobs all the way down, counterclockwise, is supposed to minimize the enhancement, but then there is no output even with the volume control all the way up, regardless of the input level! Just to set the amplifier for the rated maximum output of 100 watts into 8 ohms requires endless diddling with the input level, the SLC knobs, and the volume control. There are flashing green and red lights, nine of them, indicating various functions while this is going on, and what’s the result? See Fig. 1. I have never seen anything like it.

Fig. 1: Spectrum of a 1 kHz tone at 100 watts into 8 ohms.

With the 1 kHz fundamental at 0 dB, referenced to 100 watts, the 2nd harmonic is at –31 dB (2.82%), the 3rd harmonic at –40 dB (1%), the 4th harmonic at –44 dB (0.63%), the 5th harmonic at –45 dB (0.56%), and so on, with significant harmonics all the way up to the 24th, where the graph stops. I must emphasize that the amplifier was far from clipping at 100 watts into 8 ohms; it just produced an amount of distortion unheard of and unthinkable in the 21st century. The rms total of all those distortion components is 3.5% (–29 dB), and I found that the distortion does not vary with frequency, only with level. Even at 1 watt it is about 0.5% (–46 dB). And that’s not all. Look at those sidebands around each harmonic. Is that perhaps the Signal Loss Compensation? Could be; anything is possible. One thing is certain—there is not a word about distortion on the SA website or in the literature, even though THD is the most highly touted spec of other amplifier makers. I also found, by the way, that the volume control creates a very significant ultrasonic resonance unless it’s all the way up, fully clockwise. By the time I got to that point, however, I had given up on the amplifier.

Conclusion

After my distortion measurements, I decided not to listen to the SLC-A300. Harmonic distortion from 0.5% up to 3.5% is unquestionably audible, and then I’d be knee-deep in the fruitless argument about the alleged superiority of the Soaring Audio’s processed sound to the unprocessed sound of a neutral amplifier. Forget about it; I won’t go there. Processing, whether it’s Signal Loss Compensation or any other, belongs upstream of the power amplifier, where it can be turned on and off for reference. The power amplifier itself should be a totally transparent signal path. Soaring Audio also lists the Signal Loss Compensator as a separate, standalone box (SLC-500, priced at $1200), which is the way I would have preferred to test it; inextricably linked to a power amplifier it is an electronic abortion.

I don’t particularly like to write reviews like this, and if Soaring Audio were a bit more forthright about the design of the SLC circuit it would probably not be necessary. I could agree or disagree with the design concept without throwing their baby out with the bathwater.

---Peter Aczel

Table of Contents

2-Channel 192/24 D-to-A Converter
Benchmark DAC1

Benchmark Media Systems, Inc., 5925 Court Street Road, Syracuse, NY 13206-1707. Voice: (315) 437-6300. Fax: (315) 437-8119. E-mail: sales@benchmarkmedia.com. Web: www.benchmarkmedia.com. DAC1 2-channel 24-bit 192-kHz digital-to-analog converter, $975.00 (direct from factory). Tested sample on loan from manufacturer.

Standalone digital-to-analog converters make sense only if they are significantly better than the DACs built into typical integrated players. This one is. It is made by professionals for professionals, for reference-quality playback in studio control rooms and mastering rooms, although it is perfectly suitable for audiophile use. What it clearly does not aspire to is audiophile bragging rights based on price and tweako/weirdo features. The cloud-cuckoo-land high-end DACs at ten times and fifteen times the Benchmark’s price are no better and in most cases not as good. Their owners may delude themselves with fantasies of ineffable sonic superiority, but in reality the only fact-based brag they are entitled to is that they paid more than the poor peasants with Benchmark DAC1’s.

Why can I confidently make that statement? Because I measured the DAC1 up and down and sideways with the Audio Precision SYS-2722, possibly the most sensitive and accurate audio-test instrument in the world, and found it to be as nearly perfect as a digital-to-analog converter can get at the present state of the art. Totally perfect 24-bit converters, with the theoretical noise floor of –146.24 dBFS and a perfect monotonicity “staircase” waveform at the ten lowest LSBs, do not yet exist, at any price, and probably never will. Still, the DAC1 yielded the best measurement figures that I have ever obtained out of a digital processor on my test bench, nor have I ever seen better measurements on other units in other publications. Of course, those who believe that the best-sounding electronic components are not the ones that measure best will pooh-pooh the Benchmark. At this point I’m too old to get upset by these audio cultists, any more than I am by the advocates of crystal power or creationism. The fact is that Benchmark designed the DAC1 on the Audio Precision, as witnessed by the 15 AP graphs in the instruction manual; the whole design is obviously measurement-driven. That the audible performance tracks the measurable performance is a given.

The Design

No circuit schematic was available for the DAC1, and in view of the stunning measurements none was really necessary for evaluation. What I do know, from various sources, is that the basic architecture of the circuit is very similar to the one described by Robert W. Adams of Analog Devices in his article on clock jitter in Issue No. 21 of The Audio Critic, published more than ten years ago. The heart of the circuit is a second-generation asynchronous sample-rate converter, a refinement of the original ASRC discussed in the article, followed by a stereo 24-bit/192-kHz multibit-sigma-delta DAC. These Analog Devices chips are the latest-and-greatest, state-of-the-art ICs. Further discussion of the DAC1’s technology would get me into deeper technical waters than I wish to wade through (is that a mixed metaphor?), but I must mention that the unit incorporates Benchmark’s ultralow-distortion HPA2 headphone amplifier. No separate analog input is provided for the HPA2, so I did not test it separately, since its performance would have been conflated with that of the D-to-A circuit itself. Basically, it makes no difference which output of the DAC1 you measure; the results will be virtually the same. The chassis of the DAC1 is a very elegant little black box, no bigger than a smallish book and weighing only 3½ pounds. The front panel has various indicator lights, a switch for the digital inputs, two identical headphone jacks, and a gain control that makes it possible to feed a stereo power amplifier directly, without the need for a preamplifier—most unusual and very useful. The rear panel has three digital inputs—coaxial, optical, and XLR (professional)—and two pairs of analog outputs—balanced (XLR) and unbalanced (RCA). An output level switch toggles between calibrated, off, and variable settings. In the calibrated setting, two 10-turn trimmer pots are accessible through tiny holes for fixing the left and right output levels. They have a range of 20 dB. The complement of inputs, outputs, and controls covers the requirements of the most advanced professional applications, as well as just about any audiophile hookup.

The Measurements

The Benchmark DAC1 accepts any digital input up to 24 bits word length and 192 kHz sampling rate. For these measurements, my Audio Precision settings, insofar as they had any influence on the readings, were 24 bits and 96 kHz, unless otherwise specified.

Most measurement protocols start with the frequency response, which in the case of the DAC1 is so flat that I won’t even bother to reproduce the graph. The response in both channels is ±0.01 dB from 10 Hz to 6 kHz, –0.2 dB at 20 kHz, –0.5 dB at 30 kHz, –0.9 dB at 40 kHz, and –4.5 dB at 48 kHz. That’s easy to visualize.

Perhaps the most revealing measurement of a digital-to-analog converter is THD + N across the frequency spectrum, shown in Fig. 1. Most publications don’t show this curve because it is in the majority of cases a little embarrassing. Above 7 kHz or so the curves aren’t particularly meaningful because of the 22 kHz measurement filter, but below that frequency the –105.5 to –107 dB readings are astonishing, especially since they were taken at –3 dBFS to duplicate the graph in the instruction manual and can therefore be extrapolated to –108.5/–110 dB for the full-scale output. (That 3 dB below full scale is a kind of safety guard band, which is actually unnecessary in the case of the DAC1.) The graph in the instruction manual is even a couple of dB better, perhaps because I wasn’t terribly careful about cabling, fluorescent lights, etc. As it is, –110 dB is equivalent to 0.0003% and that’s far below anything I’ve ever measured or even heard of.

Fig. 1: THD + N @ –3 dBFS, left channel blue, right channel red.

Perhaps the most sensitive distortion/noise test is what I used to call the Rob Watts Test (named after Rob Watts, a U.K. engineer), consisting of the FFT spectrum of a dithered 1 kHz tone at –60 dBFS. This is shown for one channel of the DAC1 in Fig. 2. The largest blip sticking out of a bin-by-bin noise floor of –146 dB is no taller than –134 dB. (Is that clean enough for you?)

Fig. 2: Spectrum of a dithered 1 kHz tone @ –60 dBFS in one channel.

I also measured the single-point noise figure, which came to –112.4 dB in both channels. The SNR of a theoretical 19-bit digital system is 116.14 dB, so according to this measurement the DAC1 is about “18½ bits clean.” Even 16-bits-clean equipment is rare.

Intermodulation distortion at full scale, with the most punishing test frequencies of 19 kHz and 20 kHz, is shown in Fig. 3. The sidebands are below –105 dB; all other spurious frequencies are below –114 dB.

Fig. 3: IMD at full scale, 19 kHz + 20 kHz, in one channel.

Crosstalk at full scale is shown in Fig. 4. The graph speaks for itself; I have never seen better channel separation in any piece of audio equipment.

Fig. 4: Crosstalk @ 0 dBFS, left channel blue, right channel red.

Gain linearity error is –0.15 dB at the –90 dB level and –0.5 dB at the –100 dB level, reaching a maximum of –1.25 dB at –108 dB. Errors that small would be hardly visible on a graph with lower resolution than Fig. 5.

Fig. 5: Gain linearity (blue) and deviation from linearity (red) in one channel.

As for low-level linearity pushed to the ragged edge, there’s the dreaded monotonicity waveform, available as a special setting of the digital generator on the Audio Precision SYS-2722. I specified 24 bits, 192 kHz sampling, and 10 samples/step for the monotonicity test. Thus the square-wave half-period was 10/192,000 seconds, or 52.1 microseconds (corresponding to a full square wave frequency of 9.6 kHz); each equal-amplitude section was ten half-periods long, or 0.521 ms; and each eleven-step repetition occurred at a rate of 11×0.521 ms, or every 5.21 ms. As Fig. 6 shows, no 24-bit DAC in the world is perfect, not even the Benchmark, but the envelope still narrows gradually from left to right and there is still the suggestion of a staircase. (I must admit it took a lot of averaging to get even that far.)

Fig. 6: Monotonicity waveform, 24-bit resolution, 10 samples/step, 192 kHz sample rate.

A word about jitter. Some high-end reviewers flap their wings very vigorously on the subject, but as Bob Adams pointed out more than ten years ago in the abovementioned article, there is no reason to single out distortion components caused by jitter as distinct from those caused by other circuit mechanisms. Distortion is distortion, no matter where it comes from, and the tests above cover that ground in sufficient detail. To isolate and measure jitter, one would have to remove the cover and go inside the Benchmark DAC1, because it doesn’t have a digital output (nor does it need one). The instruction manual goes into great detail about jitter, with four different graphs to prove the DAC1’s immunity to it. Just for the hell of it, without much hope of significant results, I ran a hi-rez FFT of a full-scale 20 kHz tone to see if there were any noise sidebands in its vicinity that would indirectly indicate the presence of jitter. As Fig. 7 shows, there weren’t any, at least not under the conditions of my quickie test. And that’s all, folks.

Fig. 7: Spectrum of a full-scale 20 kHz tone, 8 kHz to 32 kHz, one channel.

The Sound

It should be obvious from the above discussion, at least to those familiar with The Audio Critic, that the Benchmark DAC1 has no sound of its own, transparently passing on to its output the quality of its input. Whatever sonic peculiarities may perchance be audible are due to the input signal, not the DAC1 circuit. Even if the circuit were a lot less perfect, that would still be the case. Absolute sonic transparency is a concept innocent audiophiles are uncomfortable with, believing that all audio components—CD players, preamplifiers, amplifiers, tuners, all of them—exhibit varying degrees of soundstaging, front-to-back depth, grain, air, etc. That it isn’t so, except in the case of loudspeakers, is a fact calmly accepted by professional engineers but not by the high-end pundits and high-end manufacturers, who would be out of business if the truth were to sink in universally.

I am by now a little tired of harping on this subject but was still amused by John Atkinson’s comments on the Benchmark DAC1 in the May 2004 issue of Stereophile. John made sure to tell his readers that the Mark Levinson No. 30.6, which cost $17,500 before it was discontinued, still sounds better than the Benchmark, despite the latter’s perfect measurements. In a December 2004 followup (“2004 Editor’s Choice,” namely the Benchmark!) he adds the Theta Generation VIII ($10,000) and the Wadia 27ix ($9959) to the of-course-sounds-better list. I wonder what quality the Mark Levinson, Theta, and Wadia engineers dial into their products—above and beyond flat frequency response, low distortion, low noise, and the other usual suspects—that mysteriously makes them sound better. Maybe I should stop wondering after 28 years as an audio journalist and 57 years as an audiophile.

Now that I got that off my chest I need to add that I actually listened (yes!) to the DAC1 in a no-preamplifier hookup. The digital output of my CD player was plugged directly into the DAC1’s coaxial input and its unbalanced outputs were plugged directly into the electronic crossover of the Linkwitz Lab “Orion” speakers. The sound, needless to say, was mind-blowing (for lack of a better word), but mostly because of the quality of the speakers, though admittedly aided by the total absence of electronic distortion. I would gladly have set up a double-blind listening comparison of the CD player’s analog outputs versus the DAC1, but then I realized that it would be an apples-and-oranges situation. Red-book 16-bit/44.1-kHz PCM against a 24-bit/192-kHz converter? It’s not very meaningful. I’d have to scare up a late-model DVD-Audio player for a valid ABX test. I’ll do it, soon, but you know something? I don’t think I’ll hear a difference. Even so, I’ll take electronic perfection, any day of the week, if it costs $975 instead of $17,500.

---Peter Aczel

Table of Contents

Mapleshade CDs:

Hog Heaven for Audiophiles 

For a number of years now, I have been shaking my head in disbelief over the sound of Mapleshade CDs. They had no right to sound that good because Pierre Sprey, the recording engineer, is what we used to call a TWAD—a “Tree-Worshipping Analog Druid.”

Sprey is also the boss of Mapleshade and he masters everything live to 2-track analog tape at 15 ips, as if the year were 1958! The signal, when it’s transferred to CD, gets converted to 44.1/16 PCM anyway, so why not start with PCM, especially with the options available today in digital recording? It seems to be pure mysticism, but the results vindicate Pierre Sprey’s strange methods. I can honestly say that I have never heard more startlingly real-sounding recordings than his. They are closely miked, because that suits the program material, and the sound is in your face, as lifelike as if you were there. I have shut my eyes on occasion and could easily pretend that I was there. Is it because of “no mixing board, filtering, compression, equalization, noise reduction, multitracking or overdubbing,” as it says in the blurb? Maybe so, maybe not, but the sound speaks for itself.

The music on the Mapleshade label is another matter. It’s mostly jazz, of various schools, and the melancholy truth is that jazz is dead, through no fault of Mapleshade. Sure, many of their recording artists are extremely capable musicians and they go about their business with the utmost seriousness, but the basic jazz culture is a thing of the past, and no straggler, bringing up the rear, can possibly sound like Louis or Bird or Diz or Miles or any of the other greats. It’s just not the same, much as it hurts to admit it.

Here are a couple of the more recent Mapleshade CDs I have listened to:

“A Touch of Evil.” Windmill Saxophone Quartet: Clayton Englar, Jesse Meman, Ken Plant, Tom Monroe, with guests Ran Blake, Paul Murphy, Frank Kimbrough, and Ben Allison. #09432 (recorded in 1989).

This is amazing. Recorded more than 15 years ago but released only in 2002, it’s the ultimate saxophone sound, so real you think you can touch the bells of the instruments. The bass sax, especially, is awesome. The acoustical perspective is close; the saxophones could be in the same room with you. The music is modern jazz, originating from Charles Mingus, John Coltrane, and others. I won’t say it’s as good as if those gents were playing it, but it’s highly listenable, and the unique sound holds your interest to the end.

John Previti Quartet: John Previti, bass; Rick Whitehead, guitar; John Cocuzzi, vibes & piano; Big Joe Maher, drums & vocals; Marianna Previti, vocals. #09652 (recorded in 2002).

Another amazing-sounding jazz recording, subtitled Swinging Lullabies for My Rosetta (based on the first track, an Earl Hines number). The bass and the vibes are particularly palpable. The studio acoustics are a bit more spacious than on the CD above. These are not especially well-known musicians but you can’t say they ain’t got that swing; their rhythm and phrasing are quite convincing. For example, Track 5, a Duke Ellington number, is really very elegant jazz. The vocalists are good but not great. But ah, that plucked bass, those mallets on the vibraphone, those brushes on the cymbal… Audio heaven.

---Peter Aczel

Table of Contents

CD and DVD Reviews 

A note about CD and DVD reviews by the Editor:

Why am I writing these capsule reviews of CDs and music DVDs instead of having a professional music critic write them? Because I just can’t find a suitable successor. There are plenty of practitioners with a greater knowledge of music than mine, but they are all so boring! (B. H. Haggin, where are you now that we need you?) I read the magazines, I read the newspapers, and their music critics all seem deliberately convoluted in their style, noncommittal, covering their asses, making only unverifiable observations, and nearly always completely humorless. I wish it weren’t so. Tell me if I’m wrong.

I also want to clarify why there are only classical music and occasional jazz reviews here (and in the previous print issues of The Audio Critic). Not because I am a music snob. I sincerely believe that, for instance, the Beatles’ “Hey Jude” is better music than any number of concertos by Manfredini. It’s just that present-day pop/rock music is at an all-time low. I cannot believe the depths to which popular music has sunk. No melody, no harmony, no texture, no rhythm (other than boom-boom), no nothing, just unstructured noise to accompany the inane words. I think the marketeers have discovered that the tin ears far outnumber the people with even a shred of musicality. What upsets me even more are the pretentious discussions of this garbage in print and on the air, the deep analysis of shallow music. What a farce!

(Please note, in the reviews below, that the year in parentheses after the CD number is the year of recording, not the year of release.)

DVD-V from BBC Opus Arte 
W. A. Mozart: Die Zauberflöte. Will Hartmann, Tamino; Dorothea Röschmann, Pamina; Diana Damrau, Queen of Night; Franz-Josef Selig, Sarastro; Simon Keenlyside, Papageno; Ailish Tynan, Papagena. The Orchestra and Chorus of the Royal Opera House (Covent Garden), Colin Davis, conductor. DVD-Video OA 0886 D (2003).

Is The Magic Flute Mozart’s greatest opera? Probably not. Is it the most delightful, the most lovable, the most disarming? I would have to say yes. This production certainly does not contradict that point of view. Colin Davis knows his Mozart and leads a stylistically impeccable performance. One is not even aware of the conductor’s presence; everything is just right; nothing is “interpreted”—Mozart addresses us directly, without an intermediary. The singing is uniformly excellent, close to world-class I would say, with Simon Keenlyside particularly outstanding as one of the most engaging Papagenos in my experience (the German of this Englishman is flawless!) and Diana Damrau also impressive, histrionically even more than vocally, as an exceptionally convincing Queen of the Night. The staging is very stylish (the stagehands are intentionally visible as they manipulate the props!), and the Dolby Digital surround sound is state-of-the-art in both localization and envelopment. All in all, a highly successful production.

CDs from EMI Classics
J. S. Bach: “Legend”—Famous Transcriptions by Leopold Stokowski. Symphony Orchestra, Leopold Stokowski, conductor. 7243 5 57758 0 1 (1957–58). Also includes a DVD-Video (see text).

Recorded by the 75-year old Stokowski with an unidentified orchestra (most probably top-notch New York freelancers) in New York City’s Manhattan Center, these half-century old tapes were digitally remastered in 1997 and reissued in 2004. The audio is absolutely stunning, fully up to present-day standards and then some. The soundstage is huge and the creamy Stokowski sound is unmistakable. Perhaps in 1957–58 they didn’t know yet how to foul up the stereo image and the instrumental sonorities. As for Stokowski’s orchestral transcriptions of Bach organ works and chamber music, purists have always rejected them and less uptight music lovers always enjoyed them. I think Johann Sebastian, with his early 18th-century instrumental palette, would have absolutely wallowed in the almost ridiculously rich orchestration of the Passacaglia and Fugue in C Minor. It’s over the top but it’s great—only an effete snob could object to it as an occasional indulgence. All eleven compositions on the CD are performed with tremendous gusto, vitality, and beauty of sound. Stokowski could make any orchestra sound like the Philadelphia; he was an orchestral magician. To top off the treat, there is a bonus DVD in the package. It shows the 90-year old Stokowski, in 1972, conducting the London Symphony Orchestra in Debussy’s Prélude à l’après-midi d’un faune. He doesn’t exactly look young but he conducts with authority, and the sonorities are exquisite. What a maestro!

Ludwig van Beethoven: Symphonies 1–9. Wiener Philharmoniker, Sir Simon Rattle, conductor. 7243 5 57445 2 4 (2002). Ludwig van Beethoven: Fidelio. Angela Denoke, Leonore; Jon Villars, Florestan; Alan Held, Don Pizarro; László Polgár, Rocco. Berliner Philharmoniker & Arnold Schoenberg Chor, Sir Simon Rattle, conductor. 7243 5 57555 2 0 (2003).

Every first-rate conductor in the world leads off his repertory with Beethoven and every one of them does a creditable job, with various individual emphases, mannerisms, and idiosyncrasies. Even Toscanini had his little quirks in this music; Furtwängler had a lot of them, and so did all the other greats. Sir Simon Rattle is no exception, and depending on whether or not one likes his personal signature one can call these performances great or merely competent. I’ll leave it to the professional music critics to fight that out; there has already been a lot of controversy about these recordings; that they are world-class is pretty much a given. I’ll restrict myself to pointing out some interesting specifics. The symphonies were recorded with the Vienna orchestra rather than Sir Simon’s own Berliners to honor a firm commitment that predated his current association. He explains that he would have lost all credibility if he had switched. The Viennese are not quite as disciplined and precise as the Berlin orchestra but they play magnificently, giving it everything they’ve got. (Let’s not forget that Beethoven was also Viennese, by choice if not by birth.) The recorded sound is excellent—clear, detailed, sufficiently reverberant and dynamic, but there exists one step up in depth, juiciness, and general realism (à la Lew Layton or John Eargle) that it lacks. (No big deal.) The Fidelio is somewhat small-scale, almost Mozartean, because Rattle believes that most interpretations are too bloated and Wagnerian. The singers are good, highly professional, but far from great. The traditional entr’acte, the great Leonore No. 3 overture, is omitted! The audio is again highly satisfactory, nothing wrong with it, but also rather small-scale in its soundstage and unexciting. I have always regarded this opera as a disappointment, considering that it was created by the greatest composer who ever lived (see the CD reviews in print issue No. 25 for a longer discussion), so you may want to take my lack of enthusiasm under appropriate advisement.

CDs from Sony Classical
Franz Joseph Haydn: Piano Sonatas No. 31 in A-flat Major, Hob. XVI:46; No. 34 in D Major, Hob. XVI:33; No. 29 in E-flat Major, Hob. XVI:45; No. 49 in C-sharp Minor, Hob. XVI:36; No. 35 in A-flat Major, Hob. XVI:43. Emanuel Ax, piano. SK 89363 (2000).

Emanuel (“Manny”) Ax is the most genial, relaxed, and unpretentious classical superstar I’ve ever met personally, in addition to being, almost incidentally, a superb musician and a great piano virtuoso. His pixieish appearance and evident sense of humor contribute to the nonintimidating impact. In this Grammy-award-winning recording he plays five “middle-period” piano works by Haydn with tremendous panache and fleet-fingered precision. It is in every way a wonderful performance. He uses very little pedal and is recorded rather close-up, although the acoustics of the American Academy of Arts and Letters in New York City would have been quite flattering with a more ambient pickup. I happen to think that this drier recorded sound suits the music perfectly. Haydn’s musical ideas are always delightfully unpredictable and satisfying. These sonatas are not as frequently played as some of the symphonies and string quartets but they are far from second-rate Haydn. All in all, an outstanding recording that moves right to the top of the pile in the Haydn/keyboard category.

Franz Schubert: Piano Sonatas in C Minor, D. 958; in A Major, D. 959; in B-flat Major, D. 960. Murray Perahia, piano. S2K 87706 (2 CDs, 2002).

There is no better pianist in the world today than Murray Perahia. Different and equally good, yes; better, no. And there are no greater piano sonatas in the literature than the last three of Schubert. Different and equally great, yes; greater, no. When Murray Perahia plays, you hear the composer, not the pianist. That’s because the playing is so flawless, so musical, so natural, so egoless, that nothing comes between the composer and the listener. I have heard many outstanding performances of these sonatas, but none that I would clearly prefer to Murray Perahia’s. Beautiful, beautiful piano playing. As for the music, these three sonatas—along with the great C Major symphony, the G Major quartet, the C Major quintet, and the E-flat Major mass—represent Schubert’s quantum leap from excellence to transcendence at the very end of his tragically short life. I am inclined to believe that if Schubert had not died at the age of 31 but lived to be, say, 60, he would have surpassed Bach, Haydn, Mozart, and Beethoven. Mere speculation, of course. At any rate, this recording is on the highest plane of musical experience, except perhaps for the audio, which is a bit too resonant for my taste. The big fortissimo chords sound great that way, but some of those perfectly fingered pearly runs and trills get blurred in the acoustic soup created by the engineering team in Germany. Drier recording would have been better, but it’s a superb CD even so.

Peter Ilyich Tchaikovsky: Piano Concerto No. 1 in B-flat Minor, Op. 23.
Sergei Rachmaninoff: Solo Piano Works. Arcadi Volodos, piano; Berliner Philharmoniker, Seiji Ozawa, conductor. SH 93067 (2002/2003).

Another recording of the most played-to-death piano concerto in the world? Yes, but with two differences. One is Volodos, the other is SACD surround sound—both of which are distinctly different. Volodos is the other young Russian giant of the piano, one year younger than Evgeny Kissin and equally phenomenal, albeit a little less publicized. He has been characterized as a Horowitz with soul (or words to that effect), and indeed his technique is absolutely astonishing without in any way coarsening his lyricism in the tender passages. Really an amazing pianist, a complete package—and he was only 30 when he recorded the concerto. The shorter Rachmaninoff pieces, for piano alone, came a year later, including a “concert paraphrase” by Volodos of the composer’s Polka italienne, a bravura showpiece. The Tchaikovsky remains one of my favorites, despite the obviousness—some would even say vulgarity—of the music; I am reliably in tears when the big theme of the last movement is restated fortissimo near the end. Volodos plays the concerto with tremendous virtuosity but also the most intense expressiveness. Ozawa and the Berliners also perform on the highest level, and the solo Rachmaninoff selections are equally brilliant. As for the surround sound, it definitely makes a difference in this music, especially in giving the piano its own separate space as distinct from the orchestra’s. There is adequate depth and envelopment; indeed, all sonic elements are first-rate. I recommend this CD even to those who own several versions of the music.

---Peter Aczel

Table of Contents

Powered 3-Way Dipole Loudspeaker System

Linkwitz Lab “Orion”

Linkwitz Lab, 15 Prospect Lane, Corte Madera, CA 94925. E-mail: sl@linkwitzlab.com. Web: www.linkwitzlab.com. “Orion” loudspeaker system, $5300.00 the pair (custom-built, complete with electronic crossover and cables, but without the required power amplifier and not including S&H). Kit versions available in various stages of completion at lower prices. Tested samples on loan from manufacturer; later samples purchased by The Audio Critic.

Advancements in technology can proceed along one of two routes: (1) the improvement of old, standard concepts or (2) the pursuit of new and different concepts. The first is safe and generally productive, but the second can lead to considerably more brilliant results. The Linkwitz Lab “Orion” loudspeaker system clearly falls into the latter category. The dipole approach is not exactly a new concept, but using conventional dynamic drivers that are not enclosed in a box (i.e., open front and back) is still a fairly radical approach. And, yes, the results are brilliant.

I’ll go further. I have never listened to a loudspeaker that sounded as musically convincing, in stereo, as the Orion. The soundstage is simply wider, deeper, and more precisely layered—and therefore more lifelike—than I have ever heard out of a pair of transducers. It’s startlingly 3-D sound out of two channels. I feel it is necessary to make that statement before I go into details. The Orion was a totally new kind of audio experience in my life, one that required the reshuffling of established values and assumptions.

That said, I must immediately point out the sad fact that this particular speaker, of all speakers, is not available commercially. You can order it as a kit or you can buy a one-off custom-made pair from Linkwitz Lab. Isn’t that typical of today’s audio scene? You can walk into a store and buy just about anything—except the best! The best just isn’t sufficiently commercial.

The Orion is an “active” loudspeaker, requiring a multichannel power amplifier driven by an electronic crossover/equalizer. That may be one reason why it isn’t commercial—audiophiles stubbornly resist the inherently superior powered-speaker approach, although in this case the power amplifier is not built in and thus a number of alternative amp choices are possible. In any case, other things being equal, an electronic crossover is preferable to a passive crossover, and direct coupling of the drivers to the amplifier is preferable to coupling through a network. The active (powered) format is simply better, in every way—but who will believe me? Only those who already know.

The Design

I’ll have a lot more to say about the sound of the Orion, but first let’s look at the design. Each speaker incorporates four drivers. Two 10" woofers—Peerless Xtra Long Stroke (XLS) “subwoofers,” model SWR 269, from Denmark—cover the range from 120 Hz down. They have a fundamental resonance frequency (fS) of 19 Hz, are wired in a push-pull configuration to cancel distortion, and are not baffled—they fire into open air forwards and backwards. The all-important range from 120 Hz to 1.44 kHz is covered by an 8" woofer/midrange—Excel W22EX001 by SEAS, from Norway—which is also mounted completely open, front and back. This driver has an exceptionally stiff and light magnesium cone and a natural rubber surround. Above 1.44 kHz a 1" soft-dome tweeter—Excel T25CF002 “Millennium” by SEAS—covers the range up to about 27 kHz (that’s right!); it is also open, front and back, but the rear radiation is blocked by the magnet structure. These driver models represent the current state of the art; they were not available until quite recently and are extremely costly. (I understand Linkwitz Lab pays $1200 for the eight drivers in a pair of Orions.) The cabinet housing the four drivers is simplicity itself (see photo), consisting basically of a rigid box with an internal dividing wall but no front and back, topped by an open baffle board. Of course, the dimensions of the cabinet are supercritical, since the exact back-to-front cancellations are part of the fine-tuned acoustical design. It requires a six-channel or eight-channel power amplifier to drive a pair of Orions, six channels if the 10" woofers are paralleled, eight channels if they are driven separately. Not much power per channel is needed for optimum performance; 75 or even just 60 watts into 8 ohms will be sufficient, but twice that power is also safe (and actually needed in the woofer channels with the parallel connection).

A crucial part of the design is the dedicated crossover/equalizer that drives the multichannel power amp. It is an elaborate circuit on a single PC board with many op-amps, powered from an outboard ±12V power supply and available either as a kit or fully wired. All connections are through single-ended RCA jacks. The electrical crossover points are at 120 Hz and 1440 Hz; all slopes are 24 dB per octave; and of course each driver except the tweeter is boosted on its bottom end to compensate for the open-baffle acoustic cancellation. (In the case of the woofers that’s a very substantial boost to yield flat response down to the lowest audio frequencies, so you can imagine the excursion capabilities the woofers must have in this kind of design.) A toggle switch on the front panel cuts the low-frequency boost in if there is unwanted subsonic content in the program material.

The Orions need to be set up quite differently from box speakers. The more air there is around them the better. They must be placed as far as possible from the back wall and not too near the side walls. Consequently, they are not suitable for all rooms and all decors. That’s the bad news. The good news is that they require no room treatment—the average live room is a perfect match to their dipole launch pattern. Also, they can be listened to at much closer range than typical box speakers. The angle subtended at the listening position by the distance between the speakers can be much larger than the conventional 60° and yield an awesome sonic perspective. More about that below.

The Measurements

A dipole speaker like the Orion cannot be measured exactly the same way as a monopole box speaker. For one thing, the usual nearfield measurement of the low-frequency response isn’t valid because it only indicates what the driver (with equalization) is doing without including the acoustic cancellation from the back wave. Moving the microphone further back will include the acoustic cancellation but also allow the room to intrude, so the measurement will no longer be quasi-anechoic. The open-air woofer resonance of 19 Hz is a good indication of the actual bass-response profile, as there is nothing to push the turnover frequency higher.

At the higher frequencies there are also some complications; even so, my indoor MLS measurements are certainly valid as far as the midrange-to-tweeter transition is concerned and unquestionably valid for the tweeter response. The irony of the situation is that the MLS curves, despite their limitations (or possibly because of them?), indicate virtually perfect frontal frequency response (see Fig. 1 and Fig. 2), and I have no reason to believe that totally anechoic swept-sine-wave measurements taken outdoors would be significantly different. (If I were younger and not as lazy as I have become, maybe I’d take the trouble to do those labor-intensive measurements.)

Fig. 1: Frequency response (blue) and phase response (green) at 2 meters on midrange axis.

Fig. 2: Frequency response (blue) and phase response (green) at 2 meters, 45 degrees off midrange axis.

Distortion is also very low, an example of which is shown in Fig. 3. That’s a pretty damn loud 40 Hz tone, and the second harmonic is only –43 dB (0.7%), with the third harmonic, which is more important, down to –60 dB (0.1%). I could have made many more measurements, but it’s fairly obvious that they would not have changed the overall picture. (Linkwitz is quick to point out on his Web site the shortcomings of simplistic measurements such as mine, but it’s nice to know that the numbers are good even if they don’t tell the whole story.)

Fig. 3: Nearfield spectrum of a 40 Hz tone at a 1-meter SPL of 91 dB.

So—the Orion is flat (at least in terms of my particular frequency-response curves), wide in dispersion, deep in the bass, low in distortion, and in general picture-perfect according to my admittedly limited measurement techniques. That, however, says very little about the audible qualities of the speaker, which apparently have more to do with the boxless design and the dipole wave launch.

The Sound

Evaluating the sound of the Orions requires the same readjustment of conventional mindsets and assumptions as the measurements. I have already referred to the positioning of the speakers in the room, but that’s just the beginning. The big surprise comes when you first put on a CD, preferably of orchestral music. What’s going on here? The depth of the soundstage has tripled in comparison with high-end box speakers! Everything is more spacious, more solidly defined, more clearly directional, more 3-D. It’s not a small difference. It’s close to a revelation. The orchestra is laid out with crystal clarity behind the speakers, with respect to both left/center/right placement and front/middle/back placement. The simpler and more straightforward the stereo recording, the greater the clarity—messy 20-microphone recordings are less precisely delineated. If the recording is of a single instrument, say an acoustic guitar, the sound is no bigger than it should be, just the correct size within the recorded acoustics, but with a large sound source, such as a full choir, the reproduced sound becomes huge, as if the speakers were eight feet tall and four feet wide. The subjective impression is that the speakers are no longer there; the listener is looking into the soundstage—big or small, as the case may be—without any intermediary.

Note that I said looking into the soundstage. That’s the limitation of stereo. In the case of 5.1 surround sound, one is immersed in the soundstage. The remarkable thing, though, is that the Orions render a more precisely defined soundstage in stereo, i.e. in front of you, than my excellent Waveform Mach MC 5.1 surround system produces all around you. That I really didn’t expect; I had been pretty much convinced that the best surround system always trumps the best stereo system. (Of course, the fairest comparison would be with five Orions plus a subwoofer.) The fact is that I now prefer to listen to the CD layer of SACDs through the Orions rather than the 5.1 DSD layer through my surround system. I get more auditory information about the performance that way, albeit with an inevitably more limited acoustic perspective. (Oh yes, I forgot to mention than I ordered and purchased a pair of Orions from Linkwitz Lab after I completed the evaluation of the loaner pair.)

Experimenting with the placement of the Orions yields some unexpected results. The conventional equilateral triangle setup with the listener at one of the apexes is fine; the sonic perspective is excellent that way, as long as the speakers are at some distance from the walls. But listening from much nearer, almost centered between the speakers but not quite, say at a 135° angle, brings the soundstage almost frighteningly alive, especially with a wide sound source such as a symphony orchestra. Needless to say, it would not be practical to have your sofa or armchair that near the speakers on a permanent basis, but it’s a very interesting temporary deployment. As I said, the Orions make you revise your established assumptions and procedures.

What about the tonality, the basic audio texture? The lack of coloration or neutrality of the sound, as well as its transparency, will be determined primarily by the quality of the drivers, which in this case is as high as it gets—if better drivers had been available, Siegfried Linkwitz would have chosen those instead. Since the Peerless and SEAS drivers are not exclusive to this design—a few other speaker systems also use them—I cannot say that the Orion is lower in coloration, more neutral, or more transparent than any other speaker. It is certainly equal to the very best out there but no better. Its superiority lies in the structure, rather than the texture, of the reproduced sound. The texture is merely superb but not different; the structure is entirely different from the usual. As for bass response, I find the reproduction of the deepest orchestral and even organ sounds to be fully satisfactory without a subwoofer, but bass freaks may disagree with me. The lowest notes are there, but if you are looking for very high-level playback below 40 Hz or so—maybe for movie soundtracks, etc.—you may want a subwoofer. I don’t.

Playing some of my familiar CDs through the Orions produces new insights. Recordings that I did not consider particularly outstanding suddenly sound magnificent. Others that I rated especially high sound nice, with the expected deep soundstage, but nothing more. The speakers, to say it once again at the risk of being repetitious, make you reevaluate all your previous value judgments. Also, the volume at which each recording sounds the most real, the most alive, is much more critical on the Orions than on other speakers. The sound almost seems to click into perfect realism as you hit the right level. One surprise after another.

Availability

“It is easier for a camel to go through the eye of a needle than for a rich man”…to obtain a pair of Orions—well, maybe not, but certainly easier for him to purchase a pair of Wilson Audio Alexandria X-2’s or Von Schweikert VR-11SE’s for 23 times as much money. They are more readily available but won’t yield equal musical satisfaction because they are still box speakers and don’t benefit from Linkwitz’s special insights. If you just want a set of Orion construction plans and an empty PC board, that’s relatively easy; Linkwitz Lab will ship them to you as soon as you pay for them. If you want precut cabinet panels shipped to you in a flat package, there is a source, and you can also order the eight drivers separately. If you just want to sit back and wait for a finished pair of Orions to arrive at your doorstep, you’ll have to put in a special order and wait a month or more while Linkwitz Lab has them custom-built for you. There are so many different options and so many different prices that it’s easier to refer your to www.linkwitzlab.com than to list them all here. The basic price of a finished pair of Orions with fully wired crossover and all necessary cables, but not including the required six-channel or eight-channel power amplifier, is $5300.00 plus shipping and handling. Considering that it’s arguably the world’s best loudspeaker system, that’s a freaked-out bargain—not even counting all the DIY opportunities to pay a lot less. No, I haven’t tested all the good loudspeakers in the world; I’m just basing my opinion on what I know. My enthusiasm for the Orion knows no bounds. I never had its equal in my listening room; I am totally sold on the “boxless” approach; and I am planning to keep my pair forever. What more can I say?

PS: “Revision 0.1”

After the above review was written and presumably completed, Siegfried Linkwitz posted a modification of the Orion on his Web site under the title of Revision 0.1. The modification consists of removing the screws that hold the midrange driver to the baffle board and holding the driver by the magnet instead. The magnet is glued with silicone adhesive to a light bracket which is fixed to the top of the open woofer box; the rim of the midrange driver remains in exactly the same position as before but floating, with a very small gap between it and the baffle, which is sealed with a soft foam strip. Thus there are no forces whatsoever transmitted from the rim of the driver to the baffle, and the minuscule “pendulum” effect of the magnet in free suspension on the basket is also stopped. It’s a deader, more completely damped mounting system than before, similar to the mounting system used in Linkwitz’s old, expensive Audio Artistry speakers.

Don Barringer, Linkwitz’s associate who researched and developed Revision 0.1, was kind enough to make the changes on my Orions personally, so I know they were done right. The $190.00 conversion kit, available from Linkwitz Lab, is just a bit too tricky for duffers like me to apply without errors. Minimal carpentry skills are necessary.

The difference in sound before and after is not dramatic, as you may have guessed by now. Before-and-after listening tests are not as reliable as side-by-side double-blind tests, so I cannot swear to my conclusions, but I believe there is a definite difference. The total acoustic inactivity of the baffle board results in a slightly more neutral, more uncolored, more natural sound on certain instruments and voices. It’s as if the speaker had totally disappeared, leaving no intermediary between the music and the listener. Or so it seems to me. The trouble is that the sound was pretty terrific even before Revision 0.1.

---Peter Aczel

Siegfried Linkwitz, His Web Site, and the Issue of Credibility

If you haven’t heard of Siegfried Linkwitz, at least in the context of the famous Linkwitz-Riley crossover, you haven’t been an audiophile very long. That crossover, however, is the least of his accomplishments. The man is a walking encyclopedia of audio design, one of the most enlightened audio technologists in the world. Go to his Web site, www.linkwitzlab.com, and convince yourself. It is a whole universe of loudspeaker facts, speaker design concepts, and general audio philosophy, with innumerable links and all sorts of entertaining digressions. I haven’t so far found anything remotely comparable to it on the Internet. It’s an evolving book, a labor of love, and—here’s the most important point—you can take every statement in it at face value because it’s science, not tweako subjectivism.

Siegfried Linkwitz was a Hewlett-Packard scientist before he branched out into speaker design, with three and a half decades of electronic instrument designs to his name. Now, in his late sixties, he has an unequaled overview of loudspeaker engineering possibilities, having considered all available options, all transducer types, all “what-about-this-other” alternatives—he makes a special point of this on the Web and in his other writings—and ending up with the Orion design as the best of all possible tradeoffs. He had ample previous design experience with open-back speakers, among them the Audio Artistry “Dvorak,” “Vivaldi,” and “Beethoven” (Audio Artistry is meanwhile out of business), and the more recent Linkwitz Lab “Phoenix,” but he considers the Orion to be the culmination of his engineering efforts. Read all about it on linkwitzlab.com.

There is a very interesting and very fundamental issue raised by the persona of Siegfried Linkwitz and his Web site. When a scientifically educated professional talks to him or reads him, his statements will not be disputed because they are supported by incontrovertible science and properly qualified by his reservations or exceptions, if any. On the other hand, when a typical audiophile, let us say a music-loving dentist, reads his statements, his credibility will be called into question with a “maybe it’s so and maybe it isn’t—who knows?”. And then—and this is the point I’m trying to make—when a half-educated, self-appointed audio guru in one of the audiophile journals makes a statement, the typical reader reaction will be the same—maybe it’s so and maybe it isn’t. In other words, to the average audiophile, Siegfried Linkwitz and (let us say) Michael Fremer or Jonathan Scull have the same level of credibility! This is an intolerable situation, a symptom of sickness within the audio community. I have no idea what to do about it but I’m highly sensitized to it because I, too, try to be scientifically incontrovertible, although I’m no Siegfried Linkwitz, not by a long shot. Still, The Audio Critic should have a higher level of credibility than the tweako/weirdo subjectivists, and Siegfried Linkwitz should have absolute credibility, without ifs, ands, or buts.

I think this credibility issue is more problematic in audio than in nearly all other disciplines. When people read something in, let us say, Scientific American, they don’t tend to conclude that maybe it’s so and maybe it isn’t. There should be some kind of widely accepted arbiter of authority in consumer audio, but I see none. In professional audio there is the Audio Engineering Society, but the typical audiophile doesn’t even know that it exists. In the case of loudspeakers, especially, qualified and unqualified opinions appear to have equal weight. A sad state of affairs and apparently impossible to remedy.

Table of Contents

Powered 4-Way Digital Loudspeaker System

Bang & Olufsen BeoLab 5 

Bang & Olufsen America, Inc., 780 West Dundee Road, Arlington Heights, IL 60004. Voice: (847) 590-4900. Fax: (847) 255-9064. E-mail: beoinfo1@bang-olufsen.dk. Web: www.bang-olufsen.com. BeoLab 5 loudspeaker system, $16,000.00 the pair. Tested samples on loan from manufacturer.

I have a dream. I dream of the Ultimate Loudspeaker. It can only exist in my dream because in the real world no manufacturer would have the overarching vision and multifaceted expertise to incorporate each and every one of its ideal features in a single design. No way; it could never happen. It would of course have to be a powered loudspeaker because perfect matching of the amplifier channels to the drivers is possible that way and because separate, free-standing power amplifiers are hopelessly twentieth-century. It would have to be a 4-way loudspeaker because 3-way design always stretches the capabilities of the drivers to the limit. It would have high-order digital filters in the electronic crossover because they are linear-phase and just plain superior. The four power amplifiers would be extremely powerful yet small enough to be tucked unobtrusively inside the speaker enclosure, thanks to the most sophisticated switch-mode design. The various functions and protection modes of the speaker would be controlled by a powerful internal computer and DSP processor, which would also permit a single digital S/PDIF connection from a stereo signal source to produce music from the all-in-one amplifier/speaker system. One of the capabilities of the DSP would be to tune the bass response of the speaker to its specific location in the listening room. (I can dream, can’t I?) Also, the midrange and treble response of the speaker would be much wider in dispersion than the usual 60° or 90°, extending essentially to 180°, so that the location of the listener would become totally unimportant. (Asking for the moon? What are dreams for?)

Am I still dreaming? What are those two strange-looking monoliths in my listening room? Could they be loudspeakers? Introducing the Bang & Olufsen BeoLab 5, the speaker that makes my dream come true in every detail, bar none. I can hardly believe it. Amazing. Speakers will never be the same again.

The Design

Bang & Olufsen really did it this time. The Danish firm whose philosophy always seemed to be cosmetics (“Dansk design”) first, engineering second, has leapfrogged the whole speaker industry with an engineering design so advanced and so imaginative that it leaves everyone else in the dust. The BeoLab 5 is the most sophisticated piece of loudspeaker engineering in the world today, at least in the world known to me—and they didn’t neglect the cosmetics, either. Just look at it! It’s a tour de force both technically and visually. There is simply nothing else like it. For once, the high price tag (sixteen big ones) does not appear to be excessive when you “look under the hood.”

The drivers around which the BeoLab 5 is designed are a 15" subwoofer, a 6½" upper woofer, a 3" soft-dome midrange unit, and a ¾" soft-dome tweeter. The downward-facing subwoofer is in a 29-liter (just over 1 cubic foot) sealed enclosure, remarkably small for the high-level 18-Hz capability, and proof of both the elaborate electronic equalization and the large excursion range of the driver. The upper woofer is in a 5-liter (0.18 cubic foot) sealed enclosure, facing forward. The midrange driver and tweeter each face upward and fire into very special acoustic lenses, one larger and one smaller, licensed from Sausalito Audio Works. This Acoustic Lens Technology, invented by SAW president Manny LaCarrubba, permits 180° dispersion of each driver’s frequency band in the horizontal plane. Three circular disks—below, between, and above the lenses—are part of the dispersion system. (The SAW lens reminds me a little bit of the Karlson speaker enclosure of half a century ago—who is old enough to remember it?—which also had the driver firing forward through an aperture that narrowed to a point on top, but of course it was a much more primitive technology, designed for full-range drivers and with a two-dimensional aperture as against the three-dimensional cavity of a SAW lens.) The ultrawide dispersion of the upper midrange and treble is one of the features that distinguishes the BeoLab 5 from all other speakers.

The four amplifiers feeding the four drivers are of a uniquely efficient switch-mode design by ICEpower, a Bang & Olufsen subsidiary. They are so small that they fit easily inside the speaker (along with all the associated electronics), yet they can deliver 1000/1000/250/250 watts, respectively, into the four drivers. Distortion and noise are as low as in the best conventional amps. (This is the technology of the future, not the Halcro kind of thing.) The electronic crossover that feeds the four power amps is part of the elaborate DSP module. Digital filters split the audio spectrum into four bands, without the phase nonlinearities that are inherent in high-order analog filters. Frequencies above 500 Hz are directed to the two acoustic lenses, below 500 Hz to the two cone drivers. (The two other crossover frequencies are not specified.) The DSP is of an advanced design, with 32-bit floating-point processing. S/PDIF outputs from any external source, with sampling rates up to 96 kHz and word depths up to 24 bits, can be fed directly to the speakers. All signals are up-sampled to 192-kHz/24-bit before being converted to analog for the power amps. All you need is some kind of digital player, without any other electronics, to hear music out of the BeoLab 5’s. The DSP also effects delay compensation for each driver, monitors voice coil temperatures and dynamically corrects output levels accordingly, and—most remarkable of all—manages the unique Adaptive Bass Control. A substantial handheld remote control commands all these functions—and more.

The Adaptive Bass Control works as follows. You press the center of the top circular disk on the speaker. A tiny microphone snakes out from underneath the subwoofer. A sequence of tone bursts begins. When it’s over, the microphone changes its position by a short distance. A second sequence of tone bursts follows. When it’s over, the microphone retracts. (Showmanship reinforcing technology!) From the two readings, the onboard computer calculates the low-frequency equalization required for flat room response at that particular location and applies the calculated curve. Flashing red and green pilot lights monitor the entire sequence. Then you repeat the same process with the other speaker. Thus, no matter where the speakers are located, you can have totally optimized in-room bass response. No additional instrumentation needed. The difference in low-frequency smoothness before and after the procedure has to be heard to be believed. Most rooms are terrible bass messer-uppers.

The foregoing is just a superficial summary of the BeoLab 5 design. There is a lot more to it. Go to www.bang-olufsen.com, where you can download various PDFs that will answer whatever question you are likely to have. Sausalito Audio Works’ web site, www.sawonline.com, also has some excellent technical information on the BeoLab 5.

The Measurements

Fig. 1: Frequency response (blue) and phase response (green) 2 meters on axis.

Fig. 2: Frequency response (blue) and phase response (green) 2 meters off axis (approx. 45 degrees).

The most obvious difference between the measured response of the BeoLab 5 and that of any other speaker is the complete absence of high-frequency rolloff at any angle off axis horizontally. Compare Fig. 1 and Fig. 2. Those are my quasi-anechoic (MLS) curves from 300 Hz up, taken at 2 meters on axis and approximately 45° off axis, respectively. There is virtually no difference between the two curves, all the way up to the highest frequencies. Fig. 2 could actually have been taken at 85° off axis with little or no effect on the response—the speaker just doesn’t care! That’s unique to the Acoustic Lens Technology. There is, however, something slightly disturbing about the curves. They ought to be quite flat but show instead a shallow S-shaped bend, dipping in the upper midrange and peaking in the treble range. That vaguely S-shaped profile is evident at all output levels and no matter how close or far the measurement microphone is placed. It appears to be the signature of the BeoLab 5. I asked the design engineer at Bang & Olufsen whether this was deliberate “formatting,” perhaps based on listening-test preferences. He vehemently denied that and sent me B&O’s own frequency-response curves to prove that the speaker is flat. The funny thing is that the B&O curves also suggest the S-shaped profile in amplitude response but, on the other hand, indicate dead-flat power response above 1 kHz. Now, I don’t have the technology (or the patience) to do the large-scale reiterative solid-angle averaging required to derive an accurate power response, so I’ll have to give B&O the benefit of the doubt. I’m wondering, however, why the BeoLab 5 can’t have flat amplitude response—at least on axis and at moderate angles off axis—like some of the other top-notch speakers out there that also have excellent power response.

As far as vertical dispersion is concerned, it is severely limited by the three circular disks that constitute the tops and bottoms of the acoustic lenses. This is not just a byproduct of the design but an essential part of the design philosophy. B&O believes in preventing floor and ceiling reflections, in preference to the arguable sonic benefits of greater vertical dispersion. More about that below.

The Sound

The ultimate purpose of world-class engineering in a loudspeaker is world-class sound, and the BeoLab 5 has got it. It was the immediate successor to my 1996-to-2003 reference standard, the Waveform Mach 17, in my main listening room, and after a very brief initial listening session I said, “Yes, this is better.” Now, the Waveform was gone by then, so I could not do an ABX double-blind listening comparison (which would have been invalid, in any event, because the size and vastly different launch geometries of the speakers would have precluded same-position substitution of any kind). The fact is that no side-by-side comparison was necessary. The Adaptive Bass Control alone, once activated, produced obviously superior results. The low-frequency response was utterly smooth, effortless, powerful, without the slightest lumpiness, and extending all the way down to dc—or so it seemed. Unbelievable bass. The wide horizontal dispersion of the frequencies above 500 Hz also resulted in a unique listening experience—you can sit anywhere in the room with these speakers, as long as they are somewhere in front of you. The aiming of the midrange and tweeter has become totally uncritical. That alone is worth the price of admission.

As for the slightly S-shaped frequency response curve—was it audible? I am not at all sure. Just the slightest suspicion of extra zing at the highest frequencies, of reduced brightness at around 3 kHz—maybe, maybe not. The fact is that the overall sound was superb, as transparent, defined, and alive as I have ever heard out of any speaker, and definitely less boxy than the remembered sound of the Waveform, which had been really minimally boxy to begin with. The BeoLab 5 is, on the whole, a masterpiece in sound, as well as in electroacoustic theory. About the only reservation I ended up with had to do with the height of the soundstage, which should have been taller, ideally. Those circular disks were doing their job—no floor or ceiling reflections, somewhat restricted height on the soundstage. It’s a tradeoff. In all other respects, the soundstage was extremely well-defined.

I must add that the BeoLab 5 does not have the unique 3-D soundstaging characteristics of the Linkwitz Lab “Orion”—no other speaker does. (See the Orion review.) On the other hand, the BeoLab 5 is a brilliant all-in-one engineering package unlike any other design in the world and also unique in sound, in its own way. If I could afford both, I would own both. I particularly appreciate the BeoLab 5’s total repudiation of the tweako audiophile worship of 200-pound amplifiers, the “sweet spot” for stereo listening, and strictly analog design. If an audiophilic orthopedic surgeon happened to be fortuitously persuaded to purchase a pair of BeoLab 5’s, he would be automatically spared every bit of idiotic propaganda about triode amplifiers, silver cables, biwiring, burn-in, spiked feet, etc., etc. Isn’t that wonderful? Bravo, B&O!

---Peter Aczel

 

<< Previous Page | Table of Contents