Hence my quandary on receiving for review the Linn Majik DS-I: What, precisely, is it supposed to do? Does the Majik DS-I contain a hard disk and music-ripping software, so I can use it to store all the music in my CD collection? Does it have a graphical user interface (GUI) that at least matches the one provided by the endearingly free Apple iTunes? Does it include a DAC that allows it to play the music files I've already put on my computer?

The answers to those questions—no, no, and sort of—served only to put an edge on my wondering: What the heck is this thing?

Description
The answer, excepting for a moment the Majik DS-I's integral phono section, preamp, and power amp, is: a perfectionist-quality D/A converter that conforms to Linn's well-defined ideas on the subject of music storage and reproduction:

• For domestic playback, archived music files have the potential to be much more faithful to the original recording than digital media played back in real time (ie, CDs and CD players), given that the former are subject to more deliberate error correction. (Err in haste and correct at leisure, as your grandmother used to say.)
• A music system in which files are stored, served, and played within a local-area network (LAN) is superior to even the finest computer-music systems in which files are streamed to a D/A converter via S/PDIF, USB, FireWire, or other data-transfer technologies.
• The best dedicated ripping programs can create files that are audibly superior to those ripped by, say, iTunes.
• D/A chip quality matters.
• D/A implementation matters, too—which is why Linn forgoes the internal auxiliary features of the Wolfson chips they use, preferring instead to write their own filtering algorithms and such.

Thus, considered as a piece of hardware, and with the enduring, temporary exception of that built-in preamp and amplifier, the Linn Majik DS-I ($4200) is an outboard D/A converter designed to communicate with other music-playback gear by means of an Ethernet LAN, streaming files in accordance with current UPnP specifications. Also provided are S/PDIF (coax) and optical (TosLink) digital inputs, as well as line- and phono-level analog inputs, for direct connection to various non-network source components, but USB and FireWire digital inputs are conspicuously absent. Absent, too, are a hard drive for archival storage, an optical drive for ripping CDs, a video display of useful size, and GUI software, all of which must be separately purchased, installed, and made available to the Majik DS-I via that Ethernet network. (A source-only version of this product, the Linn Majik DS, which lacks the integrated amp, is available for $3500.)

To be fair, I should note that Linn doesn't refer to the Majik DS-I or their other DS products as music servers, but simply as digital players. Pressed for a more technical term, Linn uses the more universal if opaque term media renderers.

Setup and installation
Assuming one wants a Linn Majik DS-I for playing high-resolution digital files ripped from one's own CD collection—a safe enough assumption, I would think—some minimal requirements must be observed in terms of both hardware and software. With regard to the former, the prospective user must also have: a network-accessible storage unit, or NAS (Linn suggests that this can be the hard drive in one's own computer, though they'd rather you buy a high-capacity outboard hard drive for the task); a control unit (which, we're told, can also be one's PC, or even an iPod or handheld "personal data assistant"); an optical drive for ripping additional music files (quite possibly, the one on your computer); and an Ethernet router and switch, which plays conductor to all of those network elements. And, of course, the user needs a few Ethernet cables, which generally aren't too expensive—yet.

Mystère
Mystère is a brand of Durob Audio, a Dutch company known for its PrimaLuna and Ah! Tjoeb brands and the Kiseki phono cartridges. Kevin Deal, of Mystère USA told me, "Durob is one of the largest distributors of the most respected brands of audio throughout Belgium and the Netherlands." While all of Durob's products are designed in Europe, they are, like much audio gear these days, made in China.

The obvious comparison for the Mystère line is PrimaLuna. Some folks assume that Mystère is simply an upper-crust version of the PrimaLuna line, as Lexus is to Toyota. Not so. First, the most significant difference between the two lines is that Mystère runs its output tubes in pentode, PrimaLuna in the more conventional triode/ultralinear mode. "I, for one, have a real affection for pentode amps," said Deal. "You let them run without a lot of negative feedback, so the output impedance is slightly higher. You get the color and glory of [a single-ended triode tube], but with power, and the resulting sound is very tasty. When I think about some of my favorite amps over the years, many of them were pentode designs, or a variation on that. Everybody loved the midrange color and richness. Then most amps went to an ultralinear [topology], and some brands used more feedback to get a lower output impedance. Better measurements, but everyone said, 'Something is missing.' I agree."

There are also big differences in the looks of the Mystère and PrimaLuna products. Not only are the Mystère models clad in a gorgeous gloss-black finish (which I usually don't dig, but this one is amazing), they're completely void of bells and whistles. While you can get a remote control, tape outputs, and other conveniences with a PrimaLuna integrated, the Mystère ia21 integrated amplifier ($2995) is a more minimalist product. On the rear it has four RCA inputs, and 4 and 8 ohm taps, and on the front are a selector knob, and, to control the volume, a stepped attenuator built by Mystère. On the side is a rocker switch to auto-bias its KT88 or EL34 tubes. That's it. If you want to hook up a subwoofer, you need to go through the amp's speaker outputs. You'll also need to get your lazy fanny off the couch to change the volume and input. Though Mystère boasts a mysterious-looking Stonehenge logo on its front plate, the ia21 is actually a simple and straightforward amp.

The ia21 is Mystère's most powerful integrated amplifier. Running KT88 tubes, it's claimed to put out 50Wpc of pentode power. The ia21's Adaptive Auto-Biasing makes keeping the output tubes perfectly biased a snap. In my experience, an amp that mismatches the bias between tubes by even 1–2mV doesn't sound as good as one that's been properly biased. The auto-biasing will ensure that your tubes are always in the zone, and eliminate any neuroses you might otherwise develop by worrying if your tube biases are matched. The auto-bias also makes tube rolling fun and easy. In addition to KT88s and EL34s, the ia21 will work with KT66s, 6L6GCs, KT77s, 6550s, and the new (to me) KT120s. Auto-biasing will let the hobbyist in you have a blast while keeping you from constantly having to tinker with your amp. The best of both worlds, I'd say.

"I am your father's brother's nephew's cousin's former roommate."
Setting up the Mystère ia21 was very easy, especially for a tube amp. Take it out of the box, plug it in, select the correct bias setting for the tubes being used, hook up a source and speakers, and turn it on.

The ia21 comes with a tube cage. Actually, cage is the wrong word—it's more like a tube helmet, with two long, skinny slits that provide the only venting for the heat produced by the tubes. The cage's possible inability to dissipate heat worried me, so I didn't use it. Nor did I like the way it looked. With the cage in place, it seemed as if the decapitated head of the Black Knight from Monty Python and the Holy Grail, or Rick Moranis's Dark Helmet character in Mel Brooks's Spaceballs, were sitting there on my audio rack. However, with its cage off, the ia21 was one classy, sexy beast. I found its qualities of fit, finish, and build to be beyond reproach. Though I'm not big into piano-black gloss finishes for my audio gear, I just loved the look of the ia21. That black gloss was inviting yet scary—I knew that if I touched this thing with my bare hands, I'd never get the smudges off. I was afraid that, after one touch, this pretty ia21 would look as if a snotnosed two-year-old had gotten to play audio for the day. So while I find the entire idea of wearing white gloves to touch audio gear a sign of audio fetishism, I did use the provided white gloves whenever I had to touch the ia21.

The last word is up for grabs: Wilson? Levinson? Linn? Maybe. But for me, whenever I'm in pissing-and-moaning mode, the choice is easy: Why hasn't the average consumer heard of the Audio Note Ongaku?

After all, the Ongaku was, for a while, the most expensive amplifier you could buy: £30,000 in 1988, the year of its release. Just as important, the Ongaku was good—in many ways, the best I've ever heard. Other audiofolk appear to have been similarly impressed. Once upon a time, Gordon Rankin and Don Garber, of Wavelength Audio and Fi, respectively, created a smaller but similarly wonderful-sounding amp dubbed the Baby Ongaku. There's even a guy in Canada who's built and sold a few (reportedly quite good) Ongaku copies.

The Audio Note Ongaku has long been the product to beat among people who prize the immediacy, palpability, and musicality of the finest single-ended triode (SET) amplifiers. And last year, the challenge of competing with Audio Note was taken up by yet another company: Audio Note. Close on the heels of their own redesign of the classic Ongaku integrated amplifier, which is now priced at $121,500, the English manufacturer introduced two amps that employ the same circuit, implemented with humbler parts. The least expensive of the two is the new Audio Note Jinro, which sells for $26,500.

Description
The original Ongaku was created by the late Hiroyasu Kondo, founder of Audio Note Japan, and was sold by Audio Note UK as part of a business relationship so complex—and, in the end, so controversial—that the full story defies a fair telling in less than a thousand words. Suffice it to say, that first version of the Audio Note Ongaku is a thing of the past. (Though there remains the Kondo Ongaku, built in Kanagawa, Japan, by the company founded by Hiroyasu Kondo.)

Still, a number of key elements remain today: enormous output transformers (wound with silver wire) that are intended to resist saturation and thus extend low-frequency response; a reasonably simple, tube-rectified power supply; star ground design, with a solid-copper chassis plate that doubles as a ground plane; silver wiring throughout; and, of course, one 211 (aka VT-4-C) directly heated triode tube per channel. One could add that the Jinro operates in pure class-A, but that goes without saying: In a single-ended output section, the only way to get the job done is for an output tube to conduct current at all times.

The Ongaku and Jinro of 2011 retain the original amp in their DNA, but credit for the new circuit goes to Audio Note's chief designer, Andy Grove. Distinctions abound: Whereas the first Ongaku integrated amplifier used four small-signal tubes, the Jinro and its dearer brethren use only two: After traveling through an Elna source-selector switch and stepped volume control with a nimbus of 44 individual resistors, the input signal for each channel is amplified first by half of a 5814 dual triode, then by half of a 5687 dual triode. As Grove explains, using two different tubes for voltage gain helps to balance the sound—"so you don't get too much of an accumulation of character."

The Jinro's driver stage is also a departure from that of Kondo-san's early design. Whereas the first Ongaku used a cathode follower to drive the output tube, Andy Grove opted for a 1:1 bifilar-wound transformer, designed and built in-house. That, he says, eliminates the slowness of the old approach, and gives the input stage a far more linear load to drive.

From there, the Jinro design appears close to that of Kondo's original Ongaku, with a simple voltage-doubler power supply—complete with an enormous, high-inductance filter choke—and an equally simple output section, built around the legendary 211 triode tube. The output transformers are also made in-house (a transformer that can handle a 20k ohm plate in a full-range single-ended amplifier is not exactly an off-the-shelf item), and the parts throughout the Jinro are in keeping with what Audio Note refers to as a Level Three product: Beyschlag resistors, copper wiring, and Audio Note's copper-foil-in-oil capacitors in the signal path, plus German-made Beyschlag Centralab capacitors in the power supply (although I did notice four fancy-schmancy Black Gate capacitors) and steel cores for all transformers and coils. By contrast, the Ongaku gets silver hookup wire throughout, tantalum resistors, and silver-foil-in-oil signal caps, as well as transformers made with nickel-alloy cores and hand-extruded Italian silver wire—plus a lot more of those pricey Black Gate caps.

Luckily, I'd brought along my Hearos Xtreme Protection earplugs—the best earplugs I've tried. They have a noise-reduction rating of 33dB, and their supersoft foam is light and comfortable in my ears. More important, they manage to preserve enough of the music's tonal color and dynamics, lowering the perceived volume without destroying the sound. A pack of 14 earplugs with a small carrying case costs about $6. Every audiophile (and anyone concerned about their hearing) should keep a pair of earplugs on hand at all times—you never know when you'll find yourself in a dangerously loud environment. Being the only one wearing earplugs in a crowded bar made me feel like a dork, but whatever. My career, and my love of listening, are far more important to me than looking cool. The DJ was spinning a mix of familiar 1980s pop and more recent indie rock, and, after a few $5 vodkas, we were all moving to the music.

At one point during the set, Nicole noticed that, instead of using an iPod or laptop, the DJ was actually spinning vinyl.

"Is his turntable better than yours, Stephen?"

"What?"

"Is the DJ's turntable better than yours?"

Sometimes I wonder about Nicole. Her sweet exterior camouflages a devilishly pointed wit and cunning. She's always leading me somewhere. I smiled and considered my response. How do I answer this question without sounding like an elitist audiophile? Luckily for me, I didn't have to answer at all. Natalie jumped in and rescued me, as she so often does. Moving to the beat of the music, she shouted, "I love the Music Hall turntable!"

Natalie was talking about the Music Hall USB-1 (discussed in the May issue), which had done an admirable job at her party as the dedicated music source. I was extremely happy—and relieved—to hear that she loved it. I smiled and gave her a hug. What else could I do?

We laughed, drank, and danced until 3am, when the music was replaced by the muscle-bound bouncer's awful nightly alarm, a strangely high-pitched wail: "Let's go, let's go, let's go! Let's go, let's go, let's go!"—my least favorite sound in the world. So we went.

It always happens just like this: I walk Natalie and Nicole back to their place. Along the way, we laugh about the time we had, and maybe make plans for later in the week—promises and reassurances for more good times in the future. There, at the steps of their building, we hug and say goodnight.

But there's no satisfying way of saying goodbye—my farewells always feel awkward, rushed, incomplete. Knowing that the girls are safe inside, I turn and walk away. Somewhere along the angled stretch between First Street and Third, beneath the pale yellow light of streetlamps, I'm reminded of my loneliness. I wish I had somewhere else to go, I wish there were someone there for me. (Sometimes, you don't realize how thirsty you are till you've had a sip of water—or, in my case, five vodka tonics.)

In the middle of these dim reflections, usually around Hollywood Fried Chicken, my mind is flooded by the last song of the night at Lucky's: Sir Mix-a-Lot's "Baby Got Back," which will surely corrupt my thoughts for days. I walk again past Lucky's, weaving through tousled partiers as they're spit out onto the rainbow-puddled street. "Let's go, let's go, let's go! Let's go, let's go, let's go!" I turn left, onto Third, make my way to the old glass door of my apartment building, walk into the putrid green light and gingerly up two flights of stairs, and into my empty home. Maybe I'm sober enough to play a record and scribble some notes before I surrender to sleep. Maybe I'm not. It always happens like this.

It happened like that just a few hours ago. It's now 8am and I shouldn't be awake. I'm sitting on the orange couch, staring into this gray Sunday morning, hoping I didn't say anything stupid last night. I'm extremely tired, but at least my ears aren't ringing.

Nicole had sent me a text message a few nights earlier, I remember now.

NICOLE: Are your ears ringing, Stephen?

ME: No.

NICOLE: They should be.

I didn't know what she was talking about and I didn't know how to ask, so I decided not to respond. Sometimes, not knowing is fun; it's good to be surprised every now and then. And now it's time to listen to two very different integrated amplifiers—one cubical and tubed, the other rectangular and solid-state. It'll be at least three hours before Natalie and Nicole are awake. I hope they'll call and ask me to brunch.

So it was with the Micromega AS-400 digital source/integrated amplifier ($4495), the anonymity of which was compounded, in my case, by a generous helping of forgetfulness: I suppose I was told, ahead of time, that this was a class-D amplifier, but at some point in time before my first at-home audition I apparently killed the brain cells responsible for remembering that fact. So I was innocent of conscious prejudice when I listened to this elegant cipher of a box and wrote, in my notes: "Dynamic, dramatic, and almost relentlessly exciting with some recordings. Imbued pianos with almost too much dynamism for the room—too much being very good!—but lacked some 'purr' in the die-away. Basically fine and fun. Wish it had a little more color and spatial depth."

All enduringly true. And you could stop there if you wanted. But the thing is, there's a lot more to the Micromega AS-400 than just that.

Description
Rather like the Linn Majik DS-I before it, the Micromega AS-400 combines a solid-state preamplifier and power amplifier with a custom digital-to-analog converter, the latter tailored specifically to computer-music files (more on that in a moment). Also like the Linn, the AS-400 comes complete with its own phono preamp: a lovely trend, and one that would seem to allow the buyer to take advantage of both the new and the old in terms of cutting-edge music media.

Yet one could argue that the AS-400's real calling card is its implementation of something that Micromega calls their AirStream module—essentially, an Apple AirPort Express WiFi receiver that has been reworked as a perfectionist audio component. Micromega uses three different feeds from an R-core transformer to supply its main module, master clock, and D/A analog section. The incoming digital stream is referenced to the AirStream's own custom-made timing clock, then fed to a 24-bit/192kHz Cirrus Logic CS4351 chip, supported with various perfectionist-quality parts. (The D/A in the AirPort Express is not used.) First seen in Micromega's WM-10 standalone digital source, the AirStream module is intended to allow the owner of an AirPort-equipped Apple Mac or similar computer to wirelessly stream his or her iTunes music files to a perfectionist playback system, thus making an end-run around the whole USB thing.

Not only is the AirStream module in this new product said to be more advanced than those in Micromega's past, but computer-music technology in general has progressed in such a way that a new frontier is available to the prospective AS-400 owner: At the end of 2010, Apple released v.4.3 of their iPod operating system, which incorporates a new wireless-transmission protocol called AirPlay (footnote 1). The long and short of it: One can now wirelessly stream full-resolution 16-bit/44.1kHz, iTunes-compatible music files from an iPod to a Micromega AS-400. Compare that with the Chordette Gem D/A converter (reviewed in the January 2011 issue), which uses Bluetooth wireless technology and a necessarily lossy codec to accomplish the same thing.

The Micromega's phono section deserves special mention: Its sensitivity is appropriate for moving-magnet cartridges, but moving-coil types will require additional gain (and, for most users, a load impedance considerably lower than the AS-400's MM-appropriate 47k ohms). For me, that's no hindrance, as I far prefer loading my MC cartridges with an outboard step-up transformer. Additionally, Micromega has engineered the AS-400 so that, when its phono inputs are selected, power to the AirStream module is interrupted, so that the latter's own power-supply feeds won't add noise to the delicate phono signal. Nice.

Finally, no discussion of the Micromega AS-400 would be complete without mentioning its amplifier output section, which is class-D—perhaps the most misunderstood of the classes, second only to the working poor. The D doesn't stand for digital—although there is, coincidentally, a digital-like concept behind this 60-year-old design: Its output devices are always switched either on or off. The resultant wave is shaped via pulse-width modulation (which is not nearly as digital as it sounds) in an effort to mimic the original signal.

All of the above is housed in a metal enclosure of average proportions and with an above-average level of finish. Perfect is not too strong a word to describe the fit of the casework, the powder-coat finish is uncannily smooth, and the front panel is the very model of understated elegance. Above all other adjectives, the AS-400 looks mature.

Setup and installation
When I open the carton of a new review sample and see that it contains a software disc, the first words out of my mouth are usually "Oh, shit." The Micromega came packaged with a set of discs, but I needn't have panicked: It turned out to be the software and documentation Apple supplies with every AirPort Express they sell; chances are, the AS-400 user will never have to break the seal on their packaging.

The setup procedure for the AS-400 was nonetheless more involved than that for a step-up transformer or a cable riser—more, even, than for most integrated amps, assuming they don't contain wireless music streamers of their own. But as someone who has, in recent months alone, worn on his sleeve a bilious disdain for needlessly difficult setup regimens, you can take my word: The Micromega AS-400 was relatively easy.

Here's how it went for me: At power-up, the AS-400's pilot light glowed blue and the word AirStream glowed red in six-point type on the digital readout. Approximately 65 seconds later, that word changed from red to blue, suggesting that the AirStream module was ready to go. And it was: When I clicked on my iMac's WiFi icon, in the upper-right portion of its display, I saw that Music was now an available network selection. I duly accepted it.

The idea of a good-quality integrated amplifier—a power amp and preamp put together in a box of reasonable size—has never ceased to appeal to me. I owned three good integrated amps before I ever bothered to buy a stereo receiver (itself purchased before countless hearings of Linda Ronstadt's cover of "When Will I Be Loved?" and James Taylor's of "How Sweet It Is" left my enthusiasm for FM in a permanent body cast), and to this day I'm embarrassed less by the humility of an integrated amp than by the expense and expanse of some separates.

It has occurred to me: Some day, my retirement system will surely be built around a good integrated amp.

That in itself makes me wonder: If I could retire today, which integrated amplifier would I choose—assuming I still wanted tubes? And a phono section? And I had about $6000 to spend?

God is good. Over the transom came word of the interesting new Quad II Classic integrated amplifier ($5999). In the wake of that news—literally within a day or two—came an offer to review the current edition of the Luxman SQ-38u integrated ($5990). Coincidence and momentum then conspired to spark my memory of the Leben CS300 integrated ($3395), which, while scarcely a newcomer to the market, had yet to appear in our pages (footnote 2), either single or in concert with its newer sibling, the Leben RS-30EQ phono preamplifier ($2695). (The more expensive Luxman and Quad integrateds have phono sections of their own.) Thus the idea took shape of a small survey, starting with the . . .

Leben CS300: $3395
The Leben CS300 bears an uncanny resemblance to every integrated amplifier I owned between 1970 and 1975. Seeing it today—let alone having it in my home—is like discovering a packet of the same incense they used to burn in my favorite record store of that period (which, as God is my witness, was called Love Records). Thus the Leben may already have an unfair advantage over its competitors, at least on the playing field of my subconscious.

This line-only integrated amplifier contains a total of six tubes: one pair of 5751 dual-triodes, and two pairs of EL84 power pentodes—the latter a tube noted for its ability to work well with very low-voltage input signals. The EL84s run in class-A/B in a straightforward autobias circuit, for a total of 12Wpc. Global feedback is used, and the output transformers are designed in-house, and manufactured by what is described as a subsidiary company.

The Leben CS300 is a split-level amp, the steel floor of which is fastened inside the chassis at a point roughly four-fifths of the way down from the top; the hand-wired circuit and virtually all of the smaller component parts are mounted under that surface, with the mains transformer, output transformers, shielding "wall," and tubes mounted above. Construction quality is stunningly good—not just for this price, but for any price. In my sample, a dozen or so old-style terminal strips supported most of the amp's capacitors, diodes, and (carbon-comp) resistors, and all of the component leads and connecting wires were neatly dressed and soldered perfectly: There wasn't a single stray strand of wire or cold-solder joint in sight.

The vintage theme carries over to the Leben's selection of front-panel controls, which include not only the usual volume and source selectors but also a balance knob and a three-position bass-boost knob, the latter allowing the user to select among flat, 3dB of boost, or 5dB of boost. Additional controls on the front panel allow the user to switch between speakers and headphones—logically, a ¼" phone jack is located close by—and to activate the tape monitor loop. Uniquely, the Leben also has a rear-panel output control: a three-position knob that allows the user to switch between transformer secondaries (or combinations thereof) that are optimized for use with speakers with impedances of 4, 6, or 8 ohms. Damn good engineering.

The CS300's line stage is said to provide approximately 20dB of gain. Because the Leben integrated amp lacks a phono stage, I used it in tandem with the Leben RS-30EQ, the company's relatively new standalone phono preamplifier.

Installation and setup
I used the Leben CS300 in place of my Shindo Masseto preamplifier and Shindo Corton-Charlemagne mono amplifiers. Because the former has a MM phono stage of its own, I supplemented the CS300 with the Leben RS-30EQ, using a 1.5m-long Audio Note AN-Vx interconnect to go from the outputs of the latter to the inputs of the former. I preceded the Leben phono preamp with either an Auditorium 23 SPU Standard or Silvercore One-to-Ten step-up transformer, for use with low- and high-output MC pickups, respectively. The Leben amp drove my Audio Note AN-E/SPe HE dynamic loudspeakers and, on occasion, my reconditioned Quad ESLs, ca 1959.

Footnote 2: Leben's CS600 integrated amplifier ($5795), which uses 6L6GC or EL34 output tubes, was reviewed by John Marks in his June 2010 "Fifth Element" column (Vol.33 No.6).— Ed.

The Luxman uses two EL34 tubes per channel in a class-A/B Ultralinear circuit, for a total of 25Wpc into 8 or 4 ohms, or 30Wpc into 6 ohms. (The amp has no apparent means of switching between different output-transformer secondaries or combinations thereof—but it does have those two sets of speaker connectors, plus a control knob for selecting either or both pairs, combinations of which may account for different load impedances and power specs.) The fixed-bias design, with a rail of about 420V (at power), makes use of both local and global feedback.

In addition to the four power pentodes, the Luxman SQ-38u contains seven small-signal tubes—four ECC83 and three ECC82—plus a pair of custom step-up transformers. Gain specs are 14dB for the line section and 37dB for moving-magnet phono. (A spec for moving-coil phono gain was not provided.)

Image may be NSFW.
Clik here to view.

The Luxman's component parts are distributed across a number of sturdy PCBs, the largest of which contains the preamp and phono-preamp circuitry, along with those seven dual-triodes—some protected with spring-loaded cans—and the step-up transformers for the phono section. Very chunky output transformers and a similarly large mains transformer are encased in metal covers; these, like most of the chassis' interior surfaces and even those tube cans, are finished in textured black enamel. Parts quality is good to excellent for this price range, and build quality is fine. It appeared that my review sample had been around the horn more than once, yet it still worked without a hitch.

Installation and setup
In my main reference system, the Luxman SQ-38u took the place of both my Shindo Masseto—itself a full-function preamp with an MM phono section—and my pair of Shindo Corton-Charlemagne mono amplifiers. The Luxman's own MC step-up transformers, with switchable gain for low- and high-output MC cartridges, performed acceptably well—and without detectable noise or grain—yet for consistency's sake I tended to precede the Luxman's MM phono section with my own step-up transformers: an Auditorium 23 SPU Standard for my low-output Ortofon SPU cartridge, and a Silvercore One-to-Ten for my higher-output EMT pickup heads. My primary line-level sources were a Decibel (v.1.0.2)–equipped Apple iMac with Wavelength Proton USB D/A converter and, as a CD player, a Sony PlayStation of indeterminate vintage. At the other end of the system were my Audio Note AN-E/SPe HE loudspeakers and, on occasion, my original Quad ESLs. I used only the Luxman's stock AC cord, and avoided tweaks of every sort.

Image may be NSFW.
Clik here to view.

The Luxman is supplied with a small remote handset, with Up and Down volume buttons and a Mute switch. As with most such things, I appreciated the gesture but tended not to use it.

Listening
From record to record, the Luxman SQ-38u was a big-sounding and altogether musically agreeable amplifier—yet one whose overall tonal signature wasn't egregiously tube-like. Tonally, the Luxman was nicely balanced from its bass through its trebles, with well-defined sonic contrasts. Musical sounds all had weight and force—even in the trebles, where higher notes had timbral body and, when called for, lots of impact. The amp didn't appear to invert absolute signal phase through either its line or its phono inputs.

When I listened to Jascha Heifetz's recording, with Fritz Reiner and the Chicago Symphony, of Brahms' Violin Concerto (LP, RCA Living Stereo LSC-1903), the soundfield was huge, instruments therein having good physical presence. Yet far from sounding opaque, there was a decent level of openness throughout the imaginary stage: individual instruments were present and clearly defined, the timpani at the rear of the stage scarcely less so than the fiddle up front. Timbrally, Heifetz's violin had the right signature and tactile "bounce"; the strings had just enough "bite" to sound realistic. It was a tubey sound but not a gooey one, and the orchestra sounded big and powerful, with very good momentum and flow.

On R.E.M.'s "How the West Was Won and Where it Got Us," from New Adventures in Hi-Fi (LP, Warner Bros. 9362-46320-1), the SQ-38u was on a par with the Leben CS300, also reviewed in this issue: rhythmic nuance and impact were good, but not up to the megabuck standard. The Luxman's bass depth was also acceptably good, although the Leben edged it out in that regard.

Taking advantage of the Lux's Mono switch—of the three integrated amps I reviewed for this issue, it was the only one that had such a feature—I decided to try some of my favorite 78s. The SQ-38u delivered decent punch and presence from Louis Armstrong's "You're Driving Me Crazy" (Parlophone R866), but not as much as my Shindo separates. Classical music on 78s fared better, such as the original 1932 discs of Yehudi Menuhin, Georges Enesco (sic), and Pierre Monteux leading an uncredited Paris Symphony Orchestra, in J.S. Bach's Double Concerto (Victor DM 932): perhaps the greatest Bach recording in my collection, and a stunning testament to Monteux's oft-overlooked greatness.

Image may be NSFW.
Clik here to view.

In a general sense, the Luxman's phono performance carried over to its use with line-level sources. A 44.1kHz rip of Mindru Katz playing the Bach-Busoni Organ Chorale and Prelude, BWV 659, had good musical flow, with decent timbral color and a nice, subtle sense of drama. Likewise, Marianne Ronez and Affetti Musicali performing Biber's Mysteriensonaten (Winter & Winter 910 029-2)—easily the best-sounding "Red Book" CD I've ever heard—was hypnotically musical and sonically colorful through the Lux (although sonic texture was somewhat lacking): Every instrument—even the bass and organ continuo—sounded almost eerily three-dimensional.

The Luxman sounded okay with the Quad ESLs, although the match didn't strike me as ideal. Compared to the Leben CS300, the Luxman allowed the piano's left hand to sound tighter and cleaner in Elgar's Violin Sonata in E Minor, with Midori and Robert McDonald (CD, Sony Classical SK 63331). Still, turning the Luxman's Bass knob a little to the left didn't hurt at all, especially with pop CDs with unrealistically equalized bass content.

Conclusions
To persist in my efforts to avoid bush-beating, I really enjoyed the Luxman SQ-38u. The Leben CS300 integrated, reviewed elsewhere in this issue, delivered a somewhat more tactile, believably textured sound with notably better bass extension, but the Luxman was consistently engaging—if not always quite as exciting—and its user interface was the best I've encountered in a contemporary amp.

I'd make some droll remark about high-end audio's anti-minimalist craze being right on schedule, but at Luxman Corporation—an 86-year-old Japanese company that ushered in the first age of high fidelity—domestic products with tone controls, mono switches, headphone jacks, solid wood cabinets, and tubes have remained in production for decades. The SQ-38u is no mere vintage reissue: At Luxman, it seems that vintage never went away. A good trip, happily recommended.

The new integrated is, essentially, a pair of Quad II amplifiers (ca 1953) powered with silicon rather than glass rectification and bundled with a perfectionist-quality control unit and a solid-state phono section. According to de Paravicini, the output section runs in pure class-A and is modeled after Walker's Acoustical circuit, in which a portion of the output transformer's primary winding is coupled to the cathodes of the KT66 output pentode tubes (as opposed to their screen grids, as in Ultralinear output circuits). Those transformers, of course, are designed by de Paravicini, and made in England to his very exacting specifications. The amp's output section is configured as autobias, and a bit of global feedback is used—although a less-than-typical amount is required, given that the Acoustical circuit itself provides local feedback, with apparently little loss of gain and power.

The small-signal tubes in the Quad II Classic—four 12AX7 dual-triodes and two 6922 dual-triodes—function as the phase splitter/driver stage for the voltage-hungry output section; the Quad's line stage is passive. The phono section is described as "a simple and elegant circuit with five transistors per channel." It's powered by a 36V drop from the rail voltage, and uses a combination of active and passive filters to achieve RIAA equalization. Gain is 42dB in moving-magnet mode, and 62dB when switched to moving-coil mode.

Image may be NSFW.
Clik here to view.

The Quad's innards are built on four PC boards: one for the input-switching relays, one each for the two pairs of output tubes, and one main board for the small-signal tubes and various power-supply components. (I never did locate a silicon full-wave rectifier, although I found a small diode bridge tucked away at the front edge of the main board.) The parts quality appeared to be fine—the output transformers were lovely to behold—and the build quality was excellent.

The casework and controls of the Quad II Classic are clearly influenced by the classic Quad 22 control preamplifier (ca 1959): The faceplate is shaped is the same manner, with a similarly big, beautiful volume control. The new Quad's source selector is a four-stop slider in a crescent-shaped slot with an old-style needle indicator above. From the outside, it gives that end of the Quad's faceplate a somewhat enigmatic smile. Not inappropriately.

Image may be NSFW.
Clik here to view.

Installation and setup
Unlike the Leben CS300 and Luxman SQ-38u integrated amps, which I also review in this issue, the Quad II Classic Integrated is physically deeper than it is wide; nevertheless, it fit well on the top shelf of my Box Furniture D3S rack, with all four feet making full contact with the wood surface (a requirement for optimal performance, I think, but that's a topic for another day). The full-function Quad replaced both my Shindo Masseto preamp and Shindo Corton-Charlemagne monoblocks, with no need for an external phono preamp. Per my usual practice, I fed the Quad's MM inputs with signals from my Auditorium 23 SPU Standard and Silvercore One-to-Ten phono transformers, rather than use just the Quad's switchable MC input with MC pickups—but the results weren't always what I expected. Whereas I tend to prefer loading and stepping-up a cartridge with a transformer rather than an active stage, the Quad's solid-state phono preamp seemed to narrow the gap between those two approaches, differences between which were subtle.

Again, my line-level sources were an Apple iMac equipped with Decibel playback software and a Wavelength Proton USB D/A converter, along with an aging Sony PlayStation for a CD player. Loudspeakers were my Audio Note AN-E/SPe HEs and, where indicated, my original Quad ESLs. I used only the Quad amp's stock AC cord. During all of my listening sessions, it was apparent that the Quad amp did not invert absolute signal polarity.

Listening
The first things that struck me about the Quad's sound were its up-front presentation—this wasn't at all a laid-back performance—and its readily apparent sense of drive with up-tempo music. The musical timing of all recordings was well preserved, although the Quad II Classic sounded especially fleet of foot with quick, snappy acoustic music such as David Grisman's "E.M.D.," from the first, eponymous album by the David Grisman Quintet (LP, Kaleidoscope F-5).

With line and phono sources alike, the Quad's midrange was lush and warm, with decent contrast between timbrally light and dark sounds in the mix, albeit less natural texture than I expected. In my system, driving the Audio Note cone-and-dome speakers, the Quad's most notable shortcoming was a lack of extension at the frequency extremes—especially the bottom end. Through the Quad II Classic, the dB's'"Happenstance," from their Repercussion (LP, Albion ALLP 400032J), sounded overly midrangey: the inverse of the rightly maligned hi-fi "scoop." There was fine punch and presence in the electric guitars, and the drumming sounded acceptably forceful, but the electric bass lacked that last bit of extension and power, and the Quad's slight lack of top-end sparkle conspired to make the voices sound a bit dull and lacking in presence.

Chamber music and very well-recorded symphonic fare were better served by the Quad's timbral balance. Of the former, the II Classic did a splendid job with the richly recorded sonorities of the Affetti Musicali's recording of Biber's Mysteriensonaten (CD, Winter & Winter 910 029-2). It missed a lot of the natural texture in that recording, but the Quad did allow the music to unfold with near-hypnotic intensity. As to the latter, recordist René Laflamme's masterful recording of Hindemith's Escale Romantique, performed by Daniel Myssyk and the Ensemble Instrumental (ripped from the Fidelio sampler Escales, no catalog number, www.fidelioaudio.com), was clear and spacious through the Quad amp, yet engagingly rich and warm.

Image may be NSFW.
Clik here to view.

Back to popular fare: Stevie Wonder's "Pastime Paradise," from Songs in the Key of Life (LP, Tamla T13-34062), sounded acceptably good through the Quad, but a little flat compared with my reference Shindo electronics; the latter did a distinctly better job of unraveling the sounds of the combined handclaps, strings, and percussion instruments. I also found myself wishing that those and other sounds would stand out more from the mix, and attain a greater sense of spatial presence.

Thus far I was charmed but unexcited by the Quad II Classic Integrated; replacing the conventional Audio Note loudspeakers with my original Quad ESLs turned up the passion considerably. Interestingly, the amp's lack of treble air—and commensurately midrangey sound—persisted, yet I found those characteristics to be far less objectionable when coupled with the naturally spacious presentation of the electrostatic panels. More important, I think, was the manner in which the amp and speakers jelled throughout the bass and lower midrange. "Überlin," from R.E.M.'s Collapse Into Now (CD, Warner Bros. 525611-2), worked very well through this combo, with fine instrumental sound and excellent momentum. And MoFi's recent reissue of the Band's Rock of Ages (SACD/CD, Capitol/Mobile Fidelity Sound Lab UDSACD 2046) sounded snappy, impactful, tonally well-balanced, and surprisingly spacious. The Quad-Quad combo let me enjoy that recording more than ever.

I was reluctant to turn back—and wound up spending more time with this pairing than I did with the Quad II Classic Integrated driving my Audio Notes. Through the electrostats, the Quad amp sounded brilliant with my favorite collection of Elgar's shorter works, by Paul Goodwin and the English Chamber Orchestra (CD, Harmonia Mundi HMU 907258). String tone was marvelous throughout, and the percussion that opens the third section of the Nursery Suite was appropriately startling, with more impact than Quad skeptics would expect. Even an admittedly grotty bootleg recording of Procol Harum playing "Repent Walpurgis" at the Fillmore East in 1969 (CD, Third Eye, no catalog number) sounded magnificent through this pairing—again, with surprisingly good impact on B.J. Wilson's masterful drumming and guitarist Robin Trower's feedback-drenched dive-bomb ending. The finest accolade I can offer any domestic playback machine applied here: It was satisfying.

Conclusions
Inasmuch as Quad Electroacoustics set out to recast their nearly-60-year old Quad II as a full-function integrated amp, they have thoroughly succeeded: The Quad II Classic Integrated offers the same basic sound—perhaps tidied up a bit—in tandem with a modern solid-state phono section, itself remarkable for blending well with its tubey surroundings. Also like the Quad II of old, the new amp lacks the openness and texture of some modern designs—but makes up for it by singing with Quad's own ESL, itself not known for playing well with modern children.

If Quad ESLs—perhaps even other electrostats—are your everyday speakers, and if your aesthetic sensibilities are tuned to its very vintage look and feel, the Quad II Classic Integrated will no doubt delight you.

Image may be NSFW.
Clik here to view.

The HK 990 is a product for the serious audiophile: a full-fledged, analog, two-channel integrated amplifier with serious output clout (claimed instantaneous output current of ±200 amps). It also has digital inputs to handle modern source components, and offers bass management as well as home-theater bypass for the left, right, and up to two low-frequency effects (LFE) channels. Some other integrated amps have some of these features, but the HK 990 is, as far as I know, unique in being a two-channel product that includes measurement-based room equalization. And it's gorgeous.

Out of the box
Having seen the glossy catalog sheets, I was unprepared for the HK 990's weight and solidity. My sample had been handled by other reviewers, and its volume knob had come off in the carton. But after I'd maneuvered this 43-lb baby out of its swaddling, I popped the knob back on, settled the amp in place on its four damping pods, and plugged it into the AC line.

A thin black line divides the HK 990's front panel into upper and lower halves. On the upper half are the large alphanumeric display and illuminated volume knob, while in the lower half are only the Harman Kardon logo and a headphone/microphone jack. Close examination reveals that much of the thin black line comprises eight thin pushbars whose labels are dim and small. The reticence of this concession of convenience to design leaves the front panel looking cool and clean, and anyway, the most common controls also appear on the multifunction remote control, where they're more handy, and more easily differentiated by shape and position.

The rear panel's many connectors are grouped into analog inputs (including XLR, moving-magnet, and moving-coil), digital inputs, and HT processor bypass inputs, to be directly connected to the HK 990's power amp stage. A look inside the HK 990 revealed that these groups reflect the array of specialized circuit cards on the main chassis. One input, looking curiously like an RJ-45 Ethernet jack, is labeled HRS, for High Resolution Synchronization, a proprietary H/K connection with which the HK 990 can be linked with the digital outputs of H/K CD players, such as the matching HD 990. The HRS link carries both the system and data clocks of the amplifier and the datastream from the CD player, which locks itself to the amplifier's clock to ensure a jitter-free connection. (I am not a fan of proprietary interfaces.)

The only modern input sources the HK 990 seems to lack are Ethernet and USB. While I would certainly welcome their inclusion, I sidestepped the omission by plugging a Logitech Squeezebox Touch music server into one of the H/K's digital coax jacks. The HK 990 also offers L/R analog and dual subwoofer RCA outputs, and a digital output jack for recording. In addition to power, RS-232, and trigger and IR ins/outs, there are two sets of multiway speaker binding posts for each channel's output, these selectable via front-panel switches.

Setup for the new millennium
I hooked up the HK 990 to my Sony XA-5400ES SACD/CD player, Oppo BDP-95 universal Blu-ray player, and Squeezebox Touch, and to a pair of Aerial Model 7T speakers and my trusty JL Audio Fathom f113 subwoofer. When I touched the On/Off bar in the left corner of the HK 990's front panel, next to the amber standby indicator, both that indicator and the rim of the volume knob glowed a soft blue. Gracious in demeanor as well as appearance, the large blue display said, "Setting up, please wait," and in a few seconds the HK 990 was ready to go.

Image may be NSFW.
Clik here to view.

That's the simplest way to get the HK 990 up and running, but it bypasses what makes this product special. Its proper setup is a procedure akin to setting up an AVR. First, you choose which input jacks to associate with which input selection name; eg, balanced analog, unbalanced analog, optical or coaxial digital. For the phono selection, it's MM or MC (footnote 1). Next, you need to choose between Manual or Automatic setup of bass management. This is easy. With Manual, you inform the HK 990 whether you're using one or two subwoofers, then pick a single crossover frequency from 20 to 200Hz, in increments of 10Hz.

More wisely choose Automatic and you enter H/K's EzSet/EQ mode. This begins with the same enumeration of subs, and you can specify a crossover frequency or let EzSetEQ do it. From there, the Automatic route opens up a series of measurements and calculations that result in the system being equalized in three bands. Plug the calibration microphone (provided) into the mike/headphone jack on the front panel and follow the prompts on the HK 990's display (or in the owner's manual):

1) With the mike at the primary listening position, the HK 990 sends a sweep tone to each speaker, and uses the information it collects to apply a Far Field mid-frequency (20Hz–1kHz) correction to the main speakers.

2) Place the mike 60cm in front of each main speaker, and the HK 990 emits a sweep tone that it uses to apply a Near Field, high-frequency (>1kHz) correction to the main speakers.

3) Place the mike back at the primary listening position. The HK 990 sends a low-frequency sweep of considerable power through each subwoofer, then applies a BassQ correction to the subs.

EzSet/EQ also sets channel levels and crossover frequency, but these can be easily tweaked afterward. H/K also gives the user considerable freedom in applying the EQ. You can switch, on the fly, with the remote control among DSP (no EQ), EQ1 (BassQ only), EQ2 (BassQ and Far Field), and EQ3 (BassQ, Far Field, and Near Field). In fact, since the HK 990 can digitize analog inputs, the list of options is longer and runs from Direct Path, DSP Path, EQ1, and EQ2 to EQ3.

Anthem is different. The company has long enjoyed success as a manufacturer of high-end pre-pros, and only recently has introduced its first range of AVRs, whose technology and philosophy are derived from Anthem's pre-pros (footnote 1). Not so surprisingly, in view of market demands, Anthem's MRX line of receivers is much less expensive than the Anthem or Anthem Statement pre-pros. A significant portion of this must be due to the fact that they're made in China rather than in North America like their pre-pros. Anthem's current top-model AVR is the MRX 700 ($1999).

Unpacking the 35.4-lb MRX 700, I was encouraged. It was more solid and rigid than I'd expected from the price and its front panel, though busy with controls, seems more businesslike than that of a typical AVR. The controls are clearly and usefully labeled. At top left is a large navigation compass, at top right a volume control of similar size. Below the compass are buttons for Info, Setup, Menu, and Presets Scan, and below the volume knob is an identical array for Dolby Volume, Channels, Audio, and Mute. Between these groups is a row of eight source-selection buttons. At bottom left are four jacks (Phones, USB, composite Video, and L-Audio-R), and at bottom right are four more buttons: Display, Zone, Zone 2 power, and Main power. While the front panel seems the very model of modern organization, the remote control was less intuitive in its organization; it took me a couple weeks to master it. Unfortunately, the backlight control button is clustered with others of similar size and shape, which makes it hard to find in the dark. A second, simpler remote is provided use with Zone 2.

The MRX 700's rear panel has the requisite multiplicity of connections; with Canadian practicality, Anthem provides full-page illustrations of it and the front panel in the manual. That made setup easy, and was helpful as I changed connections along the way. I hooked up the MRX 700 with HDMI cables to my cable box and high-definition disc players, and with coax for digital input from the players. I used six Belkin Silver Series RCA cables to connect the outputs of my Oppo BDP-83SE universal Blu-ray player, and a set of Kubala-Sosna Anticipation RCAs to the inputs of my Bryston 9B-STT power amplifier. The MRX 700 has no balanced inputs or outputs.

Image may be NSFW.
Clik here to view.

After power-up, I dove into the setup menus, whose graphics and red-and-black color scheme suggest a brochure from the Royal Canadian Mounted Police, and overlay the ongoing video display. Audio signals, too, remain audible during menu access; I could immediately hear the effects of changing the audio mode. After assigning and naming inputs, I was able to assign different default processing modes for each audio format. I also set up bass and speaker management from experience (and tape measure), and spent a week or two using no room correction, to get a handle on the MRX 700's inherent sound quality. However, the outputs of my subwoofer and satellite speakers didn't smoothly integrate, so I reset the front speakers to Large.

The overall sound was very impressive in both two channels and surround, with the timbres of voices and instruments quite natural. The MRX 700's fairly good spectral balance allowed for the influence of room modes in the bass. I also switched briefly to the MRX 700's own amplifiers, and, as you might expect with my all-Paradigm speaker system (Paradigm and Anthem share the same corporate parent), there was sufficient oomph, and the balance was little different from those with the Bryston 9B-SST—another Canadian product, eh? The Bryston had the stronger grip on the bass and, despite the room modes, was a bit cleaner through the low and midbass. I suspect that the differences might have been even smaller with Anthem Room Correction (ARC), the company's room-correction software (included), engaged, but I didn't get around to trying it with the Anthem's own amps.

The MRX 700 comes with a complete ARC toolkit: software, calibrated microphone and cable, and tripod mike stand. It's nearly the same version of ARC as is used in the Anthem Statement D2v processor: the MRX receivers have slightly less DSP power, and correction is limited to below 5kHz (the D2v can correct up to 20kHz). I've used ARC with the Statement D2v processor and, more recently, in its PBK incarnation, the bass EQ software for the Paradigm Sub-15 subwoofer, so I'm nearly as familiar with ARC as with Audyssey. ARC runs on a laptop and connects to the AVR via a USB or serial port, while Audyssey's connection is determined by the particular device it's installed in and can be run with or without wires. ARC can handle up to 10 mike positions; while Audyssey MultEQ Pro can handle up to 32 positions, less sophistication versions of the software handle three, six, or eight mike positions.

Used as fully automated procedures, ARC and Audyssey are equally easy to use, but ARC's Manual mode is friendlier than MultEQ Pro's, as it doesn't require a "key," doesn't ask for details of system setup (though this can be ignored in MultEQ Pro), and makes saving and retrieving settings a bit simpler. In addition, all ARC-equipped devices can store two sets of EQ corrections, one each for Movies and for Music—but, of course, you can repurpose them. On the other hand, MultEQ Pro permits the hands-on editing of individual target curves, albeit only within Ò3dB. With ARC you can adjust the room gain (a small mid-bass boost that preserves the warmth caused by the acoustics of the typical listening room) and dictate how high in frequency the EQ is applied. The latter is particularly attractive to the many who believe that the correction of room modes should be limited to frequencies below the Schroeder Frequency, and that a system's high-frequency performance should be determined by speaker choice and placement and the user's taste.

I ran ARC with the same nine mike positions I usually use. I chose a 2kHz upper limit for EQ to go along with the 3.9dB of room gain detected by ARC. The results were a definite improvement in the extreme bass—the subwoofer plot was smooth and extended—while the wider frequency-response swings in the main channels were restricted to the region below the crossover frequencies, which ARC put at 60Hz for the three Paradigm Studio/60s in the front, and at 80Hz for the two Studio/20s in the rear. Perhaps because the low end was cleaned up, or perhaps because I know that these speakers have sounded better, I became increasingly critical of a slight nasality in male voices, as well as a lack of presence in flutes, cymbals, and upper violin strings.

The ARC graphs told me why: All of the main channels suffered from a 2–3dB trough from just below 1kHz to 6kHz. ARC corrected this up to about 2kHz, which resulted in an even more sharply defined depression above that point. So I extended the ARC correction to 5kHz (the MRX's default setting) but cut the room gain to 2dB and increased the subwoofer's upper cutoff to 120Hz. VoilÖ! This sounded, and measured, pretty close to ideal, with the frequency response of every main channel within Ò2dB, from above its crossover up to about 20kHz. The Paradigm Sub-15 measured nearly flat down to 15Hz, and the XTZ measurement system revealed that ARC had virtually erased a large mode at 31.5Hz, as well as several smaller ones. Only the rear speakers still had a small modal bump at around 150Hz, though this was now 8dB lower than without ARC. I could sit back and just play the music.

With everything I tried, in stereo or multichannel, I found it hard to find fault with the MRX 700. Voices were distinctive and rounded, with fricatives clearly defined. The low and midbass were tight and extended, as one might expect, but also packed a good wallop when that was called for. However, the real payoff was the overall integration of the sound, both harmonically and spatially.

I have lived with this combination of speakers and room for many years, and it has never sounded better than with the Anthem MRX 700. That's not to say that other pre-pros haven't been as satisfying, but there were differences. The Integra DHC-80.2 seemed a bit more open, but the slightly warmer-sounding Anthem wasn't lacking in transparency. If anything, the MRX 700's reproduction of inner voices was more subtle. In terms of soundstage depth and the seamlessness of the illusion of surround sound, the MRX 700 was excellent, and shared with the eminent Classé SSP-800 the ability to re-create eerily distinct lateral voices without smearing.

Although it is not multichannel, Internet radio streaming was more reliable and sounded cleaner through the Anthem MRX 700 than through the Arcam, Marantz, and Integra pre-pros I had to hand. In particular, RCO Live, the streaming service of Amsterdam's Royal Concertgebouw Orchestra, would cut out every so often with those models; the Anthem never did. Moreover, the imposition of AnthemLogic-Music, a proprietary algorithm that expands the signal to up to 6.1 channels, opened up the soundstage of these lossy compressed streams without affecting the tonal balance or frontal imaging. The reasons for the success of AnthemLogic-Music may be that it doesn't add a center channel, and that its use of the surround channels is very subtle. Whatever the factors, with the MRX 700 I've been enjoying music via the Internet more than ever.

The MRX 700 is a fully competitive preamplifier-processor that lacks only the XLR outputs commonly found at this price although the inclusion of an AVR's power amps makes for added value. However, the MRX 700 ($1999) is so similar to Anthem's MRX 500 ($1649) and MRX 300 ($1099) that a single user's manual serves for all three. The differences include the power output (respectively, 90W vs 75W vs 60W into 8 ohms, five channels driven) and a few features. Only the 700 includes HD radio and iTunes tagging, and the 300 lacks a USB input for flash drive or hard disk, and Internet radio via Ethernet. The MRX 700 also has a toroidal power transformer instead of the E-I cores in the 500 and 300. Remarkably, all three models share all other specs and features, including ARC. It might be safe to extrapolate from my conclusions about the MRX 700 and consider one of the less expensive models to use as a pre-pro or AVR, assuming its features and power output suit your needs. Bravo.

Ortofon, the Danish firm that's been in business longer than any other manufacturer of phono gear, has confounded all that with the Xpression ($5399): an entirely new moving-coil pickup head designed from the ground up. It has surprised even me.

The Xpression derives from the Ortofon MC A90, a technically advanced moving-coil cartridge that our own Mikey Fremer has called revolutionary. That limited-edition product combined a number of innovations, including a tiny cylindrical field-stabilizing element (FSE), to counteract disturbances the magnetic field; and a wide-range damping (WRD) system, made of tiny rubber and platinum discs, said to enhance both tracking and timbral neutrality. But the A90's real calling card was the manufacturing process used to create its body: selective laser melting (SLM), whereby individual particles of stainless steel are welded together, one layer at a time, to create a complex, homogenous structure in which density and self-damping ability are more than merely random.

Lest you think that SLM is just another initialism cooked up by a manufacturer or its advertising agency, I can assure you that it isn't. This computer-driven manufacturing technique, though still in its infancy, has already gained a foothold in the manufacturing of titanium-alloy orthopedic appliances, where the need for precision and consistency is obvious (footnote 1).

Thus the Ortofon Xpression is a unique blend of the new and the old. Its compliance is on the low side, and the pickup head's 28gm mass is commensurate with that. The recommended downforce is a substantial but not scary 2.6gm. Impedance and output are lowish, at 4 ohms and 0.3 mV, respectively, and the stylus profile is among the most advanced on the market: a highly polished sample of Ortofon's Replicant 100.

This new Ortofon is designed and built as a drop-in replacement for any G-style pickup head. (I measured a collet-to-stylus dimension of precisely 52mm.) It has an SME-standard four-pin connector at one end and an axial finger-lift at the other, both gold-plated. The Xpression looks decidedly equine from some angles, but when viewed directly from its left side it resembles the head and neck of a friendly, googly-eyed Brontosaurus.

Used in either my EMT 997 or Schick tonearm and loaded with my Auditorium 23 step-up transformer, the Xpression proved itself to be much more explicit than my original SPU—more detailed, more open, more tactile, more revealing of nuance and technique—without sounding the least bit hi-fi. The new Ortofon sounded every bit as solid, colorful, dramatic, and forceful as the old one. (I admit, I wouldn't normally have expected such solidity, such lack of fussiness, from a pickup with other than a spherical stylus tip.) The Xpression offered insights at which my Bakelite-bodied SPU has only hinted. The one that stands out in my memory—chiefly because I'm still listening to the record as I write this—is the manner in which drummer Dave Mattacks draws out his more broadly spaced cymbal crashes throughout Fairport Convention's House Full (Hannibal HNBL 1319): difficult to describe, easy to appreciate and enjoy.

Playing Ravel's Ma mère l'oye, with Ernest Ansermet and the Suisse Romande Orchestra (LP, Decca SXL-2062), the Xpression astonished me from the first few measures. Each orchestral swell came across with a degree of force and impact suggested by no other cartridge I've owned: It was almost as if the Ortofon were magnifying the dynamic contrasts within the recording—an effect not unlike that of the Hommage T1 and T2 phono transformers I've written about in past columns. Tonally, the Xpression was more extended in its treble range than my SPU, but not to the point of brightness, nor at the expense of low-frequency richness. The contrabassoon that makes its entrance during the Prelude was just as deep and weighty with the Xpression as with the older SPU—and was better defined in pitch and presence.

The Ortofon Xpression was so outstandingly dynamic and communicative that I began to mistrust my senses: During its first day in my system, did I select, by chance, recordings that just happened to show it off? I stopped that afternoon, and swapped back in my standard Ortofon SPU. The difference was real: Love my older Ortofon though I do, the Xpression was clearly more dramatic, with no penalty in texture or color.

Bear in mind: While the Ortofon Xpression found more and wider dynamic contrasts within otherwise average-sounding records, it did not improve the sound of records that were poor to begin with. (File under: This shirt will not make you fly.) Many selections on Crosby, Stills & Nash's debut album remained dense and woolly. Fritz Reiner and the Chicago Symphony Orchestra's The Reiner Sound still sounded dull. Eno's delightful Taking Tiger Mountain by Strategy still had a little too much bite in the treble. Sir Colin Davis's recording, with the English Chamber Orchestra, of Mozart's Symphonies 28 and 38 still had a peculiar and hard-to-define congestion and midrange glare. But the Xpression's high-tech stylus profile was as quiet in the groove as anything else I've tried, making it easier than usual for me to enjoy heavily worn samples of otherwise good recordings.

The Xpression confounded more than my expectations regarding new-vs-old technologies (see "As We See It" on p.3): Delighted though I am to see and hear such a product in the second decade of the 21st century, the Xpression brings with it a certain disregard for convention and for the staid logic of commerce—not unlike the best music. That such a technologically advanced company can still take a chance such as this is a blessing.

A Haut with a heart of gold
The products of Shindo Laboratory occupy an uncrowded space in the audio market: not quite mass-produced, not quite bespoke. Virtually all of Ken Shindo's amplifiers and preamplifiers are designed around parts from his extensive collection of vintage tubes, capacitors, resistors, and the like, and of the necessities that remain—especially the distinctive steel casework, made to order for each model—Shindo orders only 10 or 20 at a time. Subsequent production runs are determined by a combination of consumer demand and sufficient reserve supplies of vintage parts.

Image may be NSFW.
Clik here to view.

That approach brings with it the opportunity for Shindo-san to revise every model virtually at will. Those changes can be major or minor—a single-ended amplifier called the Lafon, which has been built with three very different power tubes over the years, is a fine example of the former—and the designer appears to regard them as artistic variations rather than as improvements per se. Just as there are different sonic and musical characteristics to every Shindo amplifier model—many of which would otherwise seem similar, based solely on power output—so there are often distinctions between different samples of the same Shindo model.

When Steve Leung, of importer Valve Audio System (VAS) Industries, commissioned the design and manufacture of the SP-10A, he was driven by his love of the sound of the 6L6GC output tubes in his vintage McIntosh MC-240. Leung believes that the smooth, fluid midrange character of the 6L6GC power tube is superior to that of either the EL34 or KT88 tubes, and that this makes it a better choice for recordings of vocal music. The 6L6GC also "feels more powerful" to Leung, which leads him to think it better supports an amplifier driving difficult speaker loads. In addition to the four 6L6GCs, the SP-10A also has one 12AU7 phase splitter/driver tube per channel, and one 12AX7 input tube, which is shared between channels. The amp, which is wired point to point, has four line-level single-ended inputs, and both 4 and 8 ohm speaker outputs. The minimalist remote control has buttons for input selection, volume, and mute.

Image may be NSFW.
Clik here to view.

Image may be NSFW.
Clik here to view.

The SP-10A is a stunning piece of artistic machinery. Its faceplate of brushed aluminum fronts a cabinet of gorgeous cherrywood (walnut and piano lacquer are also available), which made it a natural visual match for the Epos M5i bookshelf speakers I used for the test, driving the speakers from the Cayin's 4 ohm output transformer tap. Not only did the cherry finishes match precisely, I'd swear that the high-quality wood used for all three cabinets was cut from the same tree.

The Cayin is not only solidly built; by a wide margin, it's the densest component I've reviewed in 25 years: an integrated amp weighing 44 lbs but measuring only 16" wide by 8" high by 13" deep.

I found two features of the SP-10A quite odd. First, it has a small speaker, which emits a long, high-frequency beep at turn-on, and shorter beeps when an input is selected, either with the knob on the front panel or with the remote control. Second (Mikey Fremer will love this), the Cayin automatically resets its input selector to CD on power-up, regardless of which input was engaged when the unit was powered down. As I spent much of my time with the SP-10A listening to LPs, I found this somewhat annoying.

Listening
In my very first listening session with the SP-10A, I was hit between the ears by what turned out to be its greatest strength: a seductive, lush, natural, palpable midrange with uncanny resolution of inner detail and ambience. I immediately grabbed some albums with well-recorded vocals. In the a cappella rendition of "Red River Valley" on Cassandra Wilson's Thunderbird (CD, Blue Note 72438), her voice was silky, rich, and voluptuous, every nuance clearly delineated by the Cayin's linear resolution of low-level dynamics. I had a similar experience with "The House of the Rising Sun," from Bob Dylan (LP, Columbia CL 7579)—all the subtle, husky, nasal details of Dylan's unique voice were preserved intact.

Image may be NSFW.
Clik here to view.

And it was obvious how engineer Dave Hassinger had bathed Mick Jagger's voice in way too much reverb against a background of pristine harpsichord in "Lady Jane," from the Rolling Stones'Aftermath (UK LP, Decca 820 050-1). But what really floored me was listening to "Some People's Lives," from Janis Ian's Breaking Silence (CD, Morgan Creek/Analogue Productions CAPP 027). That track—probably the most overplayed recording in my house—is quite revealing of subtle nuances. Through the Cayin, I heard Ian's intake of breath before each sung line more clearly than I ever had.

Classical recordings that spotlight woodwind instruments also showed off the Cayin's midrange capabilities. Antony Michaelson's clarinet, in his performance of the Mozart Clarinet Concerto with Robert Bailey leading the Michaelangelo Chamber Orchestra (CD, Musical Fidelity MF 017), was vibrant, airy, and holographic, with the purest rendition of his instrument's midrange timbre that I've heard in either of my reference systems. The Cayin could also unravel complex midrange textures. Pianist McCoy Tyner's chunky, dissonant chord clusters in "Passion Dance," from The Real McCoy (LP, Blue Note BST 84264), were clear and clean—it was very easy to pick out each note within the natural dynamic envelope of his phrasing.

I found the SP-10A's high-frequency reproduction to be detailed, airy, and clean, though I've heard other integrated amplifiers with more top-end sparkle. Well-recorded strings were slightly sweetened, in a tubey, slightly euphonic way—whether it was the interplay between violinist Jenny Scheinman and cellist Hank Roberts on Bill Frisell's Richter 858 (CD, Songlines SGL SA 1551-2), or the massed strings in the more dramatic passages of Fritz Reiner and the Chicago Symphony Orchestra's reading of Hovhaness's Mysterious Mountain (LP, RCA Living Stereo LSC-2251). In either case, strings sounded silky, attractive, and not overly cloying.

At the bass end of the spectrum, George Duvivier's bass solos on Eric Dolphy's Eric Dolphy (LP, Prestige 24008) were reproduced with natural woodiness, tactile definition, and not a trace of plumminess. The SP-10A was definitely not a hard-rock amp, however. I cranked up "Government Hooker," from Lady Gaga's Born This Way (CD, Streamline 0602577218385), to fairly loud levels, but the music never got much louder than forte during the accented downbeats. The Cayin didn't sound as if it was clipping, and the bass didn't distort or lose definition; the amp just seemed to run out of high-level dynamic gas when played so loud.

Image may be NSFW.
Clik here to view.

The SP-10A's ability to render ambience and a sense of room acoustic was most evident with classical vocal music. In Luc Ferrari's Chansons pour le corps (CD, Mode 81), with Yves Prin conducting the Nouvel Orchestre Philharmonique, the broad range of soprano Elise Caron was vibrant and natural—even the top of her range was perfectly reproduced, with no sense of harshness or rolloff, her voice floating on a bed of air throughout the recording.

The SP-10A's reproduction of transients was lightning-fast with every recording I tried, but with neither smearing nor unnatural sharpness. In Bang on a Can's recording of Terry Riley's In C (CD, Cantaloupe CA 21004), the transient integrity of the eclectic orchestration was preserved perfectly—the group breathed as in a live performance. The interplay of percussionists Reggie Workman and Doug James on Gaia, an early free-jazz album by pianist Marilyn Crispell (LP, Leo LR152), gets pretty cacophonous at points, but there was no trace of overhang or hardness, and the senses of air and extended harmonics were perfectly intact.

The recording that brought all of the Cayin's strengths together was Charles Wuorinen's performance of his Speculum Speculae with his Speculum Musicae ensemble (LP, Nonesuch 71300). Richard Fitz's cacophonous percussion interludes are separated by long passages of space, and every one of his battery of instruments popped out of thin air with perfect transients, and—for instruments with significant high-frequency content—a shimmering, extended, natural harmonic structure.

Comparisons
I compared the SP-10A ($2195) with my current reference integrated amplifier, a Creek Destiny ($2395) . Each was a very neutral performer with a warm, inviting midrange. However, their overall presentations of texture in the midrange, particularly of voices, differed a bit. The Cayin seemed to bathe voices in a golden glow, while the Creek enveloped them in a silvery sheen. The Cayin resolved a significantly greater amount of detail and ambience in the midrange (though the Creek was still excellent in this regard), and reproduced a linear, low-level dynamic envelope from pppp to p even better than the Creek did. But the Destiny was far superior in its rendering of high-level dynamics.

The Creek's high frequencies were more extended but less airy than the Cayin's, but both amps' overall timbral presentations were quite natural. Their reproductions of the midbass were equally natural, though I felt the Creek was capable of a bit more bottom-end slam. Each amp's reproduction of transients was very fast and natural, but again, with different textures. The Cayin's transients seemed a bit softer and more delicate, while the Creek presented slightly sharper envelopes.

Summing up
Cayin has created another winner with the SP-10A—a neutral, detailed, and involving integrated amplifier that performs at a level higher than one would expect for the price. And that cherry cabinet is dead gorgeous.

Candeias has brought to the B.M.C. line the same distinctive engineering and industrial design that distinguished his OEM creations. With its unusually large, centrally mounted, illuminated power meter, you won't confuse the Amplifier C1 with anything but another B.M.C. model.

Candeias's ideas extend beyond engineering. He's developed a new distribution and retailing model in which retailers needn't buy and warehouse expensive stock. Consumers instead purchase directly from the distributor, who returns a percentage to the dealer who brokered the purchase. This arrangement also avoids, among other brand-destabilizing retailing foibles, price-dropping selloffs as older products are replaced by newer ones.

More than a pretty faceplate
Though the fully balanced Amplifier C1 ($7990) is claimed to output 175Wpc into 8 ohms or an impressive 330W into 4 ohms, and has two pairs of balanced inputs, three pairs of single-ended inputs, and a volume control, Carlos Candeias does not call it an integrated amplifier. He prefers "a power amplifier with variable amplification and input choice." His reasons for this will become apparent in my description of the Amp C1's circuit design.

The power supply is generously built, as you might expect for a design that's claimed to almost double its output with a halving of load. At a time when some manufacturers are moving toward lightweight, energy-efficient, switch-mode power supplies, Candeias's 88-lb Amp C1 makes use of the largest toroidal transformer I've ever seen in an integrated amplifier—or whatever you want to call it. I saw this transformer only because the Amp C1 was DOA. Candeias paid a visit a few days later, removed the bottom plate, and found that UPS had literally drop-shipped the unit, shaking loose one of the clip-on connectors.

The Amp C1 is physically impressive. A large porthole that contains the two power meters dominates the faceplate, and also displays the selected input and indicates the volume control setting in 1dB increments, from 00 to 66. The rest of the design adds to the C1's visual appeal: an unusual wraparound heatsink, a central vent-like channel, and an aluminum faceplate with a satiny finish. There's a large Power knob to the meter's left; to its right, a matching Volume knob. A small button between the meter and Power knob controls the illumination level, and a matching one between the meter and the Volume knob toggles through the inputs—which can also be selected via the multi-component remote control of machined aluminum. Unfortunately, there's no direct selection of inputs.

Image may be NSFW.
Clik here to view.

The rear panel is neatly and symmetrical laid out: single-ended input 1's left and right jacks are farthest apart and input 3's jacks are closest together, to either side of the central IEC AC inlet, with pairs of binding posts at either end. Two fans set into the bottom plate send air through the interior. Though they probably did turn on at some point, I never heard the fans operating.

Everything about the B.M.C. Amplifier C1, from its looks to its build quality to how it operates, has been designed to produce a substantial, even luxurious experience at a price that's (dare I say it?) reasonable, considering its innovative circuitry and high power output.

The Unique Circuit
When we met following his visit, at the 2012 Consumer Electronics Show in January, Carlos Candeias told me that the Amplifier C1's circuit is neither class-A nor class-B; instead, he calls it Load Effect Free (LEF). He doesn't use energy-wasting class-A operation to eliminate crossover distortion and have transistors operate within a more or less linear range, because such designs still usually require some global feedback to further reduce distortion. If the distortion never happens in the first place, due to the choice of transistor operating characteristic, it doesn't need to be corrected.

Candeias explained two weaknesses of transistors: variations in voltage and variations in current. Then, rather than explain his system's technical innovations in engineering terms, he compared it with a painter standing before a large canvas, who would be able to paint "more beautifully" directly in front of him, and less so as he was forced to stretch above and below that area. Compound the painter's problem, he said, by placing around his arm two large rubber bands, one of which affects upward-going motions, the other lower-going motions. When he tries to extend his reach up or down, he'll feel how nonlinear his muscle movements are, and he won't be able to paint as beautifully. But as long as his arm remains in the middle of his range of motion, the bands won't affect his painting. Having someone telling you to "Go down! Go up! You're painting ugly!" is analogous to applying global feedback.

To avoid needing such feedback, Candeias explained, "we built a platform below the painter's feet"—in technical terms, a "floating cascode." When the painter moves his arm up or down, the floating cascode accordingly shifts his entire body up or down. So the distortion caused by the restrictions imposed by the rubber bands (voltage, in Candeias's analogy) moves from the painter's hand down to the floating-cascode platform, which is not connected to the speaker—in terms of affecting the sound, it's rendered irrelevant.

But that, Candeias said, is the easy part. The hard part is applying the power required to stretch the rubber band. Extending the analogy, he said that the painter has two friends clamping the two rubber bands, and they're watching the painter's muscles for changes in the amount of power he needs to move his arm. When the amount of power needed changes, the current needed to move the arm without restriction is supplied from a separate source. This allows the Amplifier C1 to deliver the required voltage to the speaker completely independent of the current, and to deliver it from separate sources.

What accounts for the renewed popularity of this product category? Cost is certainly a factor, but it's not the only one. You could buy a separate preamp and power amp for the cost of some integrated amps. I suspect that a major part of the appeal is the desire for simplicity: one box instead of two (or, in the case of monoblock power amps, three) cluttering up the living/listening room, and fewer power cables and interconnects—another cost saving.

When I consider the selection of products for review, I don't normally look for ones that are in the state-of-the-art, cost-no-object category, but rather ones that promise to offer high sound quality at a moderate price. Based on my experience with the PrimaLuna ProLogue Three preamp and ProLogue Seven power amp (see December 2009, Vol.32 No.12), that pretty much describes PrimaLuna's design philosophy. So when I read PrimaLuna's announcement of the ProLogue Premium—"the best integrated amp in its class just got better"—I was keen to check it out.

Description and design
It's said that an amplification component's weight can tell you a lot about its quality. Other things being equal, amplifiers that weigh more tend to sound better. Two factors are involved here: transformers and chassis construction. Higher-capacity transformers weigh more and usually result in better sound. Having a more solid chassis helps suppress resonances, which negatively affect sound.

For a 35Wpc integrated amplifier, the ProLogue Premium is surprisingly heavy: 46.3 lbs. The chassis has a very solid feel, and the fit'n'finish is of a quality that suggests a much more expensive piece of equipment. The tubes are covered by a cage that can be easily lifted off rather than requiring a tool to remove it. The front panel has two knobs: one for volume, the other for source selection. There is no balance control. Source selection has an unusual logic: no matter what source you've been listening to, when the ProLogue Premium is turned off and then turned on again, the source defaults to CD. On the left side panel, near the front, there's a rocker switch for power on/off, and on the right side another rocker switch allows selection of bias optimized for EL34 or KT88 power tubes. Accidentally setting this switch to a tube type not installed does no harm; in fact, the user manual suggests that you experiment with the switch position for personal taste. I explored matching/mismatching the tubes and bias switch positions, and preferred the matched positions.

The rear panel sports five pairs of RCA jacks for inputs, one pair labeled HT (for Home Theater). This is not just unity gain, but actually bypasses the preamp section, and allows volume to be controlled by a surround processor/receiver. The ProLogue Premium normally comes with EL34 output tubes, but can also accommodate KT88s, a set of which was included with the review sample. Except as noted, I did my listening with the EL34s. There are connections for speakers with nominal impedances of 4 and 8 ohms. The ProLogue Premium is available with an internal moving-magnet phono stage ($199), but the review sample didn't include this option.

Image may be NSFW.
Clik here to view.

The ProLogue Premium was designed in Holland by Herman van den Dungen, and, like most audio products these days, it's manufactured in China. According to Kevin Deal of Upscale Audio and PrimaLuna USA, PrimaLuna's North American distributor, the engineering of the Premium began with a circuit design in which nothing was taken for granted, and the components selected are of a quality not normally seen in cost-conscious designs. These include:

• Nichicon and Solen polystyrene and tin-foil capacitors

• Alps motorized potentiometer (said to be 10 times the cost of lesser parts)

• toroidal power transformer, manufactured in-house

• capacitor-plus-choke power supply, designed to eliminate both low-frequency ripples and high-frequency hash

• custom-designed output transformers, manufactured in-house

• heavy-gauge, ventilated chassis with a five-coat finish

• construction by hand and point-to-point wiring

The tubes supplied with the ProLogue Premium are SilverLabel, according to Upscale Audio's classification, and are said to be already "special" compared to standard tubes on the market. Other tubes, including cryogenically treated ones, are available from Upscale.

The ProLogue Premium has a number of specific, trademarked circuits designed to optimize performance and prevent damage to the amplifier in the event of a malfunction:

• Adaptive AutoBias (AAB) monitors and adjusts bias constantly. AAB was present in previous PrimaLuna amplifiers, but the latest version is said to include more extended output-stage protection circuitry, to safeguard the output transformers, resistors, and high-voltage power supply in case a tube fails.

• Bad Tube Indicator (BTI) is what PrimaLuna calls the red LED next to each output tube, though BTI is not quite what its name implies. A lit BTI LED indicates that the tube is drawing more power than it should; however, this condition may be only temporary, and not necessarily indicate a bad tube. If the BTI comes on, you're advised to turn off the ProLogue Premium, wait a few minutes, and turn it on again. If no BTI LEDs light up, you're fine. It's only when the indicator stays on that the cause is likely a bad tube, which must then be replaced. During the review period, there were a few instances of a BTI coming on, but each time, turning the amplifier off and then on again fixed the problem.

• Power Transformer Protection (PTP) is a thermal switch built into the AC transformer. If the internal temperature gets too high, the AC is switched off automatically, coming back on again when the problem is resolved. This never happened during the review period.

• Output Transformer Protection (OTP) protects the output transformers from high-voltage transients. Power in my system is supplied by a PS Audio PerfectWave Power Plant 5, which is designed to protect against transients of this sort; in any case, there were no problems in this area.

• SoftStart is yet another protection circuit, designed to extend the life of sensitive components and reduce the chance of tube failure from thermal shock.

The basic design of the ProLogue Premium follows tried-and-true principles: the output stage runs in Ultralinear mode, and tubes run well below their maximum ratings. Like other products in PrimaLuna's Premium range, the ProLogue uses double 12AU7s in each channel rather than the earlier combination of 12AX7 and 12AU7. This is said to result in much lower distortion in the first stage.

The ProLogue Premium includes a solidly built remote control that also controls the basic functions of PrimaLuna CD players. I found the remote to work extremely well to set the desired volume: a very brief depression of the Up or Down button resulted in a correspondingly minimal change in volume, with no overshoot. Like most remotes, this one has buttons of identical shape and size. Since there are only six buttons, remembering the most frequently used ones, Up/Down/Mute, is not that difficult, but I wish they'd made the Up and Down buttons differently shaped or sized from the others, so that they could be identified by touch.

Setup
Setting up an integrated amp like the ProLogue Premium should be pretty straightforward: connect the source interconnects, speaker cables, and, last, the power cord; make sure the volume control is turned all the way down; turn it on.

Sounds simple enough, but I ran into a problem at the very first stage: plugging in the interconnects from the CD player. The ProLogue Premium's RCA jacks have plastic covers on them, and I had a difficult time trying to remove the ones over the CD input pair. (The user's manual makes no reference to these protective covers; they may have been a late addition.) Pulling on them didn't seem to work, so I started twisting one of them—and then I felt it give, and the entire input jack started turning! Not good—I'd inadvertently broken the internal connection to the CD input jacks. To repair the damage, with Kevin Deal's approval, I engaged the services of a highly experienced audio technician, Roger Sherman, who came to my home and was able to resolder the connection without difficulty. I eventually found out that there's a trick to removing the protective caps. There's a part that provides for a grip (it was facing the bottom, so I didn't see it); once you take hold of that, you can easily pull the cover off. This problem solved—and both channels of the CD input working properly—I was ready to do some listening.

According to the folks who view measured electrical performance as the only criterion for the assessment of an amplifier, once an amplifier's internal components have reached operating temperature, the amplifier is performing as well as it's ever going to perform: "break-in" and "warm-up" (beyond a minute or so) are simply myths.

But most audiophiles and manufacturers of audiophile equipment know that break-in and warm-up phenomena are real, and that amplifiers vary in these characteristics. In discussing the effect of replacing tubes, PrimaLuna notes that "break-in yields improvements." In the case of the ProLogue Premium, I found that over a period of several weeks, with the amplifier on 9–10 hours a day and playing music 3–4 hours a day (I leave the CD player on all the time), there was an improvement: the music sounded more detailed, and somehow more "relaxed" and free-flowing. It also seemed to me that, playing music each day for several hours, there was a similar sonic improvement, suggesting a warm-up effect beyond the few minutes it takes the tubes to reach their normal operating temperature. I noted these effects while using the Avantgarde Uno Nano speakers; the warm-up effect was also apparent with the GoldenEar Triton Twos. I decided to evaluate it in a more systematic way.

But some products demand a complete revision of a system's architecture. Such was the case with Devialet's D-Premier ($15,995). Not only is this French product an integrated amplifier, with phono and line analog inputs; it has digital inputs and an internal D/A section. And since v.5.5 of its operating system, the D-Premier can also act as a high-resolution WiFi audio streamer, working with Devialet's Asynchronous Intelligent Route (AIR) client for Macs and PCs. At a stroke, the D-Premier replaces streaming program, USB or other computer audio interface, D/A processor, preamplifier, power amplifier, and many cables. And all is contained in a beautifully finished aluminum case about the size of a small pizza box and just over an inch thick.

I first saw and heard the Devialet D-Premier at the 2010 Consumer Electronics Show, but it was not until a year later that Audio Plus Services announced that it would distribute the D-Premier in North America. I received a first review sample in 2011, then a second, to use with the first as a pair of monoblocks, in early summer 2012—but first I needed to clear my decks of more conventional products that I was reviewing. I am now kicking myself for having waited so long.

Technology
Devialet SAS is a French company, founded in 2007 by Pierre-Emmanuel Calmel and Mathias Moronvalle, colleagues at Nortel France's R&D Lab, to develop a new type of amplifier developed by Calmel. Called ADH®, for Analog Digital Hybrid, this patented topology connects a small, high-voltage, but low-power class-A amplifier directly to the speaker, with then a parallel class-D stage providing the necessary current. This is reminiscent of the innovative "current-dumping" circuit developed by Quad in the mid-1970s, though the Quad circuit used a class-AB current amplifier. However, the AHD circuit differs significantly in detail from Quad's, and is considerably more complex. Extraordinarily, there only two resistors and two capacitors in the analog signal path!

I discussed the D-Premier's topology with M. Calmel at the 2011 CES. The analog input signals are converted to digital with an A/D converter, a Texas Instruments PCM4220, running at 96 or 192kHz—the former is the default—before being applied to the volume control, which operates in the digital domain and is implemented in a 32-bit floating-point DSP chip, along with a soft-clipping function and crossover filters when required. All signals are then converted back to analog by two Burr-Brown PCM1792 chips—a high-quality, 24-bit, two-channel, current-output device operating at up to 192kHz. Just half of the DAC is used for each channel, and the current output of the DAC is converted to voltage with a resistor and fed directly to the class-A amplifier—the analog signal path from the DAC output to the loudspeaker terminals is only 2" long. In effect, the DAC swings the high voltage required to drive the speaker output, and the class-A amplifier therefore works at unity gain, as a voltage follower, so that its performance can be maximally linear at high frequencies.

Image may be NSFW.
Clik here to view.

To provide the current to drive the loudspeaker, a four-phase, multilevel digital amplifier—four switching stages, staggered in time—is added in parallel to the class-A amplifier. It is slaved to the class-A amp much as in a car's power steering, where the driver turns the steering wheel to indicate how much he wants the wheels to turn, and a servo-controlled hydraulic system actually turns the wheel.

Conventional class-D amplifiers suffer from high levels of ultrasonic switching noise riding on their outputs, which mandate use of a hefty low-pass filter between the output stage and the speaker terminals. In the D-Premier, there is no LC filter on the class-D amplifier's output; instead, the analog amplifier provides a very wide-bandwidth correction signal that cancels the ultrasonic switching noise that would otherwise be present.

The power supply is a 600W switch-mode type offering 2100W peak and incorporating full power-factor correction. Because of the high switching frequency, the planar transformer can be tiny. There is much more to the D-Premier's innovative and elegant circuit that I don't have room to discuss here; I refer you to a white paper that can be downloaded here. But the entire package offered by the D-Premier appealed to my sense of purity—it is no bigger than it need be to do what is intended.

Setup
I had to slide off the section of the D-Premier's top plate that covers the rear panel in order to be able to use my preferred XLO Reference 3 AC cable, the plug of which would have been too big to reach the recessed IEC mains jack. The D-Premier offers extraordinary flexibility in how its inputs can be arranged—see the diagram of its rear panel. Using the Configurator app, downloadable from the Devialet website, the user sets up the amplifier as he needs and burns the configuration as a text file to an SD card. Inserting this card in the rear-panel slot and turning on the amplifier updates its internal state. I used the factory default configuration, which offers two pairs of analog inputs (one of them phono, to be tested in a Follow-Up review) and five digital inputs: two TosLink, two S/PDIF on RCAs, and one AES/EBU on an XLR jack. There is an HDMI port, currently unused, and the D-Premier is WiFi capable.

Image may be NSFW.
Clik here to view.

The only control on the amplifier itself is a discreet On/Off/Sleep button at the center of the front panel. When the amplifier is on, a gentle amber circle is projected onto the surface beneath the front of D-Premier. All other controls are carried on the remote control. A large rotary knob adjusts the volume. A single button above the volume control duplicates the On/Off/Sleep button; three other buttons control Input Select (consecutive pushes cycle through the inputs, each starting with the volume control set to "–40.0dB"), Bass high-pass filter On/Off (when configured for use with a subwoofer), and Polarity Inversion. These buttons can also be used to control channel balance and tone-control selection, when the amplifier is appropriately configured. The small, circular, color display on the amplifier's top panel indicates the input in use, the volume level in dB, and the sample rate for digital inputs. If no datastream is present, the digital input's name illuminates in red; it turns black if valid digital data are detected. The D-Premier goes to sleep if no music has been played for 20 minutes or so.

Image may be NSFW.
Clik here to view.To use the D-Premier's WiFi connectivity, Audio Plus had supplied me with an Apple Airport Express programmed to set up a network called "DevialetAudio." When turned on, the D-Premier looks for a network with this name and logs on. To play music over this network, you install the Devialet AIR app on your PC or Mac. This then handles the output of music files selected in iTunes, selecting the correct sample rate and transmitting the data over the network to the D-Premier. This runs automatically, and while it doesn't have to be open during use, as the screenshot shows, when open it displays the name of the file playing, its format, sample rate, and bit depth, and the playback time, as well as cover art and various network diagnostics. AIR is currently limited to 24-bit/96kHz, but a free upgrade to handle 24/192 files is promised.

When the D-Premier is connected to the DevialetAudio WiFi network, an iPhone/iPad app duplicates the remote control's functionality and the amplifier display, adding the words "My D-Premier" above the volume setting. Touching the input name brings up an Input Select menu, and again, selecting a different input reduces the volume to "–40.0dB." Touching the volume setting mutes the D-Premier; touching the Mute symbol unmutes the amplifier.

I can't imagine what went wrong.

Description
The EHF-100 integrated amplifier ($6350) is one of two products made by Rogers High Fidelity, a new company in rural Warwick, New York. (When I first saw a Rogers amp, at Stereo Exchange in New York City, I assumed the brand name had something to do with the late, lamented Rogers loudspeaker company of Great Britain, but that turned out not to be the case.) Roger Gibboni, who studied engineering at Drexel University and who has over 20 years' experience designing RF circuitry for the military and NASA, founded Rogers High Fidelity in 2009 with an eye to offering very high-quality, US-made tube amplifiers at less-than-extortionate prices.

In contrast with some recent integrated amplifiers, in which power amplifiers are paired with passive front ends, the Rogers EHF-100 has sufficiently high gain to be considered an integrated of the more traditional sort. Here, the majority of voltage gain is provided in the first stage: an EF86 miniature pentode tube. From there, the signal goes to a 12AX7 dual-triode tube, which is used in common-cathode mode to split signal phase for the push-pull output section that follows.

Power tubes in the Rogers amp are KT88s, operated in auto-bias mode: The cathode of each is held above ground by a chunky Dale resistor, ensuring that the KT88's signal grid maintains a steady negative charge relative to the cathode. (I measured approximately 44V across each cathode resistor; consequently, although there was 514V between plate and ground, each KT88 sees a total B+ of only 470V.) The output section of the EHF-100 runs in Ultralinear mode, with plates and screen grids connected to the tapped primaries of very large Hammond output transformers. (I don't know what percentage of primary windings are given over to the screen grid, but other Hammond Ultralinear transformers of my experience are set for 40%.) In addition to its Ultralinear operation, the Rogers EHF-100 also uses a small amount of global feedback. Gibboni says that the amp is fully class-A.

Image may be NSFW.
Clik here to view.

The EHF-100's build quality is excellent—impressively so, given its relative affordability in the context of other US-made tube amplifiers. Every component is hand-wired, most by means of silver-plated hookup wire and thick, clean tag boards. Threaded parts are held in place with Loctite thread sealant and Glyptal red enamel; long runs of cable are tied in place with something very like dental floss (the knots in which also get the Glyptal treatment); and large capacitors are incapacitated with generous dabs of silicone sealant. Solder joins are as close to perfect as I've seen. Parts choices include the aforementioned Hammond transformers (and power-supply choke), plus metal-film signal resistors, polypropylene and silver-mica signal capacitors, Micalex tube sockets, and rhodium-plated copper binding posts from Furutech.

The Rogers EHF-100 offers four pairs of line-level inputs, with one pair of jacks on the front panel—just to the right of the four-position input-selector switch—and the other three on the back panel. A phono section is not provided, although Gibboni says he's considering making one as a standalone component. As with other notable amplifiers, powering up the EHF-100 is done by flipping, in order, two switches: one, Power, to send the requisite voltage to all of the tube filaments; and the other, Operate, to apply the rail voltages. Apart from those and the selector switch, the only other control is a stereo volume pot. (A balance control is not provided.) At the center of the front panel is a power-output meter, calibrated in peak watts from 0 to 100.

Image may be NSFW.
Clik here to view.

The very substantial aluminum-alloy case is finished in what Rogers refers to as an automotive-grade finish, in glossy black (other colors can be had for an additional charge). I love the EHF-100's simple, utilitarian shape; it calls to mind such disparate products as the beautiful Marantz 8B and Shindo Montrachet amplifiers. My only complaints have to do with what seems the amp's excessive size (one thinks of Mary Wollstonecraft Shelley's line "As the minuteness of the parts formed a great hindrance . . . I resolved, contrary to my first intention, to make the being of a gigantic stature") and its disproportionately large, splashy logo and other graphic touches. But that's just me.

Installation and setup
Because my house is only three hours away from the Rogers High Fidelity factory, Roger Gibboni delivered my review sample by automobile, rather than shipping it via brown truck. That said, there was nothing extraordinary about unpacking, installing, or operating the EHF-100—although I was impressed by the exceptional quality of its carton and packing materials. The operating manual was also quite good, and my review sample came packed with a spiral-bound booklet titled "Amplifier Profile": an illustrated collection of reports on tests to which this particular amp had been subjected, including THD vs level, group delay vs frequency, squarewave performance at 1 and 10kHz, and others. I'm told that every Rogers amplifier sold includes such a report.

That ideal has been thwarted by the concept that, in order to hear what needs hearing, people who really love music must buy separate line stages, phono preamps, transformers, transports, D/A converters, master clocks, power amps, booster amps, and power supplies. An opposing notion offers encouragement for the thrifty and the sane: All that and more can be had in a single component that is still designed and manufactured to perfectionist standards, and the AVM Evolution C9 ($5750) among them.

Description
Audio Video Manufaktur, a German firm that has supplied electronics to Europe and Asia since 1986, entered the US market in 2008, apparently with an emphasis on the comparatively (footnote 1) affordable side of high-end audio. While attending the redundantly named New York Audio and AV Show of 2012, I was impressed by the two CD receivers in AVM's line, the Evolution C9 being the more expensive. (The AVM Inspiration C8, which offers 100Wpc into 8 ohms plus an array of functions nearly identical to those of the C9, sells for $4200, and is reviewed by John Marks in this issue's "The Fifth Element" column.)

In addition to its 180Wpc integrated amplifier, the Evolution C9 provides a selection of onboard source options, the most conspicuous of which is a CD player, whose transport is said to be mechanically isolated from the rest of the unit. According to AVM, the C9's D/A converter automatically upsamples "Red Book" CDs and CD-Rs to 24-bit/192kHz: a kindness that's done for all other digital signals as well. (AVM's converter chip of choice is the Wolfson WM 8741.) Alongside its CD player, the AVM C9 also incorporates an FM tuner, with a European RDS decoder for broadcast text data.

Image may be NSFW.
Clik here to view.The aforementioned DAC can also be addressed with signals from external sources, including both S/PDIF digital (a choice of RCA and optical jacks) and computer-music files, by means of the C9's USB-B jack, the latter involving the use of a 16-bit Burr-Brown PCM 2704 chip as a receiver. Another input-source selection, labeled Player, is actually a line-level audio/video input, with a quartet of RCA jacks and a USB-A jack for supplying DC to a portable music player. Although the C9's manual is sketchy on the point, this function appears intended for use with an iPod equipped with Apple's Composite AV Cable ($39), a sample of which I did not have on hand for this review.

In addition to the above, the AVM C9 provides three other line-level inputs, a moving-magnet phono input, a fixed-level line output, and a processor loop. The sensitivities of all those analog inputs—including the aforementioned Player input—can be adjusted, by means of the C9's Menu function, to match the output voltage of each device in use.

In common with other all-in-one products of recent vintage—the Micromega AS-400 comes to mind—the amplifier section of the AVM C9 operates in class-D. AVM is also au courant in having adopted switch-mode power-supply technology for all of their electronics: a total of three power supplies in the case of the C9, two of them serving as dual-mono supplies for the amplifier section alone.

Because the C9 resisted my mild attempts to disassemble it, I'm unable to comment on the quality of its parts or construction, but I found the C9's brushed-aluminum casework agreeable overall, with impressively serene, unfussy styling and well-designed controls: a large source-selection knob, an equally large and nicely weighted volume control, an endearingly easy-to-read display screen, and a horizontal row of five soft-touch buttons, the precise functions of which vary with either the user's choice of menu screen or source. Insofar as I could tell, the build quality was fine.

Installation and setup
I used the AVM C9 in two different rooms, with three different speakers: a loaner pair of Wilson Audio Sophia Series 2s, and my own pairs of Quad ESLs and Audio Note AN-E SPe/HEs—something to please everybody, as Sally Henny Penny would say. The C9 has two complete sets of loudspeaker outputs, although I never used more than one set at a time.

During its time with the Audio Notes in my 12' by 19' listening room, the AVM C9 sat on the middle shelf of my borrowed Box Furniture rack, where its temperature ranged between cool and mildly warm to the touch—the latter only during prolonged listening sessions. I used only the stock AC cable, and resisted the temptation to use any manner of accessories to isolate the C9 from an acoustically hostile environment. Speaker connections were made with my stranded-copper Auditorium 23 cables, while interconnects were my usual mix of Shindo silver and Audio Note AN-Vx.

The C9's software controls, the conventions of which will be familiar to anyone who understands the distinction between merely pushing a Menu button and holding it in for three seconds, offered some welcome choices. In addition to the ability to tailor analog-input gain as described above, the C9 allowed me to cut and boost treble and bass; to apply a loudness curve for low-level listening; to enable or disable either pair of speakers; to choose between distant and local broadcast signals; to choose between stereo and mono broadcast playback (the latter, as always, to tidy up weak signals); and, best of all, to effect small balance adjustments between the left and right channels.

Image may be NSFW.
Clik here to view.

Some specifics: The RadioShack FM antenna I installed years ago in my attic—where it has stood, half forgotten, since the last time I reviewed such a product—gave grist to the tuner's mill, while the programs Decibel and iTunes drove the AVM's USB D/A converter by means of an Apple iMac running Apple's OS X 10.7.4. (In the Sound submenu of my iMac's System Preferences menu, the AVM showed up as "USB Audio DAC.") Lack of a standalone CD transport argued against my trying the AVM's S/PDIF inputs. And although its phono-stage specifications suggest that the C9 can be directly driven by higher-output moving-coil cartridges, I nonetheless chose to use a step-up transformer between that stage and every cartridge I tried—my preferred approach in any case, owing to the superior sense of musical touch and impact it imparts to phono playback.

Listening
I began by using my Apple iMac to drive the Evolution C9's USB input (which is also the easiest way I know to break in a brand-new amplifier). During the first day of listening I was impressed by how obvious the C9 made the distinctions between iTunes and Decibel, the latter's treble openness and clarity being preferable to the former's dust and haze. Yet during that first day of listening it was also apparent that that clarity owed its existence to a somewhat bright timbral balance, and a treble range in which note attacks and textural distinctions were allowed greater-than-usual emphasis.

My idea of natural sound calls for a more reticent top end than that. Consequently I expected—and hoped—that the C9's high-frequency performance would become a little less eager, as often happens as new playback gear settles in. That happened here, but only to a very slight extent.

If that's the worst news about the C9—and it pretty much is—the best news would surely be its facility with the difficult-to-drive Quad ESL, the impedance of which drops a whopping 31 ohms between 80Hz and 18kHz. Yet when driven by the AVM amplifier, my Quads sounded as nicely balanced as they ever have. With the right gear, ESLs can also sound surprisingly and pleasantly tight, with excellent musical timing and freedom from timing distortions—and so it was here, with plucked cellos and electric bass guitars alike sounding tuneful and snappy. When listened to from an off-axis and reasonably farfield (ie, more than 10' away) seat, the C9 also afforded the Quads a more natural-sounding treble range than I heard through the Wilsons or Audio Notes—although, in the nearfield setting that many Quad owners prefer, the AVM's overly strong top end endured.

The C9's ability to express the timbral colors and natural sonic textures of recorded music, while not as good as that of my Shindo and Fi electronics (unsurprisingly, given the AVM's comparatively low cost, footnote 2), was acceptably good, especially through the Quad and Audio Note speakers. A few years ago, a friend gave me a CD copy of a digital recording made in India, of an improvisation performed on sarod, harmonium, and tabla: a timbrally rich recording, as you can imagine. Rather than sounding tonally thin or bleached out—as I might have feared, given the amp's brightness—the recording was pleasantly colorful through the C9, if not quite as fleshy and bloody as I like.

A few source-specific observations: The AVM's phono section performed admirably, without adding hum or obvious noise to the signal. (There's a phono-ground lug on the C9's rear panel, but I never had cause to use it; grounding the tonearm in use to the step-up transformer in use was sufficient in every instance.) Without an antenna, its FM tuner couldn't pick up any stations—not even the local religious station broke through—but with the antenna in place, the AVM received three stations clearly when set to Local, with the addition of a few iffy signals when set to Distant, the latter being cleaned up nicely by the switch from stereo to mono. (Remember, I live in a rural area in central New York State.) And the C9's CD player performed reliably, even with two physically worn discs that another player of mine now rejects out of hand.

The AVM's onboard CD player was, as the days played out, the source on which I most often relied—at its best it sounded pleasantly forward, with very strong center-image presence. Still, the observation that dominated my notes was the C9's overabundance of high-frequency detail. In "Ellis County," from Buddy and Julie Miller's fine Written in Chalk (CD, New West NW6158), the usually well-textured fiddle had too much texture, and drawn-out vowel sounds in Buddy's lead vocals had a little too much wheeze around the edges. In "The Last Living Rose," from PJ Harvey's Let England Shake (CD, Vagrant VR651), the tambourine sounded harsh and bright—ditto Lee Feldman's tambourine in "Halo," from his great new Album No.4 (CD, Bonafide UM-130-2). Those aspects of its sonic performance prevented me from ever really warming to the AVM in my system.

Conclusions
With a total of 10 source selections—two of which are self-contained—the AVM Evolution C9 lacks neither flexibility nor ostensible value. But while the C9 performed well enough with my Quad ESLs, and while I can think of at least four or five classic Brit-fi loudspeakers that might also suit it, owners of speakers with tipped-up or very well-extended treble ranges would probably do well to consider something else. I remain enthused about this product category—in which solid-state amplification prevails, understandably enough—but the bargain hunter who shares my taste in sound should approach the Evolution C9 with greater-than-usual care.

Footnote 2: Ibid.!

A new loudspeaker or an updated classic?
The Anima is a two-way loudspeaker from Canalis Audio, a new enterprise of longtime importer Immedia, of Berkeley, California. Canalis is thereby related to Spiral Groove, and Canalis speakers bear the Spiral Groove logo on their terminal plates. Spiral Groove, founded in 2005, makes turntables; their SG2 ($15,000) was favorably reviewed by Brian Damkroger in the June 2010 issue. Canalis makes at present four models of loudspeakers, all designed in collaboration with noted engineer Joachim Gerhard, formerly of Germany's Audio Physic. All Spiral Groove and Canalis products are made in the US.

For the time being, Spiral Groove/Canalis products come only from the electromechanical end of high-performance audio, not the purely electronic (let alone digital) end. The focus on addressing electromechanical rather than electronic challenges appears to be a consequence of Immedia founder Allen Perkins's rubric of "Balanced Force Design," which posits as the overarching design goal the optimal integration of "materials, performance, function, manufacturability, and aesthetics." This focus is, as well, a consequence of Perkins's interest in vibration and resonance control—not surprising, from a turntable designer.

The Canalis line comprises the Anima, a stand-mounted, rear-ported two-way ($3250/pair); the Anima CS, a tricked-out Anima with a ¾"-thick bottom plate of solid stainless steel and upgraded crossover parts ($6000/pair); the Cambria, a floorstanding two-way ($5000/pair); and the Allegra 2.0, a modular loudspeaker with a two-way, sealed-box upper cabinet crossed over to a bass module containing two 8" aluminum-cone woofers ($17,500/pair).

Image may be NSFW.
Clik here to view.

The Anima's driver complement consists of a ¾" hybrid tweeter that combines a textile surround with a metal dome and is used in all Canalis models, and a 5" anodized aluminum-cone woofer. The Anima measures 16" H by 10" W by 10.5" D and weighs 14 lbs. Its front panel tilts back at an angle of about 9°. High-quality Eichmann terminals permit single-wiring only.

Canalis states that the Anima has an impedance of 8 ohms, a sensitivity of 86dB, a crossover frequency of 1900Hz, and a frequency response of 44Hz–33kHz, –3dB. They also claim that the design of the crossover incorporates a proprietary "Joachim Gerhard DC-Module" that minimizes the baffle-step effect.

The Animas come very securely packed in a sturdy double carton, along with a very complete guide to room placement. The review pair was very handsomely finished (including the rear panel) in low-gloss polyurethane in a shade Canalis calls Licorice, but which I would call Jacobean—think of the chunky, heavy, very dark brown wooden furniture from the reign of James I. Other finishes are natural, which Canalis calls Honey, and a warm, reddish, medium-brown that Canalis calls Cinnamon.

The Anima is not provided with a grille, so there are no grille-mounting attachments on its front panel. The front of the bottom plate is slightly recessed, and carries a metal strip embossed with the brand name, as well as threaded metal inserts to secure the speaker to its dedicated stand ($1500/pair), or allow the use of Spiral Groove's proprietary resonance-control pods, the Strange Attractors ($800/set of eight). These go between stand and floor, or between the speaker and a shelf or third-party stand. The Attractors' metal upper hemisphere is anodized the same bright red as the blind nut that secures the bolt that pierces the three crossed legs of the stand.

Although I consider the Canalis Anima to be a new loudspeaker, reasonable minds might differ on that point. Readers with long memories may recall Wes Phillips's enthusiastic July 2007 review of a loudspeaker, also called the Anima, from a company called Sonics by Joachim Gerhard, and also imported and distributed by Immedia. Earlier, while at Audio Physic, Gerhard had designed the very-well-received Step and Step SLE loudspeakers. The Sonics by Joachim Gerhard Anima was a refinement, at that time, of his earlier design.

I assume that the Step (which remains, I assume in further-evolved form, in production from Audio Physic) was named to highlight what its makers perceived to be an unusually successful implementation of baffle-step compensation. Baffle step refers to the rise in a loudspeaker's frequency response that comes about because, above some frequency primarily related to the front baffle's width, its output is directed mostly forward. Whereas, below that frequency, heading down to the deep bass, soundwaves are increasingly omnidirectional, and less energy, therefore, is directed at the listening position. Although a baffle-step graph, when read from left to right, might suggest a rising treble, I believe the acoustical phenomenon is actually one of weakening perceived bass.

Baffle-step compensation, of course, is only one aspect of the "Squeeze it here, it bulges there" art of loudspeaker design. Time alignment of the outputs of woofer and tweeter is an issue, and then there's always the Allison Effect, aka the floor-bounce effect: a frequency-specific bass cancellation related to the distance from the center of the bass driver to the floor. And other factors.

Image may be NSFW.
Clik here to view.

Regardless how they came up with the Step name, I think it's fair to say that, in the early 1990s, the Audio Physic Step's crisp, clear, dynamic sound and remarkable imaging represented a new paradigm in minimonitor sound, and one that was warmly received by many reviewers, including Stereophile's Jack English in his October 1994 review.

The Sonics Anima that Wes reviewed in 2007 had front and rear panels of marine-grade plywood; its other panels were of high-density fiberboard (HDF). It had a grille, which Wes found made its optimal sonic contribution when removed. The Sonics Anima's tweeter and woofer, I was told by Allen Perkins, were essentially the same as are now found in the Canalis Anima. Indeed, Perkins told me that during the development work on the Canalis Anima, he took pains to ensure that the cabinet's internal volume remained essentially identical to the Sonics version. As it turned out, the new speaker could even be shipped in the same packing materials as the older.

Given all that, why would I consider the Canalis Anima to be a "new" loudspeaker? I view the Canalis Anima the way I view Harbeth's P3ESR, vis-à-vis the BBC LS3/5a from which it descends. In both cases, the footprints are the same, and the cabinets can function as drop-in replacements (Harbeth's P3ESR can drop into the soffit space left by a defunct LS3/5a). However, design changes in each case have resulted in performance that I believe constitutes a clean break with the predecessor's. In the case of the P3ESR, the cabinet remains the same, but the drivers and crossover changed. With Canalis's Anima, the cabinet material (but not its size or shape) changed radically, while the drivers remained the same (though there are revisions to the crossover). In both cases, I think, evolution turned the corner into revolution.

What is revolutionary about the Canalis Anima? In a word, bamboo. Allen Perkins told me that he'd begun thinking about and experimenting with bamboo plywood as a speaker-cabinet material 10 or so years ago, and that he still has stain and finish experiments from that time. The plywood of Canalis's speaker enclosures and the Anima's stand is ¾" thick, and distinguished by the fact that its core consists of substantial, vertically oriented fillets of bamboo rather than the usual multiple thin horizontal layers of hardwood, as in the Baltic-birch plywood often found in high-quality speaker enclosures. In most Shahinian speakers, for example, the enclosure's edges are beveled to reveal the layers of plies (and to make it clear that the cabinet is not made of MDF). The surface layers of Canalis's bamboo plywood are laminated of thinner fillets, oriented cross-grain to the core.

Why bamboo plywood? The short answer is: rigidity, plus self-damping. In this regard, it's important that Canalis chose bamboo plywood for acoustical performance reasons, which is different from choosing a bamboo veneer for aesthetic or ecological reasons. After a few weeks of living with Canalis's Anima, I found myself thinking of bamboo plywood as the poor man's carbon fiber. Indeed, the Anima's sonic coherence called to mind fond memories of various incarnations of Wilson Benesch's ACT loudspeakers. Gerhard bases his work on mathematical models by Don Keele, specifically the QB3 alignment, an overdamped woofer alignment that Gerhard claims behaves more like a sealed box, because of its longer port. Gerhard also believes that it's better to concentrate on getting the overtones of bass notes right and let the ear/brain system fill in the missing fundamental, than to make the trade-offs involved in including a large woofer in a two-way design. He chose the crossover frequency of 1900Hz to stay away from the range where the output of his 5" woofer becomes directional.

Farad, however, was effectively the sole driving creative force eight or nine years ago, when I used to regularly visit him at his London apartment and witness stages in the design of a loudspeaker that, in retrospect, would put Mission Electronics on the high-end map. As well as drinking large amounts of his liquor, night after night I would witness Farad putting record after record on his Linn, listening to what seemed to be innumerable prototypes of what became the Mission 770, trying to match the midrange accuracy of the classic "BBC-sound" Spendor BC1 but marrying it to a less loose bass region, more suited to the special requirements of modern LP replay.

And to a large extent he succeeded. I can remember almost painfully exquisite reproduction of Jackson Browne's live Running on Empty album, Emmylou Harris just taking my breath away with Quarter Moon in a Ten-Cent Town, and Dire Straits'Sultans of Swing causing us to boogie until we were disturbed by the early-morning sounds of London's pigeons taking their first cough. The 770 was one of the first speakers to use a polypropylene-cone bass/midrange unit; the combination of an upper-midrange transparency rare at the time of its launch, and low frequencies that, while never quite as tight as those of the Linn Isobarik, nevertheless were "fast" and played tunes effectively, caused the 770 to be the loudspeaker of choice for many UK audiophiles in the early '80s.

Farad and I somewhat lost contact over the following years, and perhaps inevitably, I found myself losing sympathy with the sound of Mission's loudspeakers. While always detailed and fast, and offering excellent value for money, they increasingly featured, in my opinion, a somewhat forward midrange which didn't fit with my awakening tastes for subtlety and restraint in high-end sound. If you take perfection in sound reproduction to lie at the top of a broad-skirted mountain, then you could say that Farad's and my paths toward perfection diverged around opposite sides of the mountain. The destination may be the same, but the incidental scenery is totally different.

Time passed, however, and in 1986 I found myself being drawn again toward the sounds of Farad's brainchildren. Aided by the set-up skills of one of the partners in Mission's Canadian subsidiary, Armi Leonetti, Mission had always obtained excellent sound in their CES exhibits. While the top models in Mission's new generation of loudspeakers, the 770 Freedom and 780 Argonaut, didn't look or sound anything like their distinguished ancestor, the original 770, they sounded good on their own terms. Low frequencies had excellent weight, without the almost universal propensity for box loudspeakers to lose definition in the upper bass, while sensitivity was high, so that the speakers would produce high sound-pressure levels with the relatively modest output power of the Mission amplification.

Intrigued, I spent a weekend at Mission's Canada HQ in Toronto, listening to an Argonaut- and Cyrus-based system put together by Armi. Impressed with what I had heard, I requested a complete Mission system for review, resulting in the words you are now reading. Each piece of equipment—the PCM 7000 CD player, Cyrus Two integrated amplifier, and 780 Argonaut loudspeaker—was first auditioned in the context of my own usual system, based on a Linn Troika/Ittok/Sondek feeding an Audio Research SP10/Krell KSA-50 combination which in turn drives Celestion SL600 loudspeakers on spiked Foundation stands. Interconnects and speaker cables are by Monster; beer by Corona; Scotch by The Macallan. Following that experience, the three components were auditioned as a complete Mission system.

In digital audio, time waits for no one—and so it goes here. The iDecco, which sold for $999, is no longer in the company's lineup, and while Peachtree continues to put iPod docks in various of their amps and preamps, today they're betting on a simpler yet more highly evolved addition to their stable: the decco65 ($999), in which a 65Wpc amplifier and a hybrid tube/solid-state preamp share space with a 24-bit/192kHz D/A converter.

Description
Much as they've succeeded in setting the right categories and price points for their products, so has Peachtree Audio found a winning formula for the decco65's industrial design: an attractive wood wrap—too curvy to be called a box—surrounding a chassis that's just big enough to contain its working parts and a bit of cooling space. The wrap, which comes in a choice of high-gloss black paint or, for $100 more, a veneer of cherry or rosewood, fits snugly around an elegant satin-finish panel, the latter including a soft-touch standby switch, a row of smaller buttons for source selection, a large volume knob, a headphone jack, and a clear plastic window for the decco65's single vacuum tube: a 6N1P dual-triode, used not for voltage gain but as an interstage preamplifier buffer. A button on the remote control allows the user to switch between tubed and solid-state buffering, the former indicated by a blue light at the center of the tube's socket; thus the window offers visual confirmation of an audible function that some may consider subtle.

Inside, much of the decco65's size and weight are accounted for by a large toroidal power transformer, with the remaining power-supply components located on the rear half of the main preamplifier printed circuit board. A class-D power amplifier, built around a Texas Instruments (née Toccata) Equibit chipset, occupies a 5.5" by 3.2" board of its own, approximately 30% of which is taken up by a single chunky heatsink. The amp board sits near the center of the decco65 chassis, itself formed from a single piece of thin-gauge steel and finished in textured black paint.

Alongside three other Peachtree products introduced at the same time—including the more powerful nova125 that Sam Tellig reviewed in January—the decco65 is the company's first product to offer a USB digital input capable of handling 24-bit/192kHz data, and operating in the preferred asynchronous mode. In the decco65 and nova125, the iDecco's ESS Sabre 9006 DAC chip has been replaced by ESS's 9023 chip. According to Peachtree Audio's David Solomon, the asynchronous USB datastream is clocked right at the DAC.

Installation and setup
During its time in my home, the Peachtree decco65 replaced the electronics in two different systems: For the first few weeks it sat atop a nothing-special bookshelf in my living room and drove my restored Quad ESL loudspeakers; for the serious listening that was the basis of the observations below, the decco65 drove my Audio Note AN-E SPe/HE loudspeakers and the review samples of DeVore Fidelity's Orangutan O/96 speakers, both in my 12' by 19' listening room. In the latter setting, the Peachtree occupied two different spaces: When playing CDs and LPs—the latter by means of a borrowed Leben RS 30EQ phono preamplifier—the decco65 sat on the middle shelf of a Box Furniture rack; when playing music files streamed through its USB input, it sat on a short pinewood table near my Apple iMac, to keep USB cable length to no more than 1m. Hookup was easy and logical; my only serious criticism of the decco65 from that perspective is its lack of more than one pair of line-level inputs. The hobbyist who wishes to use a phono preamplifier and a CD player—not to mention a tuner or a tape deck—will have to look elsewhere.

Image may be NSFW.
Clik here to view.

The decco65 became only mildly warm to the touch during use, no doubt aided in that regard by the metal cooling vents built into the top and bottom surfaces of its wooden wrap. Unfussy in its cabling requirements, it seemed perfectly happy with the fabric-sheathed, copper-conductor speaker cables from Auditorium 23 that have become my favorites over the past five-plus years. The decco65 didn't appear to invert absolute signal polarity when driven with line-level sources—my Sony SCD-777ES SACD/CD player, my AudioQuest DragonFly USB D/A converter, and the aforementioned Leben phono preamp—but I was slightly less sure when listening to its S/PDIF and USB digital inputs. I found myself very slightly preferring the sound with the polarity inverted at both pairs of output terminals, although those differences seemed smaller than usual, and did not, for whatever reason, call to mind to the sonic indicators I usually associate with correct and incorrect signal polarity. Go figure.