With Audio Technology Switzerland's Nagra HD DAC & MPS due back in Lausanne after their review, I volunteered to deliver the loaners in person. It'd give me opportunity to hear Nagra's resident system with their DAC and—hello!—the maturing prototype of the forthcoming HD AMP 1000*, a very high-power transistor unit loosely based on the existing MSA but scaled up for about quadruple the power. Matthieu Latour had let on that R&D had made some rather meaningful progress of late to get the design ready for some informal pre-sampling. Final cosmetics weren't decided yet. This remained a prototypical kludge for the moment. Chances of me photographing anything in flagrante were probably slim to none. Still, I was up for a fishing expedition. As it turned out, I would bag a whale but not exactly as expected.

* The actual name wasn't finalized yet but merely a project designation. Depending on which of the circuit's various power ratings it will reflect, it could become the AMP 500 (for 4 ohms) or the AMP 1000 (for 2 ohms).

To set the correct mise en scène, we must back up and consider the antidote to hifi's customary law of diminishing returns. It says that beyond a certain point, higher expenditures net proportionally smaller and smaller returns. A €10'000 DAC for example may not be twice as good as a €5'000 one. The gap between it and a €20'000 sample could be smaller still. As far as it goes, this nearly universal observation is perfectly correct. But independent of it there also operates an opposing force which we might cheekily call the law of compound gains. That says that the weakest link in the chain will hold back everything else. This bottleneck prevents everything upstream from getting all its goods past the blockade; and in turn means that everything downstream receives a lesser quota of bricks to consequently build a shorter bridge that won't reach as far or high.

The ambitious audiophile's perennial dilemma is to identify which component at any given state this weak link is. At a certain point it could be the powerline conditioner or a cable's particular length. It should be apparent that success of tracking the culprit—and we've not discussed the room which tends to be the biggest perfection preventer—is directly proportionate to the breadth of one's own access to gear. We don't know what we don't know until we do. This process is never a wholesale but always only partial revelation. The more complex and multi-stage our system is, the more variables must be kept in check; and the more serpentine and numerous the interactions become which push and pull on the final result. It's far from straightforward but one complex messy matrix of co-dependencies wherein everything matters and everything affects everything else.

Having visited Nagra before, in this and their older facility, their team trusted me enough this time not to straighten out their reference room for a formal visit but to leave it in its proper 'men at work' state. Because they'd just determined that a shorter run of the same cable that had always connected the source stack to the amp/s between the speakers sounded considerably better, said source of Nagra CD and HD DAC sat unceremoniously on the floor. They had to reach the amps before a reconfiguration of the layout would move the original HRS source rack from the side wall to between the speakers for the tidy version. The positioning of the Verity Audio 4-way towers and listening seat was obviously locked in to the millimeter. The prototype amps meanwhile sat on industrial caster platforms for easy removal after auditions back into project leader Samuel Egli's work bay for the next round of Swiss circuit massages.

R&D senior engineer Samuel Egli, Nagra's resident Djokovich

As it turned out, all the sound-producing bits and bobs had actually been locked in by the time of my visit. The next stage would involve adding the various protection mechanisms for the speaker outputs and power devices without affecting the sound; then starting work on physical layout and mechanical engineering which to us civilians could go under the umbrella of packaging job. Regardless of whether these monoblocks will end up as VPA-style towers with double or triple-decker internals; wider than deep, deeper than wide, heat sink exposed or hidden... they'll be big. The enormous power toroid which in the protos was concealed beneath a metal divider is the size of about a 2-liter cooking pot. That's because the octet of power Mosfets—four per push/pull phase—is biased for 50 watts of class A operation and doubles its rating with each halving of the load impedance. With absolutely zero negative feedback either global or local, full-bandwidth output impedance is shockingly low. One contributor to this are the eight decoupling caps with a coke can's diameter and about 1/3rd its height. Those weren't power supply capacitors which sat beneath the power iron cover. And they weren't coupling caps either since this is a front-to-back DC-coupled fully balanced circuit. As the next photo shows, the originally six caps had grown to eight by the time I visited. Their job is to lower the power supply's output impedance as it hits the circuit.

As resident golden ear and HD DAC project leader Philippe kicked off the audition with his customary Matthieu Passion Bach recording under Philippe Herreweghe—the same he'd brought during our listening session in my digs described here which he considers the most telling and challenging album in his vast collection—it rather instantly revealed itself that I was in the presence of a higher-order system than I'd heard before. Since the HD DAC fed the amps directly set to variable output mode via its remote-controlled analog attenuator, it equally instantly suggested that in the context of my own system, I'd not really heard the full extent of what the HD DAC can do. On a smaller scale, the same had happened to Nagra. The HD AMP project's initial brief had as a target the existing 60wpc MSA stereo amp's sound scaled up to a load-invariant quadruple its output. MSA x 4 in shorthand. As a result of endless circuit refinements, certain intermediate dead ends and setbacks but ongoing persistence, the monos had finally rather eclipsed the MSA on sound quality and sheer resolution. Now even Nagra's designers heard their HD DAC in a new brighter light. To explain this next stage proved rather simpler than one might expect given how according to the law of diminishing returns, advances in this context tend to get more and more subtle. Here there was nothing subtle about any of it. This was spelled out in capital letters.

From a technical perspective as the cause or enabler, we should probably look at extreme phase coherence; and extrapolate that on a good recording and contrary to popular opinion, there is an extensive amount of vertical (height) data encoded which most conventional system simply won't resolve. I'm talking about venue floor reflections which are recorded as time-delayed phase-shifted components of the sound and will be recreated if the playback hardware and setup allows for it. On the Bach choral masterpiece, the immediate impression was one of a Mayan step pyramid or an audio version of the Olympic bronze, silver and gold medal pedestals. It was clear as day that the orchestra sat below the chorus; and that the chorus rows themselves were stepped at different heights above it. There was no sound between the speakers. Rather, the sonic panorama occupied the entire front wall corner to corner, floor to ceiling. If you had drawn a grid on that wall, each little square of it would have contained sound at the same density. There was no center weighting, no fading of density to the edges of the stages, no curving into a semi arc and getting dimmer and hazier beyond the central 'stage lighting'.This wall a solid wall of sound but unlike its Rock equivalent, rendered in minute detail top to bottom.

Playing the first 15 tracks of one of my Munich HighEnd 2013 show CDs next, this very unconventional wide-screen HD panorama with very specific height data continued. A piano was clearly lower than a muted trumpet and violin whose players stood up. A drummer's cymbal tree sat above his snare drum which was more elevated than the bass drum. On a good recording involving piano, I could very clearly hear the recorded floor when the player unleashed fists of fury on his lower registers whose long paralleled strings not only energized the piano's sound body but compressed the air below it to reflect off the floor. It was just as obvious whether a piano faced the listener head-on for maximal width of the keyboard to have hand movements telegraph as lateral movement; or more or less angled to narrow the window in which the hands would move. A dancing clarinet player like Martin Frost could be seen to move forward and back and turn to the sides in momentary crouches quite regularly. Microphone positioning was revealed by the relative size of instruments. Close-mic'd positions would have them larger than real, farther mic'ing would render them with accurate dimensions. Recorded sibilance was located very specifically behind a singer's teeth, not moved to some ambiguous proximity elsewhere. Seeing all of that was effortless probably because the phase behaviour of this system—its time alignment if you will—operated with far less errors than usual. My ear/brain didn't have to sort out a scrambled facsimile.

1/ Sœur Marie Keyrouz "Touba Iahum"
2/ DIego el Cigala "Cancion Para Un Niño En La Calle" from Romance de la Luna Tucumana
3/ Nynke Laverman "De Toek-Toek Tuorren" from Nomade
4/ Aytaç Dogan "Herkes Kendi..." from Deva
5/ Martin Frost "Klezmer Dances" from Dances to a Black Pipe
6/ Johann Melchior Molter "Sinfonia Concertante in D Major"
7/ Antonio El Titi "Como Fusion (Rumba)" from Disfruto Flamenco
8/ Gianmaria Testa "Il Mio Gallo" from Extra Muros
9/ Romane & Stochelo Rosenberg "R' N Bis" from Double Jeu
10/ Time for Three "Orange Blossom Special" from 3 Fervent Travelers
11/ Luciano Pavarotti "Il Gladiatore" from Ti Adoro
12/ Oliver Dragojevic & Stjepan Hauser "Brod U Boci" from Noc Nek' Tiho Svira
13/ Miguel Poveda "La Bien Paga" from Coplas del Querer
14/ Adnan Joubran "Beat the Dream" from Borders Behind
15/ Karim Baggili "Silent Stories" from Kali City

That usually means embedded sensory conflicts which has a guitar's fundamentals arrive at different times or locations than its harmonics; which has a singer's mouth be wider than natural or vertically skewed from the diaphragm. Without any surround sound or additional height channels, this system's ability to draw a wall-to-wall canvas of consistent density and focus with very specific height assignments operated at a level of clarity and precision I'd never encountered before. This wasn't about sounding 'good'. This was about digging deeper into what had been recorded and making it visual. Obviously a presentation of such heightened data retrieval could become quite a sensory overload. And it's not the case that unless one were in an equally heightened state of expanded concentration, one perceives all of it simultaneously as separate facts. It's simply that whatever one's attention will momentary focus on—be it the staggered rows of chorus singers rising behind the orchestral forces; the tenor's narrow dynamic range and pinched timbre compared to the bass' gushier meatier range; the mechanical noises of a piano's pedals being clearly below the ivories; the lateral expanse of a drum kit; some tongue action of a vocalist that has you see the mouth or key clacks on a woodwind to suggest fingering choices—has more information to work with. That makes virtual sight so much more complete to fill in the missing bits of the 'ears only' playback experience when there's no parallel video feed.

Some internal fun at Apple's impervious expense (note all the broken arrows already fired).

The upshot of my visit was simple if humbling. One, there's something wicked brewing at Nagra which, I predict, will become their most important product of the last few years by far. That's because Nagra never had a no-matter-what muscle amp that'd do the job regardless of speaker, room size and SPL. Dealers loving their house sound but needing more power could never take on Nagra without also retaining another brand for its high-power amps. And very often that results in an either/or scenario. The HD AMP—final name TBA—should become that unconditional door opener. It should also move the company's pedigree into an even higher sphere and allow a fuller appreciation of the HD DAC's true potential than I could realize during my review of it. I'd heard it with the MSA over my own speakers. The HD AMP is another ball game well beyond just more power!

Two, going to many hifi shows over the past 15 years and crossing paths with systems wildly costlier than my own has, with very rare exceptions like Kevin Scott's superlative Living Voice Vox Olympian/Elysian setup, not relegated my own into the *§%& realms of disappointment upon return. More often than not in fact, I would prefer my own setup to be most happy to be back home. I'd been to Nagra many times before and never had been seriously rattled either. Obviously you expect a bigger system to play louder, go lower and scale up the experience. That it always had. Often it's simply not fundamentally better. It's just bigger. It wouldn't fit your own space. So it causes no envy just as visiting somebody else's palatial digs which house a big family won't have you feel crammed returning to your smaller two-person digs. This was different. For as messy as it may have looked—I could have cared less but I can appreciate that company boss Pascal Maroux on hand had trepidations about a journalist taken photos—this was a personal encounter of the fourth kind. Three-dimensional sound at varying degrees of 'holography' I'd heard before and already had. This type of height-specific consistent density wall to wall and up and down I'd not heard or had. How to identify the weakest link in my system now will be a personal task to see how much closer I can move my sound to Nagra's new rig without committing to a one-year rice 'n' beans diet. Ayee. There's never an end to it, is there? In this case that's no complaint though. It's a very real motivation now that the goal has been identified and lives just around the corner should I need any reminders.

To be sure, one doesn't need a system of anywhere near this magnification power to enjoy music with. In the right frame of mind, a shoddy car radio will do just fine to produce goose bumps or chicken pimples as a Serbian friend calls 'em. Particularly for a reviewer however who is tasked with and expected to describe sound vis-à-vis what's possible versus what's left underneath the table, it's a near requirement to advance the informativeness of one's reportage. Consider me properly primed for the next level. Whether and how soon it'll manifest on Mont Le Pèlerin is a rather different matter of course. There's desire. Then there's ability. The most important first step is always resetting the boundaries of knowing what we don't know. Here that equates to a new appreciation for what's actually possible; and what the gap remaining really is. As I said, that's humbling and motivating at the same time. It'll likely also get costly and involve some future - er, office equipment additions. For 2015, my tax accountant will roll her eyes I'm sure...
Nagra Audio website