Usually, math like 3 x 1 = 5¾ is magical thinking where 'magical' is a friendly euphemism for wishful. If we frame it less friendly and more factual, we say impossible. If we attack the magical thinker's cap to get personal, we call him or her plain nuts. So much for usually. But in hifi, unusual things can happen. Let's call them unexpected occurrences where results exceed the sum of their parts well beyond the normal. To call out such unexpected synergy relies on more than just two or three trials. After all, the first one or two attempts could have been simply inferior. That'd make the last just normal and far removed from anything abnormally brilliant. It's also not about replacing a €1'500 source with a €15'000 one and raving over how much better it plays. That's expected and normal though not always the case. No, what 5¾ is about—and its wacky sum hints at the arbitrary unpredictable nature of it all—is far more abnormal. Supra normal. It's when a well-balanced system anchored by a given pair of speakers or headphones suddenly performs not in typical sideways but other-league mode; and does so not predicated upon a capriciously spendy acquisition. This isn't about the usual spend-more-get-more math. This is about the few happy exceptions of not spending more but spending on the right stuff where what is 'right' isn't at all obvious or even part of the usual MO.

That's because beyond a certain point of well-matched component competency, sideways is what most hifi is about. It's not about unequivocal unilateral betterment. It's about different sets of strengths and weaknesses. It's like an endless progression of ice cream or perfume flavours. If those be from within the same catalogue, they guarantee that each is informed by the same attention to detail, quality ingredients and execution. That's why, beyond personal preference, it happens on the level. And such wholesale leveling is very typical for hifi once we subtract overkill build, bling and redundant specs to consider just what's audible. To come across a balanced combination that truly escapes the horizontal domain and hits the vertical shift into the next higher octave or beyond... that's rare. It's 5¾. For all of 2016, I've got just three examples; only one of which I chanced upon during the actual review. The two others happened well after. That's how this goes. At times, it's a function of something new showing up for a short while. It creates a magical combo whose discovery would never have come about otherwise. It's pure chance. Dumb luck. At other times, it might be trying something seemingly silly, for shits 'n' giggles or by sheer accident. "Everyone knows that's not gonna work". When it flips such common knowledge on its head and goes where that thing has never gone before, one looks at 5¾. It's because this happens so few and far between, even for busy reviewers who shuffle (shovel?) gear for a job, this feature came about. Whilst the room remains the big variable to still limit how much of such findings translate for others—headphones neatly eliminate that—I highlight such unexpected combinations in the hope that others may be inspired to replicate them, either precisely or conceptually.

Let's start with the Zu Druid V. Common wisdom looks at its 16Ω/98dB spec, then filterless widebander augmented by a compression tweeter. Without a second thought, it proclaims SET or tubes in general. And perhaps that crowd is right. It'd lead to recommendations from Audion, Leben, Mastersound & Co. Since I'd gotten out of valve amps, I simply wouldn't know. What I did try instead were our Pass Labs XA30.8, FirstWatt F5, F6, F7 and SIT1, Clones Audio 55pm monos, loaner Wyred4Sound SX-1000R, Goldmund/Job 225 and Vinnie Rossi LIO with DHT. They all worked. They all made sound. Some did so better than others but none were poor by any stretch. Eventual favourites were the low-power single-ended single-stage no-feedback SIT monos from Nelson Pass. That vindicated SET with transistors. It's where sleeping dogs would have snored on blissfully. Except then I had a most unexpected happy ending with the Rethm Bhaava, my second 5¾ of today. This led me to thinking. It happens, occasionally. What amp did we have that could duplicate the 15wpc Bakoon AMP-12R in general DC-coupled ultra-wide bandwidth thingness? The Crayon Audio CFA-1.2. It'd already been last amp standing in my soundkaos Libération review. When those open baffles packed it back up and the Druids returned, I leashed those to the Crayon. I suddenly realized—duh!—that of all our amps, it was the only one I hadn't yet fed to the Zu wolves. Dispatching a stout 45 watts into their 8Ω-halving loads, I had one of those vertical moments which, unlike the other kind, just lasted and lasted. Deciding between Wyred4Sound STP-SE, Pass Labs HPA-1 and Nagra Jazz preamp was the final garnish atop this dish. Ditto for cables. Ditto for EnigmAcoustic's Sopranino super tweeties. The decisive thing, the main stew, was really the Crayon. It did the SIT1's tonal and soundstage sophistication, then injected greater ballsiness, shove, dynamic conviction and scale. This wasn't about the usual 'more'. How it interacted with the entire presentation elevated it into a higher league and out of the sideways business wherein all our other hardware options had worked. Actually, this had nothing to do with dumb luck. Rather, it was the dumb resident oversight committee in action. I'd overlooked a strategic combination and only considered it in the wake of an equivalent one I'd made reviewing another not that dissimilar speaker. Better late than never!

On the wallet level, this replaced a €20'000 pre/mono combo with a €4'250 integrated. Admittedly our final setup splurges again just because. We own the €10'750 Nagra Jazz so might as well use it. That sets the Crayon's attenuator to full tilt aka bypass. The Nagra in this context is slightly juicier than the $1'799 Wyred preamp and allows use of the Aqua Formula's transformer-coupled XLR outputs which add a subtle viscosity over its RCA outputs. Incidentally, the Wyred beat out the €10'000 Esoteric C-03 to upset trophy thinking. Regardless of those shenanigans to squeeze a few more drops from the basic amp/speaker recipe, the real facilitator of this sonic 5¾ was the Austrian. Like the Job 225 though of less power, it's a fast lean circuit. Yet the Job's handling of the upper freqs and the Druid's Radian tweeter didn't get on. Meanwhile the 7.5wpc of the Bakoon which also falls into this general sonic school lacked spunk and upper-bass crunch. It was squarely underpowered. The Crayon not only split the bill and added refinement over the Job, it then took off into the fast lane. How these things work out is anyone's guess. There's nothing in the raw specs to predict it. In the end, it's a matter of trial 'n' error. The more you try, the more you err until the elusive magic shows up. Suffice to say that in our digs, the Crayon/Zu combo now is the one to beat. Having owned tube amps from Ancient, Art Audio, Octave, Woo and Yamamoto and reviewed quite a few others, the Crayon's calibre of Druid synergy removes all thinking of their kind. Which segues into my second discovery, this one predicated upon obtaining a new cable:

Final Sonorous III & VI. These full-size sealed headphones from Japan arrive with only a 3.5mm terminated cable. After I'd acquired the COS Engineering H1 DAC/amp which outshone an Auralic Vega plus 2 x Questyle CMA800 combo in fully balanced/mono mode, I grew more and more itchy. I just had to try the Final over it balanced. Ken Ball of ALO Audio to the rescue. In short order he made me up a deep-cryo'd custom leash ending in two 3-pin Furutech XLR. With that small mod, the Final VI became my favourite in-house cans. They moved to the top of a list which includes Sennheiser HD800, Beyerdynamic T1/T5p, HifiMan HE-1000 and Audeze LCD2/LCD-CX. Incidentally, the last three of those and the Senns also were recabled for balanced drive. Was this reshuffle a function of Ken's cable? To simply driving the Final's balanced, period? To some special mojo between COS and Final? All of that combined? I plead the fifth. All I'll say is that if you haven't tried it, you really have no grasp over just how good these affordable cans are. Honestly, I think Final do themselves a real disservice by not including an even basic balanced wire harness. What's more, the fussiness my review of the VI mentioned—about their dynamic/BA combo being critical in the upper midrange/lower treble—completely vanished in balanced drive. It's known that within limits, big power can ameliorate or diminish frequency domain criticisms. That's likely because more headroom fills out response reactions to impedance variations. Whatever it really is, in the case of the H1+VI pairing, my prior observations on a presence region that could veer into the frosty or glassy when the hybrid Sonorous encountered less than sterling recordings went up in smoke. Now they are omnivorous regardless of music and HD800 slayers. That and the even bigger-league performance were entirely unexpected.

Which gets us at three out of three, the Bakoon AMP-12R/Rethm Bhava combo. Once again, a 93dB widebander with active bass system predicts low-power valves as the ideal mates. As it turned out, nothing in our inventory came close to the Bakoon. With it, Rethm's $3'000/pr entry-level model upgraded to compete—here I rely on memory but given triangulations with similar speakers, I'm confident—directly with their next model up. With our remaining amps, the modest public-address driver from Boston Acoustics operated with clearly lower speed, lucidity and transparency than Rethm's more upscale custom options or a Voxativ. Yet when Bakoon took the reigns, it suddenly was a different game altogether. Because this was caught in the review proper not hindsight, I won't say more except that apparently nutty matchups should never be disqualified on principle. If you have opportunity to try one, go nuts. It could become just another told you so. Nothing lost. Or it could be that once-in-a-purple-moon hit. When that happens, 3 x 1 really do make 5¾. Just like with an abacus, it's hands-on business to find out. If you're too lazy to rely exclusively on the virtual domain, you'll never know. Just thinking about it or discussing the probable pros and cons in a chat room doesn't cut it. You have to actually do it. Then it's up to you what kind of strange math you want to call it. Personally, I'll stick by 5¾...