The Best Loudspeaker I’ve Heard? The Magico Q5

OK. I’m fully aware that I’m going to stir up a storm of criticism from Wilson owners, Rockport owners, TAD owners, Nola owners, Vandersteen owners, Focal owners, Hansen owners, Scaena owners, YG owners, Magnepan owners, Quad owners, MBL owners, MartinLogan owners, Kharma owners, Marten owners, et al.—not to mention infuriate almost every single one of the guys who make these terrific loudspeakers (and a myriad worthy others). I also know that I’m going to reignite the ridiculous canard that I’m somehow in Magico’s pay or pocket, which, BTW, is one reason why I’ve taken my own sweet time about getting around to writing about this loudspeaker, letting others in the critical fraternity have first say in print. And I also know that I’m going to catch hell from the “b-word” contingent which pretends to think that I call anything I like “the b**t” (even with a question mark, which I will explain)—funny how much less vocal some in this group are when they sell/manufacture something I commend). And I also know that I’m going to be accused of hypocrisy or stupidity or deafness or willful inconsistency or some combination of these things (and worse) for using the “b-word” about the Magico Q5 after having spent some ten pages eighteen months ago explaining why I used the “b-word” about its predecessor, the Magico M5. (And, I fear, I’m going to be pilloried by those readers who bought the M5s in part because of my rave review.)

My loins are girded. I’ll take it—and give it back, if need be. But first a word or two of preface.

The main reason I haven’t written about the Q5s until now is that I wasn’t at all sure they were great when I heard them at last year’s CES—and I expressed these reservations in my show report. While I wouldn’t have gone anywhere near as far as my colleague Alan Taffel did, when he told me in passing that he thought the Q5s were “awful” after he heard them in The Venetian, I did think they had…problems. Not all of their own making.

First and foremost, the gigantic room they were being shown in wasn’t doing them any favors. It had a huge built-in resonance around 60-80Hz that, apparently, couldn’t be fixed—no matter what Magico’s Alon Wolf and Yair Tammam tried. Second, no single stereo pair of speakers I’ve heard could have adequately filled a space that size, no matter how physically large, full-range, or fundamentally excellent they were (just think of the way big MBLs, Wilsons, Focals, German Physiks, and YGs, among many others, have sometimes sounded at loud levels in morasses half the size of this one). Third, although I didn’t know this at the time, the pair of Q5s being shown were prototypes that did not use some of the parts spec’d for them because those parts weren’t available at show time. Since CES, something like 60% of the passive components in the Q5 have been replaced in current production models with these custom-made parts. Fourth, the Q5s were being driven by my favorite solid-state electronics, the Technical Brain TBP Zero v2 amps and TBC Zero preamp, which, though more colorless and higher in resolution and dynamic range than any other solid-state electronics I’ve auditioned, aren’t inherently warm, sweet, “liquid,” “beautiful,” or gemütlich in tonal balance; nor are they big-sounding in the bass, like some of the other electronics that Magico has favored. With sealed-box speakers in a gigantic room, a little added warmth and heft, particularly on the bottom, is not an altogether bad thing; the TB gear doesn’t supply this.

Then there is this: When I went to CES, I had just got done calling the Q5s predecessor, the M5, the “best speaker I’d heard in my home.” At the time I wrote the M5 review—and for the year-or-so I spent listening to the M5s and blogging about them before writing the review—Magico had, uh, failed to inform me that their replacement was in the works. Somehow Magico also failed to inform me that their replacement was going to cost $30,000 less than the M5s did, and that it was designed to outdo them in every single parameter of performance! (Those who think I’m “in Magico’s pocket” please take note of this.

Finally, if I were completely honest, I’d have to admit that when I went to CES I wasn’t so much rooting against the Q5s as I was rooting for the M5s—partly out of pride and partly out of long affectionate experience with the latter. Not that I thought the M5s were perfect.

Rather than make you re-read my entire M5 review, let me recap why I thought the M5s were the best big dynamic speakers I’d heard, and also where I thought certain others speakers bettered them in spite of their overall superiority.

First, why the M5s were (and are) great. As I said, at the beginning of the M5 review, IMO the only reasons to buy big multiway dynamic speakers are loudness and bass. The problem is that while the ability to go louder more cleanly may allow you to play back large-scale music at more lifelike levels (and, in some cases, to expand the soundstage to fill larger spaces), higher less distorted SPLs (all by themselves) don’t necessarily guarantee superior dynamic range and scaling. The ability to play very loudly accurately doesn’t entail the equal ability to play very softly accurately, and dynamic range depends just as much (and with some music entirely) on the ability to reproduce low-level dynamics, textures, and timbres as it does on the ability to blast you out of your listening chair on fff tuttis.

Just as “iffy” with big dynamic loudspeakers is the issue of coherence in the bass (and of coherence overall). Because of their large woofer(s), big multiways do tend to go lower (or at least to give the impression of going lower with greater authority) in the bass than smaller speakers with smaller drivers, but they also (almost invariably) exact a price for that low bass (or the impression of same). As I said in my M5 review, it is rare to hear a big speaker that does not sound grossly discontinuous in the bottom octaves—to hear a big speaker in which the bass driver doesn’t sound as if it is playing at a slightly (or markedly) different tempo and with slightly (or markedly) different timbre and resolution than the other drivers it is mated to. To my ear, big speakers always sound a little like hybrids. (Subwoofed systems almost always sound this way to me, too, as do literal “hybrids” that attempt to pair electrostatic or ribbon or horn drivers with cones.) Big dynamic speakers also tend to sound as if their enclosures are playing along with their woofers, storing and releasing their own low-frequency information, which adds thickness, blurring, and confusion to bass notes, peaks up the midbass, further excites room nodes, and makes the loudspeaker enclosure itself far more audible as a sound source, fatally compromising the speaker’s “disappearing act.

There is this, as well. Some big speakers with vaunted “low” bass don’t really go that low. Certain (not all) ported speakers, in particular, can give the impression of tremendous low-end slam and extension, even though they aren’t really outputting much bass below 35-40Hz. What they are doing—and I have to admit that it can be a VERY realistic and exciting effect, particularly with rock music—is adding port, enclosure, and driver energy in the mid-and-upper bass, where things like Fender bass and toms and kick drums live. Since this is also the spot where most room resonances live, the blessing can be very mixed in many rooms—once again, reminding you (unmistakably) that you are listening to a loudspeaker in which some octaves are being selectively reinforced by room, driver, and cabinet noise.

What made (and makes) the Magico M5 so special was that it was the first big multiway dynamic in which, to quote my wise friend Andre Jennings, “the box didn’t seem to be playing along with the bass.” Here, for once, was a Big Boy that sounded like one thing from top to bottom—that wasn’t shelved in the bass or treble, like drawers that have been left partly open. On top of this, its drivers were audibly (and measurably) lower in THD at average and very loud levels, bespeaking unusually low levels of cone coloration. At the right volume, the M5 was the first big speaker I’d heard that disappeared as a sound source the way a really well-engineered little speaker does. In fact, the M5 sounded very much like an overgrown, much fuller-range, lower-in-distortion Magico Mini II, one not just with more extended bass but with higher resolution midrange and treble, to boot.

There were all sorts of other reasons why I loved the M5s. But its disappearing act and its (in my experience) unparalleled seamless coherence in the bass (and everywhere else) were primary. HOWEVER, there were some things that the M5s didn’t do as well as other speakers. I’m not going to list them all—for that you’ll have to read the entire review at www.avguide.com/review/tas-196-magico-m5-loudspeaker, where I devote a page or so to this very subject—but I am going to talk about two of them.

First, there was the issue of grain or opacity. As I noted in my review, it is our custom to compare speakers that sound like “one thing” to those paragons of speakers that “sound like one thing,” electrostats, which have a single driver (or a single kind of driver). The M5s came closer to the electrostatic/single-driver ideal than any big speaker I’d yet heard. Still and all, and here I’ll quote myself: “Cones aren’t quite as high in resolution and low in grain as ’stats; even the [Magico] Nano-Tec drivers add just the slightest overlay of texture to foregrounds and backgrounds, making the difference between listening to M5s and CLXes rather like the difference between viewing a slide enlarged and projected on a screen by a Leitz projector and viewing the same slide on a light table with a loupe. The CLXes will tell you a bit more about how a record or CD has been recorded and engineered.” In other words, electrostats (and certain planars) had slightly lower noise and higher low-level resolution than the M5s.

Though I didn’t realize it at the time, I now think that the overlay of texture or graininess that I heard with the M5s (and didn’t hear with CLXes or Quads) was also at least partly responsible for my other chief reservation about the big Magicos: their inability to fully strut their stuff at low volume levels. I mentioned this very subject a few paragraphs ago. The ability to play loudly convincingly does not guarantee the ability to play softly with equal realism. Although the M5s were and are anything but lacking in inner detail at the right volume, you had to play them a bit louder than mezzoforte to kick them into fullest life and fullest resolution. What this meant was that you never really got a true pianissimo from the M5s. By turning up the volume to the speakers’ SPL comfort point, you literally raised the level of very quiet passages (while also, simultaneously, raising the volume of very loud passages). Though the ratio between loud and soft was sensationally realistic, the level at which the softest passages were being played back wasn’t. It was as if you were losing a “p” at the soft end and adding an “f” at the loud one. Understand that this is not a drawback unique to M5s. In fact, I’d go so far as to say that it is the rule with dynamic loudspeakers and also with many planar-magnetic speakers. Indeed, the only speakers in my experience that play true pianissimos clearly and realistically at true pianissimo levels are electrostats (and a very small number of horn loudspeakers, although horns tend to have problems blending their drivers, which ’stats don’t). Unfortunately, what dynamic speakers do to pianissimos, electrostats tend to do to fortissimos. They take an “f” or two away from the loudest passages, while preserving ppp’s at precisely the right SPLs.

Now I didn’t know what to attribute this grain/opacity/low-level resolution problem to other than the fact that the M5 was a big dynamic loudspeaker and such problems (usually to a much more marked extent) come with the greater mass of the drivers, the more complex crossovers, and the larger enclosures of this kind of speaker. To give him his considerable due, it was Andy Payor of Rockport (none too pleased by my rave review of the M5s, BTW) who first pointed out precisely what it was that I was hearing as grain and opacity and a slight low-volume-level limiting of dynamics and resolution: the M5’s squared-off, birch-ply-and-aluminum enclosure. I didn’t buy Andy’s argument at the time—and I owe him an apology for that—but, after all, the M5’s had the best disappearing act I’d ever heard from a large dynamic loudspeaker, and I couldn’t see (or, rather, understand) how a speaker that disappeared so completely into the soundfield and played so non-resonantly in the bass could also be being hamstrung in certain ways by its otherwise invisible/inaudible box. In fact, I thought its box was one of its glories—and, in terms of its bass coherence and disappearing act, still do.

What I didn’t understand then was that what the box was taking away with one hand it was also, apparently, ladling out with the other. The M5’s slight overall graininess and opacity were, in fact, largely being caused by the admixture of energy stored and released (after a short delay) by the wooden enclosure. This was not a big resonant signature, mind you. (The M5 actually measures quite well in waterfall plots.) But it was enough of one to add a little Kersey-like texture to the soundfield.

How do I know this? Well…last weekend I spent three days in Berkeley listening to my own music through production model Q5s powered by two of my reference components, the Soulution 700s and 720. And, a bit to my surprise given CES and a couple of biases I will discuss in a moment, the Q5s had none of the slight grainy texture and opacity of the M5s; nor did they have the M5s’ problems with low-volume-level dynamic scaling and resolution. In fact, the Q5s managed to sound almost exactly like a Quad 2805/2905 electrostats, with slightly richer timbre (yes, richer timbre than Quads), much more extended, linear, and powerful bass, considerably sweeter and more extended treble, dynamic range and three-dimensionality that Quads can only dream about, and the kind of low-level resolution (even at low volumes) I’ve only heard before from the MartinLogan CLXes! Even a long-time Quad owner like our publisher Jim Hannon, who joined me for one of the listening sessions, was overwhelmed. “I think this might be the best speaker I’ve ever heard,” said he. Which made two of us (three, if you count Alan Sircom, editor of our sister magazine HiFi+, who is already on record calling the Q5 “the best speaker in the world,” for which go to www.avguide.com/review/magico-q5-loudspeaker-hi-fi-74).

So what changed between CES and last weekend? Well, as I already noted, almost 60% of the passive parts in the production Q5s (including several very critical, bespoke resistors designed exclusively for the Qs) are different than the parts in the prototypes Magico showed at CES 2010. And, of course, the room I listened in in Berkeley was much smaller (close to the size of room I have) and much more carefully treated than the Venetian echo chamber. The amps and preamp were also different.

But I’ve got the distinct feeling that it wasn’t so much all that changed that made the dramatic difference in the Q5s’ showing, although I’m sure these changes played their parts. It was, I think, all that the Q5s had going for them to begin with—but didn’t get a fair chance to show in Vegas because of show conditions. And what they had going for them from go was a much quieter enclosure and a far superior tweeter, although it took my trip to Berkeley to prove this to me.

As you undoubtedly already know, the Q5 uses a box constructed entirely of 6061-T aluminum panels, which are CNC-milled and gorgeously finished in Magico’s own machine shop in San Jose, CA. So by the way is the elaborate interior bracing of the new cabinet—for which see the photo. I will have a good deal more to say about the science behind this enclosure, the bracing, the way these things are fitted together and damped, and other niceties of construction and parts selection when I formally review the Q5s. But now let me confess to two biases that kind of set me against the Q5s even before I heard them in Vegas.

First, aluminum may be a swell material from a stiffness viewpoint, but it’s not so great from a damping one (which is why combining it with stacked birch plywood boxes seemed like such a nifty idea). Undamped aluminum tends to ring like a bell when excited by acoustic energy. At least, that has been my past experience with speakers with aluminum enclosures—and I’ve reviewed and heard a few.

Second, beryllium may have the highest acoustic transmission speed of any metal, but until I heard the TAD CR-1 (one of the truly great loudspeakers, by the bye) I’d never met a beryllium tweeter I really liked—and the Q5s use a beryllium tweeter. It’s not that that these things aren’t marvels of speed and extension and energy transmission and low distortion; it’s that, the CR-1 excepted, they always stick out like marvels of speed and extension and energy transmission and low distortion, the way Magnepan’s “true” ribbon tweeter sticks out from its quasi-ribbon and/or planar-magnetic brethren. Their very superiority makes them sound different than the more conventional drivers they are mated with, which tends to spoil the “single-driver” effect (which, as I’ve noted before, is as close as speakers come to a “no-driver” effect).<o p=””></o>

When I went to CES, I confess that I had both of these biases in mind, and while I can’t honestly say that I heard the beryllium tweeter or the aluminum enclosures adding metallic zip and coloration/ringing to the sound in Vegas, I secretly thought they may have played some part in the Q5s’ less-than-stellar showing. In fact, I was rather expecting to hear these metallic colorations on my visit to Berkeley, where the far less resonant, far less bass-boomy environs of Alon Wolf’s listening room would allow the Q5s’ “true” character to show much more plainly.

It is abundantly clear, now, that I was wrong. The Q5 have no metallic coloration that I can detect—in the treble or anywhere else. In fact, as I just got done saying, the Q5 seems to have less coloration (driver, enclosure, crossover) than any dynamic speaker I’ve ever auditioned, and appears to be fully equal (or damn close to fully equal) to the finest electrostats in this regard. Its resolution is phenomenal. (I played a cut from Sound The All-Clear that I’ve been listening to regularly on my home system for weeks and heard so many details I simply hadn’t picked up before that I lost count.) Its timbre is gorgeous, top to bottom. Its dynamic range is almost horn-like, going from whispery pianissimo to thunderous fortissimo and all stops between without any of the M5s’ slight dynamic-compression of lower-volume-level passages. Its bass is just right, IMO, although I can see where rocknrollas might miss some of the added midbass slam that really good ported speakers supply. (The quality of sealed box bass versus the quality of ported bass will always make for a sticking point for some listeners and some music. Nonetheless, I predict that most classical, folk, and jazz aficionados will be in bass heaven with the Q5.) Its new tweeter is a marvel of speed, detail, and (to my amazement) utter unfailing sweetness, making for some of the most gorgeous string and wind tone I’ve heard on a stereo system. Its imaging is more precise than that of the M5s, its staging every bit as vast, and its disappearing act even better.

So why did I put that question mark after “The Best Speaker I’ve Heard?” Well…I listened to the Q5s closely for the better part of three days, but I still haven’t heard them in my room, with my components, and my sources. Until I do, we’ll have to wait and see whether that question mark stays or goes. There is this, as well. As I said at the close of my M5 review, there are other speakers on the horizon—from Rockport, YG, TAD, Wilson, and others—that may prove competitive. Some of them I will audition at RMAF. Some I will hear at friends and colleagues’ homes. Some I may review. As I also said at the end of my M5 review, there is room for more than one “best” in high-end audio, and I have the distinct feeling that we will soon be hearing about other nominees.

That said, if I were a betting man (and I am), I’d wager that that question mark goes—and the Q5s stay (for a long long time) as my references.