Review: Martin Logan Summit X Speaker


Category: Speakers

PLANAR SMACKDOWN: The Martin-Logan Summit X vs. The Magnepan 3.7 and 20.7 Magneplanars

These days, it can be brutally difficult for audiophiles to audition equipment before they buy, and virtually impossible, in many cases, to actually compare two or more pieces of gear under the same conditions. Most of my equipment purchases have consisted of demo or used gear, but when I set out to replace my M-L Odyssey’s, I wanted the opportunity to do some serious auditioning, even if it meant buying new gear - and I got lucky. In hindsight, I thought it would be helpful to share my journey.

I’ve had extensive experience with both M-L and Magnepan products. Over the years, I have owned Magnepan IIa’s, Tympani 1D’s, Tympani IVa’s and then, as relocations dictated changes in listening rooms, Martin Logan reQuests and Odyssey’s. After 10 years with the Odyssey’s I was getting ready to move on and, in preparation for retirement, buy that “last” set of speakers.

But which? Magnepan and M-L have steadily improved their products, but advances in material science had dramatically benefitted piston (cone) multiway designs over the past decade. Being a long-time planar guy, I decided to begin on familiar ground and investigate the latest offerings from Magnepan and M-L. The Magnepan 3.7’s were receiving uniformly positive reviews and TAS had recently published Dick Olsher’s rave of the Summit X. Both brands still seemed to be punching way beyond their weight class, so to speak, and represented great value.

Here is where good luck intervened. Some business travel took me past Wilmington, Delaware, so I checked in with Overture A-V as they carry both lines. This turned out to be a great experience with a high-end dealer. Not only do they carry B&W, Focal, Magico, Avalon, and others in addition to Magnepan and M-L, they actually have demo models in stock to listen to. And the listening rooms are some of the best, if not the best, I have ever encountered. Period. I ended up working with two or three sales reps over an 18 month period and all were helpful, very professional, went out of their way not to let their personal tastes interfere with my listening, and bent over backwards to make both of my extended auditioning sessions a real joy. While it was obvious that they cater to a more exclusive clientele who leave my budget limits in the dust, I was always treated with the utmost respect and never received anything but courteous, professional treatment. Highly recommended!

THE MAGNEPAN 3.7
Round 1 was a listening session I requested simply out of a curiosity to hear the Maggie 3.7’s. Since I was living with the Odyssey’s, the comparison from the M-L line we chose was the Summit X. We used digital source material on hand at Overture and the electronics were McIntosh: a Mac front end with a pair of MC501 monoblocks driving the speakers.

Both speakers were impressive. From the midrange and up, and in terms of soundstage width, depth and imaging it was actually tough to pick a winner, but in that frequency range the choice is between a very fast electrostatic transducer and a killer line source ribbon tweeter and quasi-ribbon midrange.

Now the “buts …”

While there was plenty of punch in the lower end, the Maggies had a very obvious and rather abrupt drop off in bass response at 40 Hz. This was confirmed by the Audio Tools RTA app which confirmed a response “cliff” right in the 40-45 Hz region (for 20 bucks an audiophile with an iPhone is crazy not to have this app from Studio Six Digital,).

The active bass modules on the Summit X were extremely fast, tight, clean, deep and powerful, and integrated better with the stat panel than I have heard on any previous M-L hybrid design. However, there was one instance where the handoff between the panel and cone woofer jumped out. On a very high res demo recording, there was a passage by a solo acoustic guitar that consisted of a run from the upper to lower register of the instrument. When the signal was handed off to the woofer there was no change in timber, accuracy, imaging, dynamics, etc., but it was as if the “ambience” surrounding the instrument collapsed. I still heard the body of the guitar resonating, so a respectable amount of the 3D palpability and image of the instrument was preserved, but I could no longer hear as much of the “air” around it. My conclusion was that this spoke to the ability of the stat panel to resolve very low level information. It also occurred to me that with an instrument ensemble, or with source material of “normal” resolution, this change in the reproduction of ambient information would probably not be audible, or at least much less noticeable.

Putting aside the bass response cut-off, I felt that the Maggies provided a slightly more coherent presentation. At the same time I had a constant nagging sensation that I was listening to a mechanical device while with the M-L’s the sound seemed to materialize out of thin air. It occurred to me that the visual cue of two large black panels vs. a see-through transducer may have been playing with my perception, but the perception was there, nonetheless. The later audition of the 20.7’s offered a plausible explanation for this.

Bottom line: the Magnepan’s were stunners, especially given that they were less than half the cost of the Summit X’s. For a moment, I seriously considered the 3.7’s as a worthy replacement for my Odyssey’s but quickly realized that I would ultimately end up adding subwoofers with all of the attendant complexity and expense. That said, had I not already made the jump to monoblock amps in my system, I would have foregone the more expensive Summit X’s and paired the 3.7’s with a good pair of mono amps to gain the improvements in soundstage and imaging that they bring to the party.

Round 2 took place a year later, after the introduction of the Magnepan 20.7’s, and came at a time when I was now fully committed to replacing my Odyssey’s. This time I set up a more serious audition between the 20.7’s and Summit X’s with the assumption, given my previous experience with the 3.7’s, that this exercise would rubber-stamp my notion that I would be buying the new Maggies.

After some consultation with Overture, they selected a front end consisting of the Spectral SDR-4000 CD player and the Classe CP800 preamp. Amps were the Classe CA-M600 monoblocks. I wanted to listen for differences between speakers, and requested electronics that would work equally well with the reactive load of the M-L’s and the resistive load of the less efficient Maggies. Speaker cables were MIT Matrix HD 90’s, which made this the first time in my experience that the speaker cables cost roughly as much as the speakers I was listening to. (Holy Crap!)

This time I brought some of my own source material:

Lyle Lovett: Release Me, Curb Records (White Freightliner Blues, Garfield’s Blackberry Blossom)
A great sounding, but “dense” studio mix. On these types of recordings, when the volume is cranked up depth can flatten and the individual instruments can blur into a wall of sound leaving the lead or solo passages clear enough, but where the remaining ensemble is congealed into a compressed mess. I wanted to hear how the speakers handled the depth of the mix and the voices of individual instruments at louder listening levels.

Dave Brubeck Quartet: Live at Carnegie Hall, Columbia Records (St. Louis Blues, Pennies From Heaven)
My favorite Brubeck album. It was recorded live at Carnegie Hall in 1963, before a lot of studio processing nonsense took hold. This was also before Carnegie was renovated and the acoustics “improved” in 1986. I listened for the hall “air” and ambience of a small ensemble playing in a good acoustic venue which this recording captures wonderfully. It was also an incredible performance.

Holly Cole: Temptation, Metro Blue Records (Jersey Girl, Train Song, Temptation)
I wanted to hear the subtle information buried waaay down in the mix on this audiophile chestnut. For example, on “Temptation” behind Holly’s dark and somewhat satanic vocal rendition, the same melody line is sung in a subtle “angelic” voice. I listened to hear if I could hear it clearly as a distinct and separate voice positioned farther back in the soundstage.

Joe Bonamassa: Blues Deluxe, J&R Records (Blues Deluxe)
Never heard of Joe? Well, shame on you. If you appreciate electric blues, you are missing one of the best acts to come along in a generation. Even his acoustic sets are smokin’ hot (check out “An Evening at the Vienna Opera House”). Ditto his collaboration with Beth Hart (“See Saw”). This is an earlier 1994 release, but a nicely recorded studio album. Joe’s go-to axes are Fender Stratocasters and vintage Gibsons, which have distinctive sounds when being blasted through a low-wattage tube amp.

Vivaldi: Four Seasons, Philharmonia Baroque Orchestra, Nicholas McGegan cond. Even if you have umpteen versions of this Baroque warhorse, you need this one, too. This is Vivaldi with attitude, and the recording is a knockout. This is a small 20-piece chamber orchestra playing period instruments. I listened to hear the resonating bodies and the air around each instrument and whether their location remained stable in the soundstage as the volume was increased. And in the back of the ensemble is a harpsichord which, as you may know, has virtually no ability to vary its dynamics. When the full ensemble is playing, I listened to hear the distinctive, buzzy sound of the strings being plucked at the rear of the sound stage.

I listened to the Summit X’s first, then switched to the Maggies, and then back to the Summit’s, listening to each speaker for at least 45 minutes. I turned up the Summit X’s to what for me is a “satisfying” level that would be representative of a typical listening session. By “satisfying,” I mean a sound level and dynamics sufficient to provide visceral toe-tapping, party-on enjoyment without being overly loud or fatiguing. When I checked the SPL with the 6-Rivers app, it showed an average SPL of 87 db at the listening position, approximately 12 - 14 ft from the speakers. When we switched to the Maggies, I adjusted the listening level to the same average SPL to start and went from there.

THE MAGNEPAN 20.7
What a soundstage! The placement of individual instruments was solid, precise and stable at all listening levels. I first encountered the Magnepan’s ribbon line source tweeter in my Tympani IV a’s, and continue to believe that this is one of the best tweeters ever. Except that now in the 20.7’s (and 3.7’s) the handoff to the midrange panels is seamless. The bass was articulate, fast, natural, and plentiful, but also remarkably musical. “Natural” as opposed to hi-fi sonic spectacular.

The sound was decidedly more coherent than the Summit X’s, by which I mean timbres, dynamics and ambient information were presented in a consistent fashion across the entire spectrum of frequencies. I didn’t hear any obvious sound of the crossover network, or a signature of the individual drivers. Maybe if I listened long enough I would, but not during my relatively brief session.

So the sound field was amazing – a cut-from-one cloth tapestry of sound. But as I listened, I became more and more aware of the “weave” or texture of that tapestry. The individual images were razor sharp and distinct, but seemed to be tethered to the tapestry’s “fabric” and did not seem to completely bloom into the acoustic space. The words “polite sounding” sprang to mind, but this is neither fair nor accurate. By the end of the listening session, it dawned on me that what I was hearing was most likely a function of dynamics. More on this later.

THE MARTIN LOGAN SUMMIT X
The M-L’s are definitely more focused transducers. That is to say, with a smaller, 1-person sweet spot. For me, this not a big deal given my listening habits and dedicated room set up. For others, this may be a more critical consideration.

That said, the soundstage of the summit X’s seemed to be as wide as the Maggies, and even a bit deeper. The Summit X’s were decidedly better at retrieving the very low level details at my preferred listening levels. To be sure, those details were there with Maggies, but were less apparent until the volume was cranked up. The M-L’s presented individual instruments in a more holographic manner in that there was a better illusion of the individual Instruments being right in the room. Breathtaking, but the placement of the individual instruments within the soundstage was not as precise as the Maggies, being a smidge less stable with respect to relative placement. I think this is mainly due to the hand-off to the bass units at 350 Hz. The images of prominent instruments will move somewhat downwards toward the floor as they descend the scale unless they inherently produce a higher frequency component when played that is simultaneously reproduced by the stat panel, providing a cue to help locate the image (for example, the snap of fingers on the strings of a contra bass). The M-L’s do have excellent bass performance, but it is bass from powered piston subwoofers, not panels. So while it is very, very good, it is not 100% the same in character as the rest of the frequency rage. This is obviously not the case with the Maggies.

My big surprise was how I ultimately arrived at my decision, which slowly came as I became aware of my involuntary responses to the two speakers. For all of the glories of the 20.7’s, when listening to the Summit X’s I found myself becoming absorbed by the music and forgetting for the moment that I was there to make a “rational” decision. So, in spite of my pre-conceived bias, I ended up opting for the Summit X’s. Why? Because for me, they simply connected me more emotionally with the music.

Post listening session comments by my sales rep confirmed this (he offered these after I made my decision known). He noted that there was more “toe-tapping and head bobbing” on my part when listening to the M-L’s. He also offered his personal observation that the M-L bass setup was a “glass half full” situation, as the contour adjustments that these speakers feature make it easier to dial the speakers into a room.

After some reflection, I concluded that the Maggies were ultimately not as good at reproducing the subtle dynamic contrasts of the music at my preferred listening levels. And this is critical as it is the subtle dynamics that imbue music with emotion. One way to understand this is with the example of the human voice in conversation. A “monotone” voice is one without change in pitch and dynamics, and it comes across as lifeless.

To switch to an engineering metaphor, it was as if the Maggies had to be “biased” into a range of linear operation in order to properly capture dynamic contrasts. While I found the Summit X’s to provide a satisfying balance of dynamics and information retrieval at 87 db, the Maggies required a minimum of 92 – 94 db to get me to a reasonably similar satisfaction level. This brought into focus the comments made by reviewers of the 20.7’s such as the need for “room to breathe” and “bring lots of amp.” The extra db required to get me to audio nirvana required roughly 6-8x more wattage from the amps compared to the Summit X’s, and the need to play louder also resulted in a greater amount of acoustic energy being pumped into the room, which had to be managed through a combination of the room volume and by diffusion and absorption. No problem for the room at Overture. Your results may vary.

For me, the room itself (or potential lack of one) is a critical consideration. As part of retirement planning, we decided to take advantage of the recovering housing market, list our home, and rent for the few years left before heading off into retirement. So my speakers will have to adapt to a series of unknown acoustic environments before we settle down once again. The Summit X’s are a more svelte 75 lbs each as opposed to the Maggie’s ~175 lb per channel avoirdupois. They may also have to integrate into an actual living space in the event I do a stint in a rental property that does not offer the opportunity for a dedicated listening room. The size, bass controls (and WAF) of the M-L’s will make this much easier. While this situation did not influence my conclusions from the audition, it would definitely have served as a tie- breaker in the case of a dead heat.

BRINGING THEM HOME: REAL-WORLD PERFORMANCE
I have now used my Summit X’s in two listening environments, and with similar results.
During this time, the relevant portion if my system has consisted of Pass electronics (XP-20 preamp and the XA 160.5 monoblocks) and the Ayre C-5Xe(mp) disc player. Interconnects are XLR terminated HERO’s. Speaker cables – therein lies a tale.

Break in was pretty straightforward. I approached this task using the brown noise track on Ayre’s “Irrational, but Efficacious” CD and the repeat track function on the CD player. After 200 hours, I would say they were 90% of the way there.

As far as placement, in both rooms I plopped them down 3 to 4 ft from the side walls and a minimum of 3.5 ft from the rear wall (measured to the stat panel) and was able to dial in the bass very quickly. Imaging and soundstage width were pretty much good to go after getting distance to the listening position dialed in to within a half inch, which is not as straightforward as it sounds since the base cabinets are angled and the panels are raked back and curved. I discovered that the fastest way to dial these speakers into the listening position is to mark the listening spot on the floor and then use a tape measure to measure the distance from that spot to each of the front legs of the speakers. When the outer and inner legs are the same distance from the listening point, you are there.

The real learning curve for me involved rake angle and toe-in. While the bass sounded pretty good with minimal fuss (that glass was half-full, indeed!), the highs at first were bright with an almost unbearable glare above 4 kHz, and getting them to sound right took much longer than expected. M-L offers basic setup advice that essentially involves adjusting toe-in such that you are listening to the inner 1/3 of the panels at the listening position. However, even with the clever technique they recommend that involves reflecting light off of the membranes, I found that this was not sufficient to arrive at the optimum. It turns out that with respect to upper frequency response, the M-L’s are extremely sensitive to both rake angle and toe-in. The Summit X’s have an clever modular spike system that allows the rake angle to be adjusted over a 12 degree range and, while rake angle is important, it was the toe-in that ultimately made the critical difference in getting the frequency response just right.

Long story short: while exploring rake angle, toe-in angle and cables, I re-read Dick Olsher’s review and noted that he had set up his Summit X’s with no toe-in to get a flat high frequency response. So I began experimenting outside the box and discovered that a small degree of toe – OUT resulted in a smooth top end and acceptably flat frequency response in my room. This was not something that was suggested in the owner’s manual, and so was somewhat counterintuitive. In retrospect, I realized that the constraints of both listening rooms required a fairly narrow placement of the speakers to get sufficient clearance from the side walls, and the toe-out was necessary to get proper alignment with just the right spot of that “inner 1/3” of the stat panel..

As far as speaker cables, the M-L owner’s manual states: “Use the best speaker cables you can.” Believe it.

I had been using a pair of MIT 2 Biwire cables with my Odysseys that I loved, but which were now useless with the non-biwireable Summit X’s. I began with an 8’ pair of Cardas Crrosslink 6x that I had in the closet, and then moved on to a pair of Audioquest CV-8’s. The Audioquests smoothed out some of the treble issues I was having at the time, but I still missed my MIT’s. Ultimately, I bit the bullet and traded in the MIT-2’s towards an 8’ pair of MIT Matrix HD 28’s. I don’t know what kind of voodoo is involved with MIT’s “poles of articulation,” but it works for me and the M-L’s. If some interesting cables come up in the used market I may experiment further, but for right now I don’t feel the necessity.

Thus, after considerable fiddling around, I finally coaxed the same degree of performance out of my pair that I heard during the audition (Joy!) . I would caution any audiophile who finds the M-L’s to be too thin and bright after a casual audition to keep my experience in mind. As an FYI, I’ve uploaded screen shots from AudioTools showing 1/3 octave frequency response curves with a 0.5 second decay (playing pink noise) of the frequency response I was able to achieve setting them up by ear in my current listening environment . Some additional room tweaking is in order to continue to smooth out the peaks and valleys in the upper registers, and I think the roll off at 16 kHz might be the result of the “Listen” filter setting on the Ayre (not confirmed), but I’m happy with the results thus far.

Finally, I didn’t mention auditioning multiway piston designs. That’s’ because I didn’t. When I had completed the comparison of the Summit X and 20.7’s, I asked my rep to hear whatever multiway piston designs that he would recommend I also should consider. He said he would be happy to, but advised that if I wanted to hear something that sounded meaningfully better, that I would have to increase my budget to the $25,000 - $35,000 range, at minimum. I passed.

In conclusion, I don’t think you can go wrong with any of these planars, but you will face some trade-offs resulting from your own listening preferences and the limitations imposed by your room and amplification equipment. You’ll note from the playlist I used for critical evaluation that my preference tends towards performances by smaller groups and ensembles. During my high school days, I played several instruments and developed an appreciation for the intense physical connection between performer and instrument, and this has shaped what like to listen for. If my tastes leaned more toward large scale orchestral works, my choice of speakers would probably have gone the other way, as I think the Maggie’s strengths would favor such recordings.

In a perfect world, I would have bought both speakers and made my decision over a very, very long period of time. Well, OK, I suppose that’s not quite the truth. In a perfect world I would be auditioning Wilson Alexandria XLF’s. ;-)

Thanks for reading.

Associated gear
Pass Labs X-20 Preamp
Pass Labs X160.5 monoblocks
Ayre C-5Xe(mp) CD Player

Similar products
Magnepan 3.7
Magnepan 20.7
hthaller
What a remarkable job you have done explaining the differences, pros and cons! Thank you, and congratulations on the X! Check out my system's page for Odyssey modifications and especially the Mundorf crossover, which I would highly recommend and should bring your X to yet another level -see http://forum.audiogon.com/cgi-bin/fr.pl?vaslt&1249400348&openflup&23&4#23
Very interesting comparison as well as your Summit X setup in your home. I too have Summit Xs and have struggled getting their performance as desired according to the manual (I am using my benchmark media systems DAC1's volume control and the room is asymmetric and suboptimally treated so I had contributed it to that). I will try your "toe out" approach and see if it helps.
Ack: Thanks. Something to consider. I did modify the XOvers in my Tympani IVa's - had the outboard low pass boxes rebuilt with air core windings and Relcaps. I re-capped the high pass circuits in the high/midrange panels myself, as well as rewiring the panels with Cardas wire. Huge difference, as you can imagine.

Bigsam2: The first room I had them in had issues I suspect are similar to yours: squarish, irregular, and somewhat "live," which I attributed to the results I was getting. In this setting I was experimenting with speaker cables and rake angle. Optimizing both helped, but the sound was still too bright. This persisted in the second (basement) room which, due to insulation, is damped to the point of bordering on "dead." This is what prompted a re-look at toe angle. It was somewhat surprising how sensitive the X's are to this aspect of placement, but when you get it right, magic happens.