Wilson Audio Specialties Alexandria XLF loudspeaker Page 3

Have you heard Roberto Gerhard's cantata The Plague? Better to hear it than buy it. This 1973 performance, with Antal Dor†ti conducting the National Symphony Orchestra and Chorus (LP, Decca Headline HEAD 6), recorded at Constitution Hall by an engineering team led by Kenneth Wilkinson, is a gruesome affair that is every bit the horror show its title suggests. But what a recording! The timpani thwacks sound great through the MAXX 3s—but through the XLFs they went deeper, were better controlled and far more transparent, and lacked a slightly hard leading edge that I now know is a coloration produced by the MAXX 3s. I'd never heard this record sound so powerful, so spatially coherent, so tonally convincing. The XLFs' ability to reproduce an illusion of depth, despite being placed so close to the walls, never failed to amaze me. And both the chorus and the narrator, Alec McCowan, sounded eerily real, with the best balance I've heard of vocal sibilants and body.

Unlike the bass output of most large speakers, which tends to soften and lose shape at low SPLs, the ca 94dB-sensitive Alexandria could be played at whisper levels with no loss of bass structure or rhythmic integrity. The cleanness and precision of the XLF's ability to start and stop at low frequencies and low SPLs was unique in my experience.

Not that I specifically listen for such things. I notice them only when my wife screams from upstairs for me to "Turn those effing things down—you're shaking the whole house!!" Otherwise, I'd always want to listen at realistic SPLs!

A $200,000/pair speaker capable of such robust bass does need to pass certain tests—such as hearing if bottom-end weight clouds the lower registers of female voices. I played Ella Fitzgerald, Diana Krall, Joan Baez, and some of the deeper-voiced jazz singers, such as Sarah Vaughan and Nina Simone. When bass was not supposed to be present, it wasn't. When one of these chesty singers reached down to the lower end of her range, only the appropriate low-frequency energy appeared: focused, of proper size, and in context—not as general, otherwise unidentifiable "bass."

I have an original pressing of the superb recording of Canteloube's Songs of the Auvergne, by the late soprano Netania Davrath with an orchestra conducted by Pierre de la Rouche (LP, Vanguard VSD-2090), said to be the definitive version of this cycle of orchestral folk-song settings. It's the only one I've heard, but it puts Davrath in a warm, large-sounding space that has never been more apparent than it was through the XLFs—yet her voice, floating magnificently between the speakers, never mixed with it. I had never heard her small-scale dynamic vocal gestures so clearly delineated.

The XLF's harmonic presentation was as fully realized as its performance in every other parameter of sound. The full ranges of instruments in well-recorded symphonic music and orchestral film scores were reproduced with rich, full palettes of colors that would surprise skeptics who regard the Wilson "house sound" as overly lean and analytical. The XLF was anything but. Yet instrumental attacks were naturally fast and clean, sustain generous, and decay into blackness complete.

Despite its wide bandwidth and complex design, the XLF's top-to-bottom integration of its drivers' outputs was masterful, surpassing that of any speaker I've reviewed. From the very bottom, which Wilson claims goes down to 19.5Hz, to the very top, a claimed 33kHz, the speaker produced vast, seamless pictures or delicately drawn, equally unified miniatures, as appropriate.

The most familiar recordings—at least, those that I had time to play—expressed subtle, occasionally dramatic, new, and often profound musical information in just about every performance parameter. Listening to music through the Wilson Alexandria XLFs was a transformative experience. And that's the least you should expect for $200,000.

Can a pair of loudspeakers possibly be worth $200,000? Can an automobile? Can a diamond? In all three instances, the answer can be Yes. Value is in the ears, hands, and eyes of the potential purchaser.

But a loudspeaker costing $200,000/pair should represent a fully realized concept that produces the ultimate expression of every performance parameter related to musical accuracy. It should reproduce the full audioband, from 20Hz to 20kHz, and do so while seamlessly integrating the outputs of its drivers to produce exceptionally linear frequency response from bottom to top across a usefully wide listening window, along with stable, well-controlled power response in the upper frequencies.

It should set a very high standard of very low coloration and distortion, particularly in the difficult-to-reproduce low-bass frequencies. It should have unlimited micro- and macrodynamic authority, and be able to play at very high and very low SPLs and everywhere in between, and sound equally good at all points along that volume scale.

It should set new standards of transient clarity, transparency, and purity. It should accurately express the harmonic structures and timbral and textural characteristics of musical instruments, limited only by the quality of the recording.

It should take to new levels the focused reproduction of three-dimensional images, as well as soundstage width, depth, and height.

It should do all of these things in a balanced, seamless way to produce a transparent loudspeaker that gets more out of the way of the music, than other speakers.

It should be sensitive and relatively easy to drive, and be designed in such a way that its performance can be optimized for a wide variety of real-world rooms both large and small, and under difficult acoustic conditions.

Finally, it should be built to the highest standards of fit'n'finish inside and out, and look great as well—although, of course, form must follow function, and tastes will differ.

And, technically speaking, it should sound amazing, get your heart racing, and set your audiophile hair afire each and every time you sit down to listen.

For all of those reasons, and probably a few I've missed, the Wilson Audio Specialties Alexandria XLF is worth $200,000/pair.

Wilson Audio Specialties
2233 Mountain Vista Lane
Provo, UT 84606
(801) 377-2233

Mike Lomond's picture

Have you heard it Pauly.  Or are you another critic that, depite being in the "Hi Fi Industry for 30 years" ( Who with, what doing?), can judge a product without listening to it.

I have no idea what it sounds like.  It may sound awful or it could transport me to realms untravelled.

What gets me about the Stereophile comments section are the naysayers, the armchair critics, experts on products they've never seen, let alone heard.

"David Wilson, amateur!  That guy just throws a few drivers in a box and wants 200 grand.  I could do better than that".

Give me a fucking break.

Many posters here try to suck the joy out of all that's fun and positive about this marginal, slightly bizarre, yet wonderful hobby.

Thanks JA.  Enjoy the mag.





Paul Luscusk's picture

Mike among the companies I worked  for as a rep where Mc Intosh, Dynaco, Hafler, and Hafler Pro, Sherwood,Rockford  Fosgate Jim Fosgate,AR, NHT, ADS, NAD, PSB,  Ixos, and Esoteric Audio. I also worked as a Buyer for 8 years,and did  two of the LA Stereophile shows. SO you give me a break. I have heard many Wilson speakers(Not this one) and I'm not impresed by them. I have never judged a audio product without a listen.If you had read my post I was defending Micky's choice  to by the Wilsons. I just said they don't do anything for me as a audiophile.

BillK's picture

I ask this simply because I'd never heard anything I liked from earlier Wilsons, including generation of W/Ps and the Maxx 2s, and even some newer models like the Sophia 3, but to me the Sasha and Alexia have been truly special sonically, finally not suffering from the "cones in a box" disease of most dynamic designs.

billyjul's picture


it is not difficult to do excelolent speaker for good DIYER

they no mistake 

if you use corectly a driver, you can't go wrong, on internet there a quantity of utilitise to simulate parametters, listenig for learn how make a speaker

but construct a driver is far difficult, it is a crutial point, a good speaker bigin whith good drivers, and goods drivers choice to have a good intégration

whith a good reflexion you can make not a good, but a very good speaker in 2voice, a three voice is more complicate

i think, 


"David Wilson, amateur!  That guy just throws a few drivers in a box and wants 200 grand.  I could do better than that"

yes, i agree his speaker have problem, the integration of the drivers, in the cabinet and the design of this are the problem i think

an exemple, the focal berylium tweeter have to be flush mounted on the cabinet, is you don't do this , the response curve is awfull, not flat, flush mounted it is excellent, and the sound is realy better


whtih air motion transformer you can do to directive driver, i think one of the best design is the adam adio x-ART but the model whith the most powerfull magnet, for the high-end speaker

and they have an high medium whitch is a very interesting driver, he goes lower than most other, and they are no to long and vertical directive

they make reference direct radiation drivers for 800hz to 20000hz 


sommovigo's picture

As I said at the beginning of this section, there are practical limitations when measuring so large a loudspeaker. While I am confident that my measurements regime fully characterizes the performance of a small speaker (such as KEF's LS50, which I reviewed last month), with a speaker as large as Wilson's Alexandria XLF, the measurements offer suggestions rather than certainties.

I'm not sure I understand what the implication is here - because it would seem that, by this statement, you could be confident in the response of the speaker above a certain frequency (let's say somewhere above the high pass moving from the bass drivers to the mid/tweet module) - and that those measurements, per the assumptions you derive from your methods ordinarily, would be more of the "certainty" variety and less of the "suggestion" variety.

Would you say that you are cnfident in the response of the speaker above ca: 150Hz?

Michael Fremer's picture

I tend to stay away from the comment section following reviews, mine and those of other reviewers, and this thread is a good reason why. 

The arrogance, stupidity and ignorance is simply appalling and depressing.

I have had many of the world's greatest speakers in my room and I've heard others in other settings: homes, stores and shows around the world.

There are MANY different sounds that are valid and designed for different tastes. The inability of some here to understand that, not to mention understanding how to interpret measurements, is just plain pathetic.

Were I to be led around by measurements, all of which are CRUDE compared to the ear/brain, I'd be listening to CDs... 

This hobby combines science, art and human perception. 

Some of the comments here are sub-human, I'm afraid...

MVBC's picture

Beyond the finished product measurements, drivers parameters can help figure out the style of sound, especially in bass.

For instance, the difference of sound between 18" pro bass drivers JBL 2241 and 2242 can be traced to certain parameters such as BL and moving mass:

JBL 2241  BL 19,  mm 145g, 98dB/w/m versus JBL 2242 BL 24, mm 158g, 99 dB/w/m

Despite a slightly heavier moving mass, but thanks to a more powerful motor, bass from 2242 are much tighter offering better transient response and allowing more freedom in upper frequency cut off while bass through 2241 are more of the rolling type.

Similarly one can compare the same parameters in JBL pro offering versus the Focal drivers used in the Wilson family design:

In 15"

Audiom 15 BL 18, mm 137g, 92 dB/w/m versus JBL 2226 BL 19, mm 98g, 97 dB/w/m

In 12"

Audiom 13 BL 18, mm 108g, 90 dB/w/m versus JBL 2206 BL 18, mm 65g, 95 dB/w/m

Clearly, with a low BL and heavy moving mass, the Focal drivers exhibit quite a low output for such large drivers - a serious problem when trying to recreate live dynamics-, and won't physically deliver faster transients than their JBL pro counterparts, that is clean, lean bass. Other factors such as cone rigidity might help compensate but from the start, I would expect the Focal sound to be plump. Conversely, the 2226 won't go as deep as the Audiom 15, yet its bass will be tighter, punchier. My recent audition of the Focal Grand Utopia confirmed that feeling versus the quickness and tone of the JBL. And here we keep the comparison to bass/upper bass, as cone midranges versus compression would add another level of challenge for the expensive speaker as it did with the Utopia.

So the Wilsons or Focal are of course beautifully crafted, well designed speakers destined to plush interiors and lovers of a certain kind of sound, just as Cadillacs can deliver a certain style of ride. Fine. Yet at $200,000 there is plenty of space for DIY audiophiles to challenge them at more reasonable costs, especially when using active networking designs. Notwithstanding the choice of pro monitoring speakers that are much more affordable than these luxury items. Therefore, some DYI can proudly defend the quality of their bespoke work; however, others could tone down their arguments of authority, checkbook arrogance and quick tongue.

BillK's picture

I have no doubt that a good DIYer could build a speaker with better frequency response, but I'd still like to see the plots and hear what a reviewer had to say about the sound (I've heard any number of components with impeccable response graphs but that sounded simply horrible.)

That of course ignores what price you'd have to sell it for to afford a full-blown factory with staff to produce it in the US, but let's just start with that.

There are any number of "hot rodders" who can build a Porsche-beater for less than the price of a new 911, but they too tend to be one-offs rather than something you can walk into a showroom and purchase.

MVBC's picture

Check the price differential between some Watt/Puppy and a 4348 JBL pro studio monitor and compare the sound... Your answer is there. At the price of this professional gear, even DIY are almost getting not economical.

billyjul's picture

plat frequency response for a driver used in a speaker is just the beginning, but with active crossover witch contain an equalizer, you can correct the response of a speaker, to make better, butt, good driver hame flat frequency response there are other parameters, you have to look, when you make a speaker, parameters, that most audiophile , don't know and a speaker tha measure good on overall parameter can't be a bad speaker, it is not the case for this wilson audio and the jbl mansionned is better than this for much lower price, beaucause jbl know what to do and have developed all the excellent driver to achieve their desgn

NMMark1962's picture

WOW, what a load of TROLL CRAP here.....some people here need to get a life and if you hate Wilson so, then get the hell out of here....you will NOT convince one person with the anti-Wilson spew.....I agree with others....build us your own XLF and prove that Wilson is building garbage or perpetrating a fraud....

When your fabulous speaker is built, let me know so I can buy your wonderful effort and save myself tens of thousands....

To some of you.....quit feeding the idiotic trolls who shit all over these and other forums....i guess that these trolls were run out of the asylums for audio elsewhere...

By the way, for you Wilson haters....I am planing an ultra high end system for late this year...the XLF's are on the list...for the turd here who claims to be able to build an XLF for way less, let me know and I will add it to MY list....hehehehehe...and I bet it will sound oh so sexy.....




ABCDEFG's picture

Perhaps it would be enlightening for some here to consider the genuine economics of this situation.

Mr. Fremer did not pay $200,000 for his XLFs, whatever their cost of construction or subjective worth. In fact, it is very likely that he received a discount considerably greater than Wilson’s 40-45% retail margin.

Considering the dealer cost, it is likely that Mr. Fremer paid less than $100,000 for his pair.

Add a payment plan directly financed by Wilson Audio and a future resale value greater than the accommodation price and the picture of Mr. Fremer’s purchase snaps into focus with remarkable clarity.

Michael Fremer's picture

Is your obtuseness. Guess what? I can buy just about any loudspeaker known to man at an accommodation price. In fact, were I like some reviewers I could get a "long term loan" for just about any speaker known to man and just have them here for as long as I like.

However your libelous comment that there was a "payment plan directly financed by Wilson Audio" is where I tell you with no due respect to go f...k yourself.

The money came from Bank of America, not Dave Wilson or Wilson Audio.

SNI's picture

I would not pay too much attention to the speakers frequency response.
It is a very large speaker, and FR is always measured on the tweeter axis.
Low frequency measurements in smaller rooms are also questionable.
What I do find interesting is the speakers time domain behavior.
This is not very good, and I cannot imagine, that this will not ad a lot of collouration to the sound of this speaker.
It simply emits sound long time after the input signal has stopped.
In my experience this will mask the sound in a way, so that a lot of low level signal is lost.
Also the impedance of this speaker would make me worry.
Anyways I do not think this is a speaker for life, I´d believe that one would get fed up with this "Sound Of Its Own" as time goes by.

If one would like to see a clean time domain behavior, then look at the newly tested Dali Rubicon, that´s how things should behave
, if you want transparancy.


Enter your Stereophile.com username.
Enter the password that accompanies your username.