Crystal Cable Arabesque Minissimo Diamond loudspeaker Measurements

Sidebar 3: Measurements

I used DRA Labs' MLSSA system and a calibrated DPA 4006 microphone to measure the Arabesque Minissimo Diamond's frequency response in the farfield, and an Earthworks QTC-40 for the nearfield and in-room responses.

My estimate of the Minissimo Diamond's voltage sensitivity was 81dB/2.83V/m, which is both well below average and lower than the specified 83.5dB. This speaker needs a relatively powerful amplifier to play at satisfactory levels, but as it has a small-diameter woofer, its user will have to be careful not to overdrive it. However, as the solid trace in fig.1 reveals, it is a very easy load for the partnering amplifier to drive, with an impedance that remains above the specified 8 ohms for much of the audioband, and a minimum magnitude of 7.8 ohms at 177Hz. The electrical phase angle (fig.1, dotted trace) is generally benign; when it is extreme, at two instances in the bass, the magnitude is very high, ameliorating any drive difficulty.


Fig.1 Crystal Minissimo Diamond, electrical impedance (solid) and phase (dashed) (5 ohms/vertical div.).

There are no wrinkles in the impedance traces that would suggest the presence of cabinet-wall resonances, and the unusually shaped enclosure felt inert to a knuckle-rap test. However, I did find a couple of high Quality factor (Q) modes present at narrowly defined places on the enclosure's baffle and top and bottom panels, at 463, 485, and 560Hz.

Visible in the right-hand portion of fig.2 is the Minissimo Diamond's acoustic crossover on the tweeter axis. The woofer (blue trace) hands off to the tweeter (green) at the specified 1.8kHz. The tweeter's average level appears to be a couple of dB lower than that of the woofer, and while the crossover filter slopes appear gentle, the woofer's upper-frequency rolloff is clean. The apparent rise in the woofer's output in the upper bass will be due in part to the nearfield measurement technique, which assumes that the drive-unit is mounted in a baffle that extends to infinity in both planes. The sharply defined notch at 51Hz in the woofer's output indicates that this is the tuning frequency of the port that fires down from the speaker's base, and the port's response (red trace) peaks between 35 and 90Hz. The port's output rolls off smoothly, and though a peak can be seen at 520Hz, this is well down in level.


Fig.2 Crystal Minissimo Diamond, acoustic crossover on tweeter axis, corrected for microphone response, with nearfield responses of woofer (blue) and port (red), plotted in the ratios of their radiating diameters.

Fig.3 shows the Crystal's response averaged across a 30° horizontal window centered on the tweeter axis, spliced at 300Hz to the complex sum of the nearfield woofer and port responses. The response rolls off with the usual fourth-order slope below 80Hz to reach –6dB at the port tuning frequency, and again, the boost in the upper bass is mostly due to the nearfield measurement technique.


Fig.3 Crystal Minissimo Diamond, anechoic response on tweeter axis at 50", averaged across 30° horizontal window and corrected for microphone response, with complex sum of nearfield woofer and port responses plotted below 300Hz.

The tweeter's level is now 4–5dB below that of the speaker in the midrange, which is due not only to the relative levels of the two drive-units seen in fig.2, but also to the tweeter's output becoming increasingly directional above 7kHz (fig.4). Below that frequency, however, the Crystal speaker's dispersion is wide and even—which, all else being equal, correlates with stable, accurate stereo imaging. In the vertical plane (fig.5), the response doesn't change significantly for several degrees below the tweeter axis, which is 36" from the floor, but a suckout in the low treble does develop above the tweeter axis.


Fig.4 Crystal Minissimo Diamond, lateral response family at 50", normalized to response on tweeter axis, from back to front: differences in response 90–5° off axis on large curved side, reference response, differences in response 5–90° off axis on small curved side.


Fig.5 Crystal Minissimo Diamond, vertical response family at 50", normalized to response on tweeter axis, from back to front: differences in response 15–5° above axis, reference response, differences in response 5–15° below axis.

Over the years, I have found that a speaker's spatially averaged response, which integrates the speakers' direct sound with the in-room energy generated by the speakers, correlates relatively well with its perceived balance. I average 20 1/6-octave–smoothed spectra, individually taken for the left and right speakers using SMUGSoftware's FuzzMeasure 3.0 program and a 96kHz sample rate, in a rectangular grid 36" wide by 18" high and centered on the positions of my ears. This mostly eliminates the room acoustic's effects, and in fig.6 I have overlaid the Crystal's response (red trace) with that of the similarly sized KEF LS50 (blue) so that their mid–high-treble outputs are similar in level. The Dutch speaker thus appears to produce too much upper-midrange energy compared with the British speaker; if the music being played allows the listener's ears to latch on to the treble output as its reference, the Minissimo Diamond will sound a little midrange-dominant. But if the music makes the listener take the midrange as being correct in level, the upper octaves will then sound too mellow.


Fig.6 Crystal Minissimo Diamond, spatially averaged, 1/6-octave response in JA's listening room (red); and of KEF LS50 (blue).

In the bass, neither the Crystal nor the KEF have much output apparent below 50Hz, even with the help of the lowest mode in my room, around 32Hz. But the higher-Q tuning of the Minissimo's woofer results in a little too much upper-bass energy in-room compared with the LS50.

One thing I noted while doing these in-room measurements was that while the two Minissimo Diamonds matched very well in the midrange and treble, the left speaker, featured in figs. 2 and 3, had more energy apparent between 25 and 30kHz than the right. It's fair to note that this would have had no effect on my auditioning, as my hearing sensitivity dies above 15kHz.

Turning to the time domain, the Minissimo Diamond's step response on the tweeter axis (fig.7) reveals that while the woofer is connected in positive acoustic polarity, the tweeter's polarity is inverted. However, of greater importance is the fact that the decay of the tweeter's step (the sharp down/up spike at 3.75ms) blends smoothly with the start of the woofer's step. While the Crystal speaker's output isn't time-coincident—possible with a first-order crossover only if the baffle tilts back to align the drive-units' acoustic centers—it is time-coherent. Finally, the Minissimo Diamond's cumulative spectral-decay or waterfall plot (fig.8) is impressively clean, suggesting that the diamond-diaphragm tweeter's first breakup mode doesn't occur until 30kHz or so.


Fig.7 Crystal Minissimo Diamond, step response on tweeter axis at 50" (5ms time window, 30kHz bandwidth).


Fig.8 Crystal Minissimo Diamond, cumulative spectral-decay plot on tweeter axis at 50" (0.15ms risetime).

In many respects, Crystal Cable's Arabesque Minissimo Diamond offers respectable measured performance. But its deliberately tailored treble response is going to make judgments of the speaker's sound quality dependent on the music being played, as I found in my auditioning.—John Atkinson

Crystal Cable
US distributor: Audio Plus Services,
156 Lawrence Paquette Industrial Drive
Champlain, NY 12919
(800) 663-9352

Anton's picture

I like paradoxes, and understand liking "small speakers." And I agree that "big speakers have big problems."

However, later in the review, you tell us that they "sounded larger than they are."

Was that a compliment or a complaint? Good thing, or bad thing?

It also seems the vertical dispersion makes for a narrow vertical sweet spot, "a hollow quality became evident when I stood up."

How do you think you would feel with them as your long term reference? Would that erode your enthusiasm or be a neutral thing? I think that would wear on me, personally.

Anyway, those are beautiful speakers and thank you for the over-all great review!

John Atkinson's picture
Anton wrote:
. . . later in the review, you tell us that they "sounded larger than they are."

Was that a compliment or a complaint? Good thing, or bad thing?

A compliment, as the lack of low and mid bass was, to some extent, compensated by the generous upper bass - the old "LS3/5a trick."

Anton wrote:
How do you think you would feel with them as your long term reference? Would that erode your enthusiasm or be a neutral thing?

If I weren't a reviewer, I could happily live with these speakers. But that "tailored" high end - which was a deliberate design choice, according to the "Manufacturer's Comment" in the November issue - means I couldn't use these speakers for reviewing electronics and amplification.

John Atkinson
Editor, Stereophile

tonykaz's picture

Geez, I love those small monitors, especially the ProAc Tablette ( Richard Gerber r.i.p. ).

But $20K for $2K's worth of performance, hmm.

They are darn sharp look'n, even stunningly beautiful.

I suppose it's ok to have em fixed to the sharp looking stands but I'd still like to have stands to match, my wife's decorator would insist on veto power for stand design and color.

I'm in the Auto Industry and can say that Auto paint can scratch and is a bitch to repair & color match. Auto paint is a deal killer.

Still, the darn things might sound better than the ProAc, I rather doubt it though.

So, on balance, it seems like a $20,000 solution to a $2,000 problem, with the negative bonus of being about $1,000 in ergonometric & design flexibility value.

They might be the perfect thing for those Modern Art Gallery types, found in places like NY,NY ( folks that spend gobs of dinero for down right ugliness )

One more thing; how do they hide the ugly speaker cables?, these things need to be Wi-fi, like Jason's DynAudio ! ( I'd have to insist on that ). And probably run on Li-ion 36v like modern power tools.

Thanks for showing us these things, nice to see how the 1% spend our money!

Tony in Michigan

ps. I can just see some distracted person resting a burning cigarette on the top of these things, thinking they're made of some indestructible material type. ( like an Ikea Quartz counter top which can take burns quite well )

Anton's picture

The same company makes the world's most beautiful cable. If I could afford it, I would buy their cabling based on looks alone! Really perfect fit and finish.

Les's picture

I never complain or am shocked by HiFi prices. But this I find rather galling. I have no issue with the concept of spending such an amount on a pair of speakers. (I would, if I could.) I'm just flummoxed at the thought that these mini monitors (which I have not heard) are imbued with a certain special quality not available elsewhere, particularly when the budget is so generous. Is this the best one could do for $20k, I would wonder... But then again, audio is about a certain special love affair. And this would have to be a very special love affair, indeed.

Anton's picture

A shootout with the KEF LS50's would be interesting.

Xyriut's picture

Agreed since both are rather midrange-forward.

hb72's picture

yes, and no.

impedance of the 2 speakers is VERY different, on very high end with this speakers, and on very low end wiht LS50, necessetating quite different power amps. so that comparision would include mated power amps (and speaker cables for the matter).

Anon2's picture

$19,995/$6,000 = 3.3325 pairs of B&W 805 D3s (and a big mark up from $5,000 D2s to be sure). 2.499 is the ratio for the new Kef Reference 1. 9.22 is the ratio for a kit--having nominally the same components--and adding $1,000 and elbow grease for making your own (albeit non-curved) enclosure.

And, we should note, that on a major used equipment site, the same outlay might get you a used pair, on a good day, of some Wilson Sashas or 2-3 pairs of B&W 802s.

I don't want to make this a feeding frenzy but I, like others, have to question what one can get for $19,995, or a whole lot less.

Staxguy's picture


[9:1] If you are into kits, are you into PBN Audio? Pennywise.

If you are into ratios, do you like Polymer Audio Research? Polymer Logic [$24,970]

The later had a good advertisement illustrating driver cost (retail) vs speaker cost (retail) and ratio vs. Marten Design Coletrane Supreme ($250K) - diamond mid.

Just saying.

Staxguy's picture

Well my thoughts / opinion on the matter is slightly different. I don't like the diamond-like image tweeter cover, but assume it is for protecting the tweeter.

I think that they spent to develop a diamond tweeter, and included silver wiring (like Kondo / AudioNote Ongaku...) and were able to develop it and price it for $20,000 / unit.

I don't like the colour as well, but oh well, I loved the look of the previous version - maybe I'm in a bad mood? :)

It's probably that tweeter - cover - which looks good, on it's own.

In terms of price, I was thinking after having already read the previous article, why I didn't buy the Ayre MX-R @ $10,000.00 / pair and if I would get the Bel Canto REF 600M - which uses the wonderfully cheap / great? - nCore - 300W modules @ $5,000.00 / pair.

300 watts is a lot of power, for most (wattage) folk, 200W being "American" average muscle amp @ 8 ohm, so 300 might (note the look of the 600M - back) be good for cheap speakers. At least what I have here.

Reading the review, I was interested in reading a description of how this version of the speaker compared to the previous.

It's amazing that the previous included a Berylium tweeter - love the Grande Focal Utopia III EM and baby Scala Utopia III - also can only stand the sound of Focal Twin 6 be as a minimum standard in near-field recording monitors you'll hear a pro-music store room.

Didn't like the sound of the B&W 802 Diamond trebble (though it was incredibly detailed and fantastic, in terms of its delivery to the brain) - so that if the new diamond tweeter here is better than the berylium one, previously, it's interesting to me.

I also like small nice, speakers. I guess because it's a whimsy - a not needed purchase, so price is not an egregious issue.

Reading the review, I was thinking why the reviewer did not include how you get two nice speakers for the price of a gambler's ring. ;)

Well, really, to get like a little micro-pave ring today by Cartier for a lady, it's about $40,000, tax-in.

Actually, as a major interest, was reading how the reviewer liked good music and musicians - Fred Hirsch and Robert Silverman.

If you've heard these guys live, it would be majorly interesting to see if you liked the price of the Crystal Cable Arabesque MD.

It's amazing how little piano you can get for $20,000. A poor stand-up, maybe. A good wind instrument, mind you. A pair of Selmer Clarinets. Reeds will set you back, further. ;)

Now given the name, I have ballet on my mind, so perhaps this is why I have such a positive impression as to the musical taste of the reviewer.

My I wanted to hear more on the sound and the technology.

The one commend that I didn't get was the tailored upper-end response - I liked that it went to 50 KHz and was - like a pro-monitor - spec'd to - 3 dB (+/- 1.5 dB), and only so deficient in the lower bass range.

I tried to understand the reviewers psychology at this point, wonder if he was trying to justify not purchasing a pair.

The Silver Signature - a reviewer's favourite - was $8,000.00 a pair - the same price as a Selmer Bass Clarinet - great sound - today.

$8000.00 in 1995 at 5% interest for 21 years (to today) is $22,287.20 USD.

So perhaps this is relatively relevant / poignant.

The Silver Signature was an essentially beautiful loudspeaker with a response of 100Hz-15kHz ±1.5dB and an efficiency of 88 dB.

Sorry for jumping.

YukonCornelius's picture

This may be inappropriate bias, but I cannot wrap my mind around a 20k loudspeaker with a laminated paper cone. I know a bunch of people like them and I probably couldn't tell the difference if blinded, but couldn't they use something like carbon fiber? Fake it for us a little.

Anon2's picture

Paper cones are finding their way back into many expensive speakers.

I have noticed that even Wilson Audio has discreetly changed its drivers from metallic materials back towards paper.

Scan Speak, of which this speaker sports a mid-range/woofer, has many drivers, for kits and OEM applications, made of paper. Indeed one of Scan Speak's paper-based mid-woofers is reputed to be among the best in class.

My question is: what is the long-term durability of all speaker cone materials? We have not seen a lot of discussion devoted to this topic in this (or other) publications.

How long does a speaker driver last? Perhaps it's all subject to the butyl-rubber surround which, fortunately, has replaced the disintegrating foam of yesteryear.

We would like to know what the comparative durability, long-term integrity of speaker driver materials may be. Aluminum, Kevlar, fiberglass, paper, carbon fiber, plastics, other composites: how long do these materials last?

Audio equipment is expensive. Some speakers have essentially the same driver materials in the most and least expensive versions of a speaker line for a manufacturer. It would be valuable to know how long a speaker will last before its driver material begins to fail.

It would seem to me that paper, from a lay person's experience with this material, might be more susceptible to longer-term failure due to changes in home humidity than, perhaps, would a material like aluminum. Another question is how long woven-fiber drivers last until the weave starts to unravel.

It is great that so many materials have found application in sound reproduction. Some comparative assessment on long-term (over 10 year) durability would be welcome.

YukonCornelius's picture

I like some of your points. While I'm not an engineer I think we can assume a composite paper cone is, even if soaked in pitch, likely to be less durable than a properly woven cf cone in the long run. So, it is my humble opinion that this paper cone thing is a shift towards saving money and perhaps absolutely not at the expense of sound.

My personal opinion is that for 20k those drivers better be able to be passed to my kids. :)

dumbo's picture

Come on JA, your killing me here :)

Why does there seem to be increasingly selective use of the plastic tape accelerometer measurements for speaker cabinet performance? You show them on some (in the same issue nonetheless) but not on others?

I was gravely disappointed to see this measurement left out of the new B&W802 D3 review also.

What gives?

John Atkinson's picture
dumbo wrote:
Why does there seem to be increasingly selective use of the plastic tape accelerometer measurements for speaker cabinet performance? You show them on some (in the same issue nonetheless) but not on others?

When I started doing the cabinet vibration tests a quarter century ago, I designed and built a preamplifier with a very high input impedance to amplify the output of the accelerometer. That preamp has become increasingly unreliable and sometimes stops working just as I need it. I have been trying to repair it but the real solution is to build a new one.

John Atkinson
Editor, Stereophile

dumbo's picture

Thanks for the reply JA. No harm intended. For some reason I've become fascinated with seeing how serious these various manufacturers are when it comes to such things.

Marc210's picture

I wonder who's gonna buy that. But there are beautiful WBT Next Gen silver binding posts at the rear.

funambulistic's picture

And these speakers certainly are. Now for the price (from my humble perspective): $20K is play money for some people (I do not begrudge anyone that can afford them one bit) but, from my decidedly middle class perspective, is way over the top, from what they can do (according to the nice review and measurements). If they are truly performing the "LS3/5a trick" then good for them - I can mention many manufacturers that employ this (our favorite KEF model, for example) for a lot less.

The most I have ever paid (not really - I won them from a competitor's website) was $7,500 for a pair of Wilson CUB 2's back in '98. These things were monsters for stand mounted speakers, both figuratively and literally and, at the time, were Wilson's least expensive offering (supplanted now by the +/- $20K Duette). I loved their build, quality, heft, MTM arrangement, but there was always something off. I did countless set-ups, measurements, triangulations, etc. but they never satisfied. I prematurely came to the conclusion it was my sub-par equipment (Rega Planet CD player, McCormack TLC-1 preamp, Parasound amp - all favorably reviewed by this magazine). In actuality, the CUB was too (add bold or italics) accurate and, though quite dynamic when called upon, were just too sterile for my tastes. I eventually sold them for a fraction of their selling price (which was no loss to me as they were free to begin with, but there was a price to pay as the publisher of the other online mag sold out our email addresses and I am still getting spam to this day - first world problems).

TL/DR: I would rather have a speaker with some "musicality" (as someone mentioned, ProAc [which my current speakers, though not ProAc do superbly]) than an ultra-precise, warts-and-all monitor (leave those for the studios and reviewers). At the same time, I would expect $20K monitors to give one EXACTLY what was intended by the musician/engineer, etc. (though what that is is anyone's guess, unless one is involved with the process, which Mr. Atkinson has been, more than the 1% we talk about).

TL really DR: If I had the funds for a $20K speaker, they would run the entire audible spectrum (thundering bass, rich midrange, extended - but not clinical - treble). Wait, that is exactly what have now, for fraction of the price, with my two monitors and a well placed sub. As always, a great review John, and I await the next one!