Mojo Mystique X SE D/A processor

When I survey the realms of fancy-pants audio, the first thing I notice are cohorts of luxury-brand manufacturers selling pride of ownership with emblematic faceplates. After that, I spot another type of manufacturer, one that mocks the first type and aims its products at a younger, more working-class demographic, seducing potential customers with how much "truth" they are offering for only $15. But sometimes, when I look beyond the full-page ads and big rooms at audio shows, I discover a rogue manufacturer that is peddling a very specific type of listening experience, which they believe is the best. A listening experience only they could have created. I am grateful for manufacturers like this. They make my job more interesting, and I admire them for their courage in betting on their own taste in music reproduction.

I am relating these observations because this month I'm reviewing a digital converter from an off-the-paved-road audio manufacturer named Benjamin Zwickel. He operates a company called Mojo Audio, which is located in Albuquerque, New Mexico. When Editor Jim Austin asked if I was interested in reviewing Mojo Audio's Mystique X SE DAC, I told him I had never heard of Mojo Audio, but that I would check out their website. Halfway through the first website page, I realized Mr. Zwickel was one of those too-rare manufacturers with a strong viewpoint, one that inspires him to create products designed with a specific type of listening experience in mind. Zwickel explores an engineering strategy he believes is truer than others to the data in our digital files.

His well-stated beliefs reeled me in with surprising ease. I told Jim I was curious to see how Zwickel's "unusual" approach to DAC design would work in my outlier sound system.

The Mojo's mystique
Mojo Audio's $9999 Mystique X SE digital-to-analog converter is an evolution of the company's earlier series of DACs, the EVO series. Like those earlier DACs, the Mystique is what Zwickel describes as "purist design" because it only processes PCM—no DSD or MQA—and only does D/A conversion. According to the Mojo website, the Mystique X features an R-2R circuit that uses a pair of 20-bit Analog Devices AD1862 ladder DAC chips. "There is no pre-digital filtering, digital noise shaping, upsampling, oversampling, or error correction. Your DAC will decode up to 24/192 but will truncate 24 to 20-bit." (footnote 1) The SE version of the Mystique X employs a "massive" power supply that uses Lundahl amorphous-core chokes (one for + and one for –) feeding five independent secondary voltages, one for each stage in the DAC circuit. Additionally, nine ultralow-noise, ultrahigh-dynamics Belleson SPX discrete regulators are used to isolate power to the clock, DAC chips, and the Staccato op-amps used for its direct-coupled analog output.

After unpacking the $9999 Mystique X SE, I held its 19lb, 9" W × 4" H × 16" D extruded aluminum chassis on my lap, turning it around, absorbing its dark minimalist aesthetic. It looked like the DAC I'd been wishing for: a simple heavy box I can just plug in and forget about. No brightly lit display. No menu. No setup. No app for my iPad. No software updates. No plastic remote. No PCM filter choices. No external clocks. No DDCs. And! Only three digital inputs: a galvanically isolated asynchronous XMOS-compatible USB-B, AES3, and coaxial (S/PDIF). And two analog outputs: single-ended (RCA) and balanced (XLR). Also on the Mystique's back panel, beside the mains power switch and IEC socket, there is a sliding DC ground-lift switch.

The Mojo's distinctively styled black front panel looks artfully rugged with its baked-polymer finish and engraved insignia. Above the faceplate-filling "Mystique X" lettering sits three barely noticeable little buttons, each with its own tiny blue LEDs, used to select one of the DAC's three digital inputs. A fourth, red LED above these indicates S/PDIF or AES3 input errors.

A few days after I received the Mystique X SE DAC, Mr. Zwickel sent me an email with a link to his blog and a pdf of a lecture he gave at AXPONA 2022 entitled "The Seven Myths of Digital Audio Dispelled." (footnote 2) In that email, Zwickel also gave me what he called a "small glimpse" into the design strategies employed in the SE version of the Mystique X DAC I was reviewing.

"What makes Mojo Audio DACs so unique is this: We use the best of 100-year-old, 40-year-old, and modern technologies. The power supplies we use are LC choke-input, the largest, heaviest, most expensive, and least efficient type of power supply possible. It was well known over 100 years ago that this was the best power supply, the only type of power supply to provide instantaneous current, and the only type of power supply to be able to yield correct time and tune of music. Any other power supply typology is engineered to lower cost, reduce size and weight, and increase efficiency. Not to improve performance.

"As for the 40-year-old technology, that would be the vintage R-2R DAC chips we use. We're far from the only company who is using these vintage chips, but don't mistake true R-2R for the modern discrete-segmented R-2R." Here, Zwickel named several well-respected DAC manufacturers (footnote 3). "They use multiple 8-bit discrete R-2R ladders and then use a FPGA to weigh and combine the ladders to achieve higher bit depths and better linearity. Sort of like a combination of true R-2R and Delta-Sigma technologies. And that is what they sound like: a cross between true R-2R and Delta-Sigma.

"As far as modern technology, we search the world for the best in modules and components, such as Vishay's best TX2575 'Nude' resistors, SiC Silicon Carbide zero-recovery Schottky diodes, Belleson discrete regulators, and JL Sounds XMOS USB input modules. Our chassis may look like a simple black box but it is anything but. Aside from the massive aluminum construction, we use a baked-on polymerized finish, which lowers mechanical resonance by about 11% over the anodized finishes most companies use, as well as improving durability. We use an Alodine undercoating, which lowers RFI and improves finish cohesion even more. We use all nonmagnetic, non-rusting stainless steel hardware."

Zwickel concluded, "The most important thing is to be open minded and to always evaluate components in the system they will be played in using the digital source and software that will be feeding it. ... Do some blind A/B comparisons with friends, where the listener does not know what they are hearing."

The $9999 "SE" version of the standard $7999 Mystique X adds ultrafast, ultralow-noise, zero-recovery SiC Schottky rectification diodes and ups the capacitance of the four-pole Mundorf capacitors to 22,000µF; it also adds Lundahl amorphous-core input transformers on the S/PDIF and AES3 inputs.

Mojo sells its products direct to the end user and offers a 45-day, no-risk audition and a five-year transferrable warranty.

Footnote 1: See I think what Zwickel is referring to is that regardless of the number of bits in the signal path, the real-world "resolution" of the very best DACs is in the 20-bit range. JA's measurements have found that a few do slightly better (and some do considerably worse; see JA's measurements in this review), but 21 bits seems to be the current upper limit. That's a real-world limitation that's difficult to overcome—and maybe pointless, since, as Zwickel also notes in that essay, when you consider not just the whole signal path but also the listening room, the true noisefloor is much higher. The only DAC I recall that Stereophile has reviewed that intentionally truncates to 20 bits is the Schiit Yggdrasil—thanks to Technical Editor John Atkinson for reminding me of that. I'm aware of a few others that use vintage chips and so truncate the datastream to 16 bits.—Jim Austin

Footnote 2: See

Footnote 3: We decided not to list the companies Zwickel named because they will not have an opportunity to respond. Still, I respect his opinionated candor.—Jim Austin

Glotz's picture

To seek out 'rogue manufacturers" specifically. I'd really like to hear Holo Audio gear if they have it.

If I could afford $10k for a DAC, this would be near the top, on parts quality alone. Thanks a ton for your report and your transparent perspective.

georgehifi's picture
"The Mystique X SE produced a unique, sophisticated listening experience that presented digital recordings as beautiful, probing, and engaging."

Same happens to me every time I listen to a "good R2R dac", ya just can't beat em, even if "some" of JA's measurements are disappointing, as is the cost, there is something that just fundamentally "right" with R2R Ladder doing PCM conversion, that Delta Sigma misses out on.

Cheers George

RH's picture

Could have bet the farm on the "disappointing" measurements.

This seems to be a pretty reliable trend one can see in JA's measurements of gear, especially DACs, Amps etc. Whenever a design seems to go some esoteric direction, eschewing general "best current practices" and telling some story about a technical advantage for doing such technical advantage seems to show up in the measurements. They usually produce irregularities that tend to underline why most components aren't designed that way.

I know the usual refrain is "but the proof is in the listening, how it sounds!" But, again, audio gear always comes with some compelling technical story about an engineering issue "solved" by the design, so one could at least expect for measurements to show these technical "solutions" solve what they purport to solve. Rarely seems to happen.

FredisDead's picture

Art Dudley loved the Abbingdon Music Research DP777 and JA said it measured poorly. Fifteen years before that review in the mid-90's there was an uber-expensive YBA cdp that was loved subjectively and measured poorly. It made the cover of the magazine. Measurements are useful. They will never tell you how much you like the sound. There are a ton of measurements you can make with cars and with wine. They will never tell you if you like the way the car drives, feels, or looks or if you will find the wine interesting. Isn't this a damned tiring conversation? JA does his job, the reviewer does his/hers and it is up to the consumer to decide if the measurements or the opinions of the reviewer are paramount.

RH's picture

"Measurements are useful. They will never tell you how much you like the sound."

Actually...they can.

Depending on what we mean. Measurements can, for instance, tell you whether you will actually hear what is being measured, including whether it is at all plausible you will hear a sonic difference between X and Y. (That's what engineering is for..)

Of course when it comes to pure psychological phenomena, sure you can always "like" X more than Y. Some people are interested in knowing whether that is due to audible differences, or whether it's more likely due to psychological bias in the listener. That can inform purchases...for those who care about such things.

Not to mention, certainly measurements can tell someone quite a bit about the sonic character of, say, a speaker. Insofar as someone is familiar with the sonic consequences of measurements, and sees, say a 4 db rising trend around 2k, it can alert them to a design they likely won't like, if they don't like exaggerated highs.

So, while sure a lot of people won't be able to predict fairly well from some measurements what they will dislike, or like, that can as much be due to inexperience or ignorance in regards to measurements. More experienced people will be able to do so.

"Isn't this a damned tiring conversation?"

No. Not for people interested in how measurements correspond to audible consequences. Why would that ever be "tired?" One would think it quite important to the project of high end audio - that's pretty much which drives audio manufacturers, and every single bit of high end audio gear comes with a technical story BECAUSE they know how that appeals to audiophiles. Why would it ever go away?

"JA does his job, the reviewer does his/hers and it is up to the consumer to decide if the measurements or the opinions of the reviewer are paramount."

Yes, of course.

But if someone isn't interested in measurements...why complain about someone who IS? Go one's separate way, if that's the case.

ok's picture

..which is a mathematically based format, measurements are of paramount importance; the question in question is which ones and why :-)

georgehifi's picture

"Measurements are useful. They will never tell you how much you like the sound."
RH: Actually...they can.

Yes totally agree, of course they can, anyone that says they are "useless" either doesn't have a clue how to read them, or are "useless" voodooist's themselves.

Cheers George

RH's picture


Over the years I've found that when people (especially those who fall for pseudo-science-type ideas) defend their beliefs with:

"We don't yet know enough about X.."

It usually means:

"I don't know much about X..."

They are often projecting their own ignorance on to science or engineering.

I think unfortunately this is very often seen among audiophiles because it's a sort of perfect storm: Audiophiles tend to have some level of gadget and techno-lust to begin with, an interest in the gear, and that's why a high end company's offerings always come with a technically enticing story of "Here's a technical problem that will affect sound quality and Here's How Our Engineering Solves This Problem!" So there are all sorts of "technical sounding claims" thrown out that strike the audiophile as plausible sounding, but many if not most audiophiles lack the technical expertise to properly vet the claims, and recognize b.s.

(And I include myself in there. I just try to stay aware of what I don't know).

georgehifi's picture

To me it's more the "different ways" of PCM conversion that's the main factor here, as with any decent "R2R" dac I hear when compared to "Delta Sigma (DS)".
One being "bit perfect" (R2R) doing PCM, the other being a "facsimile" (DS) doing PCM.
Different story when you convert SACD or DSD music then the shoes on the other foot and Delta Sigma is the one to go for.

Cheers George

Archimago's picture

Clearly the measurements tell us that this is not a high-fidelity DAC in the 21st Century compared to many technically superior products that cost way less than $10k. No surprise since the AD1862 DAC came out in 1990 and we're seeing here an implementation without proper filtering, and poor jitter performance.

That HR likes this kind of sound and finds that he's "reeled in" with unusual claims of why this is "good" based on Zwickel's comments is his subjective right to an opinion (but just an opinion) that says more about the man than the product.

I'm sure many music lovers will enjoy the sound since distortion is still low and linearity is "adequate". This is not impressive for "high fidelity" lovers though. Nonetheless, seriously folks... $10,000 is a lot of "0"s for such a low level of performance!

RH's picture

Yes, I certainly have no problem with anyone who wants to buy an item like this. And perhaps the colorations edge in to the audible and some might find it pleasant.

But high end audio gear is very rarely sold to the audiophile along the lines "Buy our gear, it adds pleasant distortion in to the chain." Rather, it's usually some technical story about how the efforts have gone in to *lowering* distortion and getting a "more pure path to the music." All that seems at least implied in the technical brief for this product which mentions more expensive and hence "higher performance" parts. But, again, we don't see anything measurable suggesting the design has led to "higher performance" or lowering of distortion. Just the opposite.

John Atkinson's picture
Archimago wrote:
Clearly the measurements tell us that this is not a high-fidelity DAC in the 21st Century compared to many technically superior products that cost way less than $10k. No surprise since the AD1862 DAC came out in 1990 and we're seeing here an implementation without proper filtering, and poor jitter performance.

Because of the conflict between the Mojo Mystique's measured performance and its sound quality, Editor Jim Austin asked me to write a follow-up review based on my own auditioning to be published in the May issue. That follow-up, which includes additional measurements, will be appended to this website reprint once the May issue has hit mailboxes and newsstands.

John Atkinson
Technical Editor, Stereophile

Archimago's picture

You've already done the leg work to show what this DAC is doing.

Whether you like or don't like the sound from this DAC auditioned in your room really doesn't matter. I don't mind the sound of simple NOS TDA1543 DACs either but it kind of depends on the situation (including the gear and room set-up), and what I'm after from the technology.

For this DAC, frequency response with -1 or -2dB into 20kHz isn't a huge deal and might make harsh material less objectionable. Low level truncation with 24-bit material down around -100dB, linear down to something like -75dB (<1dB deviation) is also fine compared to other sources that audiophiles like (eg. vinyl). J-test looks bad but I don't think those who test out jitter anomalies would really hear much of an issue.

It's all more just a reflection of the limits of human hearing not having the resolution or precision of the test instruments that you've been able to show.

I think RH is right. Instead of always companies promoting devices as if the product is technically better because of whatever story they have, it's IMO more honest to market it to audiophiles who want a different type of coloration that they might prefer. For example, in the headphone world, the Harman curve might be suggested as a preferred EQ but many people (including myself) might not have such a preference for the bass response. Using the measured data/comparison data, we can choose for ourselves.

IMO, a reviewer's subjective audition descriptions are not that useful. Could be entertaining hearing stories, and music suggestions are good, but not sure these days if subjective opinions all that meaningful - especially for those with any kind of technical inclination and the plethora of opinions on forums and social media/YouTube if one is looking for sentiment.

jack_lint_1984's picture

It seems I'm forced to state the obvious, which is never a good thing - Stereophile provides reviews + measurements. Reviews and measurements.

What gives you the power to speak for everyone and suggest that the world is ready to do away with the reviews?

Who are you? Besides someone who appears to be confused about their influence.

RH's picture


We differ somewhat I believe on the usefulness of subjective reviews.

"IMO, a reviewer's subjective audition descriptions are not that useful....(snip).... - especially for those with any kind of technical inclination

I think that last sentence is very significant. Many, if not most audiophiles, I would argue, are not technically inclined. While we are attracted to shiny toys, it's very often a consumer-level interest, without a deep familiarity with the technical aspects.

I think the portion of audiophiles who are engineer-oriented, by trade or by proclivity, and who geek out on measurements and doing all sorts of hands-on DIY or truly diving in to deep knowledge, are a relatively small portion of the hobby. Many of us are glad there ARE engineering nerds around to do this...but it ain't our bag.

Speakers for instance vary hugely in the ways each design can sound different - resonances, dispersion characteristics, cross-over characteristics, sealed, ported and on and on. For someone to be able to look at a set of measurements, especially with speakers, and just know exactly how a speaker will sound takes some serious personal experience in correlating measurements to their audible consequences. Many audiophiles are neither predisposed nor interested in that level of commitment to the technical side.

That's why the engineering types who ARE in to that stuff come in handy. Someone not technically proficient may look at graphs and say " does this mean in terms of How It Sounds?"

This means that the people who ARE technically proficient have to be able to cross that bridge - to be able to interpret the measurements and then describe the character of the sound for the non-technical audience. So in this way the subjective aspect - someone describing the sound of a piece of gear - is inescapable in any practical sense.

And it follows that it will fall to the skill or talent of the technical person in putting sound in words. And here rises a sort of dichotomy: Generally speaking, I have found the people MOST inclined towards engineering/measuring are often the LEAST inclined or interested in detailed sonic descriptions. "Too squishy and imprecise, just tell me about the design and measurements!" To the degree they live in the world where measurements tell them what THEY need to know, they may not have needed to develop a skill of putting sound in to words.

It is rare to find someone who is both technically inclined AND who is a wordsmith able to put sound in to words for the layman audience.

(John Atkinson is one such treasure, correlating measurements to language).

If there is a resonance in the measurements at 100Hz, we want to know the audible consequences. The engineer translating to the layman listens and tells us "it had X effect on male vocals" or "while it looks bad, in practice this resonance seemed to be rarely audible across a range of music." So it still means the layperson is relying on a subjective assessment.

But what this also means is that, since "what it sounds like" is what counts, a perceptive listener can hear "what it sounds like" and if they are good with words they can describe "what it sounds like." If that's the case, in a sense you can skip past the measurements right to "what it sounds like playing music" with a good subjective description.

So while I agree measurements can be very informative, and personally the ideal for me is a subjective review accompanies by measurements to correlate between them...I find I can often get quite a good sense of the character of gear by triangulating the subjective impressions of reviews who I've found accurate before, among audiophiles, etc. I think a combo of both can mean one can winnow subjective reviews of the voodoo aspect.

For someone who is more inclined to geek-out to the point of wanting to be independent of any subjective reviews, I totally understand that too. We all have our interests and ways we care to spend out time.


teched58's picture

John- Is Stereophile planning to publish a manufacturer's response?

I would be interested to hear Mojo describe what they believe is the Mystique's value proposition.

John Atkinson's picture
teched58 wrote:
Is Stereophile planning to publish a manufacturer's response?;

Mojo Audio's response to the review was published in the May issue of Stereophile and is now included with this reprint of the review:

John Atkinson
Technical Editor, Stereophile

teched58's picture
JA1 wrote:

Mojo Audio's response to the review was published in the May issue of Stereophile and is now included with this reprint of the review:

Welp, you gave Ben Zwickel the opportunity to respond. It's not on SP that he didn't use the space for anything substantive.

John Atkinson's picture
My Follow-Up review is now appended to the review:

John Atkinson
Technical Editor, Stereophile

georgehifi's picture

Yeah the $10k price is over the top, I had a dig at it in my first post. Looking at the parts count and case of the inside shot, at best around $500-$700 worth, even with the Lundahl 160mA chokes.

Cheers George

Long-time listener's picture

"[The Mojo Mystique's] nuclear-level pheromone emissions led me amorously from one streaming track to another."

Yes, low resolution and dodgy measured performance will do that. Low resolution leaves more to the imagination, and our romantic imaginings take over. Maybe we should have a "low-resolution" setting on our DACs to help take the edge off of those overly sharp or overly digital-sounding recordings. But I'm not going to pay $10,000 for this one. Sheesh.

FredisDead's picture

I own a Thorens TD124 and a Garrard 301. On both I have Reed 3P arms and expensive MC cartridges. An under $1,000 Pro-ject with an Ortofon Blue is likely to measure better than either vintage deck and yet there are few who would listen to them and prefer the Pro-ject. This is a purposeful-on my part-reference to a type of audio component that does not lend itself to much in the way of measurements. To this day no one has come up with a way to measure how a tonearm or a cartridge will sound. Measurements as to speed accuracy, wow, flutter, and rumble don't tell one much about how well built a turntable drive system is or even how it will sound (barring significant deficits).
Those that insist that modern DAC chips are inherently superior to so-called "vintage chips" misunderstand how the chip operates and it's overall contribution to the analogue signal. They don't realize for example that modern DAC chips are designed and built to be useful for hundreds of products and not just audio components and that only a small portion of the DAC chips operation capabilities are utilized in an audio DAC. They don't understand that other factors such as the all-important output stage affect the ultimate sound.
I doubt JA will respond to this question but I think of him when I ask, do any of you question the knowledge and expertise of Gordon Rankin (Wavelength Audio). Look up an interview he gave at RMAF 2018 and listen to what he has to say about measurements vs. sound. Gordon is no longer making and selling DAC's but when he did, he did not utilize the "latest greatest" DAC chips and he did utilize, "gasp" tubes. His amplifiers are built based upon, "gasp" tubes.

JHL's picture

I like your remark because it allows for an explanation (assuming one is even needed given the focus and purpose of fine audio, which is connecting to music). It allows - it is *objective* - for information to yet enter the analysis.

Unfortunately analysis is where measurist belief goes wrong, ironically, and it goes wrong at least three ways I can think of. Let's repeat that: Measurism is a system of belief overlaid on audio technology that expects to find or even define correlation, pertinence, and linear progress toward a better goal.

I can't say I've found it to do that. In fact, it seems measurism goes wrong in that it can't establish a reliable, relatable pertinence between data and *sound*. At best it seems to find a linkage between *a* datum and *a* somewhat, somehow involved, ostensibly-related phenomenon. Fortunately that's expected in the sciences because an observation is not a phenomenon nor is it an understanding. It's a detached viewpoint tool whose relationship to the whole is incomplete and needs an interpretation and a determined relationship.

Secondly, as if in sympathy with this central problem, measurism tends to engage a roster of supporting fallacies. There are some in this thread, some circular and some simultaneously declaratory and open-ended at the same time - the begged question. They generally go: if measurement > then definitely sound.

And third, measurism will invariably encounter technical, data-side errors in methodology, or rather, differences in methodology that further obscure their purported purpose. There is no perfect measurement just as there are no hard and fast relationships between its data and sound. Granted, they're generally useful but since they have to be interpreted, when they differ so too will the conclusions made about them.

All together the measurist may want to reexamine his or her objectivity. While data is an invaluable tool and the cornerstone of research and technology, to extend that to sound ipso facto is a misaligned approach to better sound, and measurism has had almost no pertinence at all to the very best sound that I can see.

At worst is may threaten fine, musical audio if it overwrites sound itself because the data is said, subjectively, to speak for it. That is, 'I know it sounds right because of how it measures', an obvious and contradictory belief and not at all a science: 'Music sounds like this because the measurement says it shall.'

No it doesn't and no it may not.

The very best sound - this being Stereophile and not another earless journal and publication - is an applied science elevated to a fine art overlaid onto a flawed medium. It is most essentially the arranging and sorting of many interrelated technical phenomena. How or whether it obtains to a data for the sake *of* that data is virtually irrelevant.

Archimago's picture

That's a lot of concerns which IMO has no basis in reality on many levels and promotes levels of unnecessary anxiety.

The "measurist" (objectivist) perspective has evolved and been tested over time based on engineering and the science behind the products we enjoy these days. Subjective opinions based on what any individual likes simply will not move the field forward, nor will improve value to consumers (maybe some audiophiles don't care about this if $10,000 DACs like this is seen as reasonably priced?).

Unless one gathers data, there is little knowledge or wisdom accumulated based on meaningful foundations. Nothing wrong with expressing subjective preference of course... But it's one voice or maybe even a handful of like-minded opinions with the associated complexities and idiosyncrasies that might or might not apply to others.

JHL's picture

What I said is demonstrably true and fact-based: Science is abstract until it's applied - hence *applied science* - and that's where the dogmatic insistence that if it's a measured abstract it's also a broad-based reality proxy fails. That's too often the core of measurist audio.

Whether the measurist perspective 'has evolved and been tested over time' is, as a common retail system of belief, independent from exactly how it's 'based on engineering and the science behind the products we enjoy these days' until direct, reliable, and unambiguous pertinence is established. Make a clear, scientific distinction between isolated data and complex effect. I am.

Note that nobody's claiming that technology shouldn't be tested and note that I fully accept highly engineered audio. One of the pillars of measurist dogma is that the other guy is a "science-denier" and therefore objectivism's automatic lesser. The other is that objectivism simply citing The Science is authoritative enough a stance and linguistic display that resistance or perceiving opposition to it lumps an interlocutor into that science-denier category.

That's not true; a third way rightly questions the technical shorthand wherein measurism actually impairs better-sounding audio. That's why I disagree with the liberties taken far short of a necessary level of technical understanding that cite objectivity and science anyway. They're not objective and they're not scientific. Obviously; we consistently hear around them. See: the high end reviewer and his readers.

'Subjective opinions based on what an individual likes simply *does* move the field forward' insofar as a listening review is pertinent and the listener may determine a finding. It certainly does and has 'improved value to consumers', demonstrated as continuing strides into the more musically-faithful and rewarding component categories that despite the objectivist clamor over sciency $499 commodity boxes that have magically solved for all acceptable fidelity all the time, seem to refuse to die among real listeners.

You say that data, knowledge, and even wisdom pertain to meaningful foundations without defining what's meaningful. I say it's that better musical experience, however to much of measurism the goal is quite arguably the data itself, as I said. The fidelity of reproduced sound is claimed to flow from it and not from the gauge of original sound.

You'll have to nail down those opinions, Archimago, and not project them into the space as ambiguous placeholders of a general faith in a general science. Remarks like that are by now virtually tropological while your particular lede is admittedly opinionated; it only states a contradictory opinion. You see hand-wringing, which is common among the assumptions freighted with measurist audio reductionism, where I see its inevitable defensive hand-waving.

Having established that the sound at the ear is the only legitimate gauge, let's get into a real pertinence and meaning. There's just no reliable architecture tying a point of data into broad musical fidelity. That doesn't mean they do not or may not correlate; we can safely say they do correlate. But we can't continue to make such wildly broad-brushed or unmoored opinions part of any reliable syllabus that speaks comprehensibly to both entities.

Glotz's picture

More pretend from those that think measurements tell them everything about sound.

More rigidity, really, looking for validation.

As parts-quality goes up, the 'immeasurables' become more audible.

RH's picture

More rigidity, really, looking for validation.


"As parts-quality goes up, the 'immeasurables' become more audible."

How *sure* are you about that?

Glotz's picture

More sure than measurements being an indicator of 'quality sound'.

I would point to hundreds of successful audio components that follow that mantra.

PS- JHL consistently writes the most brilliant and true arguments in these pages.

RH's picture

"More sure than measurements being an indicator of 'quality sound'."

So are you just as rigid in that belief than the measurements-oriented folks?

How open are you to being wrong that measurements can't tell us how things sound?

How rigid are you on your belief that "As parts-quality goes up, the 'immeasurables' become more audible." And if you are appealing to "unmeasurable" how could someone convince you that you're incorrect?

Glotz's picture

Look at the Benchmark AHB-2 amp. There are tons of applications where the unit sounds hard and unforgiving, where other, just as a well-designed amplifiers sound and measure as accurately, sound sublime. It obviously measures among the best.

I own the HPA-4 preamp / headphone amp and think it sounds great (and measures among the best)! That being said, I would drop it like a hot potato for a preamp from, say, Holo Audio Serene with Kitsune tuning. There are other far more expensive designs, both solid state and tubed, where I would replace this very fine preamp.

I want more than accuracy in my playback- I want tailored improvements in depth perspective, reverb trails and added sparkle (or perhaps refinement) that tubes and other solid state designs that do add to the sound from a willful decision. (And yes, I am satisfied currently.)

Perhaps Lab-12 preamps and DACs are more my speed. Tube amplification can justify their existence while raising their refined noses at accuracy-focused, solid-state designs.

I feel that very high parts quality do much to improve an already smart design- I would bet that the HPA-4 would sound even better with key components upgraded with better parts, even though those additions may indeed subtract a degree of perfection from their measurements.

So, yes, I do feel that that measurements prove a smart design and provide their benefits sonically, but I also believe that other approaches bring their own set of virtues as well- and ones that prove to be more sonically and musically meaningful.

It takes someone brave to taste them all, and figure what could improve upon measured-perfection to bring out more of what music-lovers (of any and all types) want. After all, this hobby is an art form of applied science, who is anyone to say one approach is wrong or 'worse'? Only a small minority. (Sorry, a rising politically and dogmatically-motivated minority... lol.)

I am not rigid in any aspect of audio- I have 40 years in this hobby and keeping an open mind to everything in audio reproduction is the golden rule. Eliminate biases. Re-evaluate. Repeat.

RH's picture

But there are plenty of audiophiles who do not feel their Benchmark AHB2 amps sound "hard" or "unforgiving" but rather distortion-free and wonderful. They feel they have found their "last amps."

So if you listen to the AHB2 in a system and conclude "this amp is causing the sound to be hard and unforgiving" and another audiophile listens to the same system and says "No it doesn't, it sounds smooth, distortion-free and easy on the ears" do you determine who is right? Would you remain inflexible in "what you heard?" Because after all, if other audiophiles don't hear the "hardness" you can always wave that off as "well, I guess they don't have hearing as sensitive as mine is to such artifacts."

This is why I was asking the question "How would someone show you are wrong?" which speaks to whether you'd actually be rigid, or not, in your conclusions.


(BTW, I own the Benchmark LA4 preamp and love it. It seems clean and easy on my ears in my system. But I also love tubes, so I go back and forth between it and my tube preamp).

JHL's picture


MatthewT's picture

Of the Border Patrol DAC review. I can't imagine how nasty (to me) a system assembled based only on measurements must sound.

Long-time listener's picture

I've bought my last three DACs -- Benchmark, then NAD, then Topping D90SE -- based almost entirely on measurements. All three were great purchases that sounded wonderful during their respective lifetimes and time periods. I've also made very satisfactory purchases of other components without hearing them, by triangulating between the opinions of different reviewers that I'm familiar with and that I believe do a good job of describing the sound of components. Frankly, most of the time (aside from DACs), that has meant reviewers other than those from Stereophile. Rather than using consistent terminology from one review to another that would really help readers differentiate different products, Stereophile reviewers often use flowery, poetic language that I can't translate into real-world terms. "Nuclear-level pheromone emissions" -- excuse me?!

Phastm3's picture

I purchased this dac last month and returned it 2 weeks later. I was told by the owner that streaming was the reason that his dac wasn’t able to perform at the level he designed it for. He told me that his dac was just showing me all of the flaws in my streaming system. What a joke. I’m sure steam is coming out of his ears after seeing the negative comments from JA about the measurements.

georgehifi's picture

"I was told by the owner that streaming was the reason that his dac wasn’t able to perform at the level he designed it for. He told me that his dac was just showing me all of the flaws in my streaming system."

Part true, not many R2R dac can do DSD; so yes it's the dac that's at fault, but also if the streaming companies streamed PCM instead of DSD the this dac could be very good, and better for the music.

Cheers George

Glotz's picture


supamark's picture

"but also if the streaming companies streamed PCM instead of DSD the this dac could be very good"

Streaming services use either mp3 or FLAC encoded PCM files, not DSD. Maybe you're thinking of MQA, which Tidal uses for hi-res files.

Mark Phillips,
Contributor, SoundStage! Network.

georgehifi's picture

I feel that this product review would have faired way better done at Absolute Sounds, where measurements have no bearing on what the reviewers "subjective" wordings are when read, but they are his wordings and no one else's, and there in lies the problem reading about it over there and here if your a naysayer on measurements.

Cheers George

Benjamin Zwickel's picture

This is Benjamin Zwickel, the designer and manufacturer of the Mystique X DAC.

First of all I would like to thank Herb for this wonderful review. I feel like he really got what my DAC does better than most and he articulated it beautifully.

As for the various arguments in regards to measurements, I consider most audio measurements to be either inaudible or incomplete.

In terms of inaudible, considering that there are few recordings that have more than 12-bits of dynamic range (72dB), I believe that Sony and Philips knew what they were doing when they came up with the 16-bit standard.

You would have to listen to your music above 126dB in order to hear the LSB of a full 16-bit dynamic range recording (96dB) above the 30dB noise floor in your average room in order to actually hear it.

I don't know about you, but I rarely listen above 95dB, and there are very few systems regardless of price that are capable of reproducing 126dB without significant distortion.

As for the incompleteness of most audio measurements, consider a sphere, a cone, and a cylinder. All could appear to be identical when measured in 2D but in 3D they are QUITE different. Need I say more?

So I rather than debating about measurements, I would recommend that you take me up on my 45-day no-risk audition and hear our Mystique X in your own system and decide for yourself how significant or insignificant those measurements might be.

~ Benjamin

John Atkinson's picture
Benjamin Zwickel wrote:
This is Benjamin Zwickel, the designer and manufacturer of the Mystique X DAC.

Thank you for contributing to the discussion, Mr. Zwickel.

Benjamin Zwickel wrote:
... there are few recordings that have more than 12-bits of dynamic range (72dB)

While this will be true for victims of the Loudness Wars, recordings of classical music both have a wider dynamic range and that range varies with frequency, due to the background noise having a "pink" spectrum. See, for example, the spectrum of the background noise on one of my own commercially released choral recordings:

The noise limits the low-frequency dynamic range to around 13 bits. But above 3kHz, the noisefloor allows up to 18 bits of resolution for the music, which peaks at 0dBFS.

John Atkinson
Technical Editor, Stereophile

JRT's picture

...many other DACs do, and that brings intersample overs into the dynamic range calculus, increasing the maximum possible unclipped signal peaks (if well processed) at relatively high frequencies.

I understand that you (JA1) are already well aware of this subject, but some (perhaps many?) others are not. For the tech-curious, read the paper at the following link.

Phastm3's picture

I tried the 45 day no-risk audition…it cost me $500

Benjamin Zwickel's picture

Phastm3 is correct: we don't reimburse shipping or PayPal processing fees.

That is clearly stated in our terms and conditions.

Keeping in mind this is a very heavy and very expensive product and most of what it would cost you is shipping and shipping insurance.

We don't charge any restocking fees.

Of course how many other companies offer any audition period at all?

And of the few companies who offer auditions what are their terms and conditions?

Not reimbursing shipping or banking charges is very common if not universal with the companies who offer auditions.

And I can honestly say that less than 10% of the people who audition our DACs return them for a refund. And it is no surprise that those are the same people who think they can use an inexpensive digital source to feed their DAC (garbage in, garbage out).

On the other hand, over 70% of people who audition our DACs become long term customers who trade in and upgrade to our newer models every year or three.

So I would say if you don't think it is a good idea to feed a $10K DAC with a computer or disk player that costs less than $2K statistically speaking there is a good chance you'll fall in love with our DACs.

Benjamin Zwickel's picture

I can't disagree with John Atkins in his comments in regards to some digital classical recordings having significantly more dynamic range.

Of course that doesn't change the fact that you would still have to listen at a volume high enough above the background noise in your listening room (30dB average) to actually hear the LSBs and in the case of 18-bits of dynamic range (108dB) that would be at a volume which could cause hearing damage (108dB + 30dB = 138dB).

And it doesn't change the fact that there are very few systems which have the headroom to play much above 120dB cleanly even if you wanted to listen at those volumes.

The recording companies know what they are doing when they use reasonable compression.

FYI, the most dynamic range you can physically get on an LP is about 65dB and most LPs are compressed to well below 60dB...even the best of classical recordings.

georgehifi's picture

"I can't disagree with John Atkins in his comments in regards to some digital classical recordings having significantly more dynamic range."

And just don't limit that to classical recordings, just have a look at what happened to "Yellow Brick Road" the more times it was re-issued through the years (red more compressed).

Cheers George

DH's picture

I think there's a lot of missing the point here.
We have lots of hifi equipment which measures wonderfully AND sounds good.
If I'm going to spend $10K on a component, I want it to be accurate AND sound good.
If it's not accurate (measures well) I'm paying big money for distortion/coloration that's someone's idea of good sound - but is actually just euphonic distortion.
Why pay for that? I assure you that you can find an accurate DAC that you will think "sounds good".

JRT's picture

In a video of a conversation between Herb Reichert and Steve Guttenberg while Steve was reporting on Herb's playback gear and system setup, Herb mentioned, "What I do for reviews, I try not to do too much of that Oligarch Stuff. ...somebody who works 9-5 can aspire to owning something that costs two or three or four grand. It's like a range where I think a lot of really amazing stuff happens."

I very much like that viewpoint, within a context of seemingly intentionally loose generalizations and coarse approximations.

While this $10k two channel DA converter, subject of this review, does not seem to me to reach near a category that I would describe as being "Oligarch Stuff", it does well exceed the price range that you had mentioned, even if adjusted for a couple of years of inflation.

For those aspiring to acquire major system components priced roughly within your preferred target range, spending $10k on this DA converter is a reach, and may come with some significant opportunity costs which deserve consideration, maybe especially when very good performance can be had from DA converters selling for a small fraction of that price. One that comes quickly to mind is the Benchmark Media DAC3 B (priced at $1.8k), which received very high praise in Stereophile's review(s), both in the descriptions of subjective listening evaluations, and in the objective measurements and associated comments.

edit: John Atkinson has included the Benchmark DAC3 in his follow up.

Glotz's picture

and we all know as we have read the reviews here over and over again. All of the reviewers deviate from their initial intent as writers- there are too many great products at higher price points to justify a review.

These is like a car magazine... Reading about stuff we can't afford is okay. The next review from Herb will probably be a more affordable product.

JRT's picture

I would prefer to see some extensive reviews of the relatively new RME ADI-2/4 Pro SE ($2.5k), from the perspective of a variety of reviewers. It seems to be quickly gaining an excellent reputation for its sonic performance and its capabilities feature set.

RME's marketing webpage for this product:


JRT's picture

RME ADI-2/4 Pro SE, copy/pasted from the specifications tab on RME's marketing webpage for this product.

RME ADI-2/4 Pro SE
2-AD/4-DA 768 kHz, High-Performance Converter

Analog Inputs

Input: XLR and 6.3 mm TRS jack, servo-balanced
Input impedance @ 1 kHz, balanced: 90 kOhm, unbalanced: 45 kOhm
Input sensitivity switchable +24 dBu, +19 dBu, +13 dBu, +7 dBu, +1 dBu @ 0 dBFS
Digital Trim Gain range: 0 dB up to +6 dB, in steps of 0.5 dB
Signal to Noise ratio (SNR) @ +13/19/24 dBu: >120 dB (AES17), >123 dBA
Signal to Noise ratio (SNR) @ +7/+1 dBu: >120 dB (AES17), 123 dBA
Frequency response* @ 44.1 kHz, -0.1 dB: 0.4 Hz – 20.4 kHz
Frequency response* @ 96 kHz, -0.5 dB: 0.2 Hz – 45.5 kHz
Frequency response* @ 192 kHz, -1 dB: 0.15 Hz – 90.9 kHz
Frequency response* @ 384 kHz, -1 dB: 0.15 Hz – 182 kHz
Frequency response* @ 768 kHz, -1 dB: 0.15 Hz – 348 kHz
THD @ -0.5 dBFS: > -130 dB, 0.0000316 %
THD+N @ -0.5 dBFS: -119 dB, 0.000112 %
Channel separation: > 130 dB @ 1 kHz
* DC Filter None. With Filter RME @ -1 dB: < 0,5 Hz

RIAA Mode via TS

RIAA deviation 20 Hz – 20 kHz: < ±0.05 dB
Input impedance: 45 kOhm @ 1 kHz, input capacitance 150 pF
Sensitivity: +14 dB: 20 mV, +20 dB: 10 mV, +26 dB: 5 mV
Sensitivity: +32 dB 2.5 mV, +38 dB: 1.25 mV
Additional 0 to +6 dB Gain in steps of 0.5 dB via Trim Gain
Maximum input level @ +14 dB: 114 mV, -16.6 dBu
Maximum input level @ +20 dB: 57 mV, -22.6 dBu
Maximum input level @ +26 dB: 28.4 mV, -28.7 dBu
Maximum input level @ +32 dB: 14.3 mV, -34.7 dBu
Maximum input level @ +38 dB: 7.2 mV, -40.6 dBu
Digital headroom in all Gains, related to sensitivity: 15 dB
Signal to Noise Ratio (SNR), 20 Hz-20 kHz, +14/20 dB Gain: 88 dB, 92 dBA
Signal to Noise Ratio (SNR), 20 Hz-20 kHz, +26 dB Gain: 80.5 dB, 85 dBA
Signal to Noise Ratio (SNR), 20 Hz-20 kHz, +32 dB Gain: 74 dB, 78.5 dBA
Signal to Noise Ratio (SNR), 20 Hz-20 kHz, +38 dB Gain: 68 dB, 73 dBA
THD+N, 20 Hz-20 kHz, +14/+20 dB: -88 dB, 0.004 %
THD+N, 20 Hz-20 kHz, +26 dB: -81.5 dB, 0.0084 %
THD+N, 20 Hz-20 kHz, +32 dB: -74 dB, 0.019 %
THD+N, 20 Hz-20 kHz, +38 dB: -69 dB, 0.035 %

Analog Outputs
1/2 XLR

Maximum Output Level: +26.5 dBu @ 0 dBFS
Output level switchable +24 dBu, +19 dBu, +13 dBu, +7 dBu, +1 dBu @ 0 dBFS
Signal to Noise ratio (SNR) @ +7/+13/19/24 dBu: 120 dB (AES17), 123 dBA
Signal to Noise ratio (SNR) @ +1 dBu: 119 dB (AES17), 120 dBA
Frequency response @ 44.1 kHz, -0.1 dB: 0 Hz – 21 kHz
Frequency response @ 96 kHz, -0.5 dB: 0 Hz – 44.9 kHz
Frequency response @ 192 kHz, -1 dB: 0 Hz – 90 kHz
Frequency response @ 384 kHz, -1 dB: 0 Hz – 181 kHz
Frequency response @ 768 kHz, -3 dB: 0 Hz – 285 kHz
THD @ 0 dBFS: < -120 dB, 0.0001 %
THD+N @ 0 dBFS: -116 dB, 0.00016 %
Channel separation: > 130 dB
Output impedance @ 1 kHz @ +24/+19/+13 dBu: 213 Ohm
Output impedance @ 1 kHz @ +7/+1 dBu: 113 Ohm

1/2 TS (rear)
As output XLR, but:

Output: 6.3 mm TRS jack, impedance-balanced
Maximum output level: +21.5 dBu @ 0 dBFS
Signal to Noise ratio (SNR) @ +19/13 dBu: 120 dB (AES17), 123 dBA
Signal to Noise ratio (SNR) @ +7 dBu: 119 dB (AES17), 121 dBA
Signal to Noise ratio (SNR) @ +1 dBu: 118 dB (AES17), 118 dBA
Output impedance @ 1 kHz: balanced 213 Ohm, unbalanced 106 Ohm

Phones 1/2, Phones 3/4
As Output 1/2 TRS, but:

Output: 6.3 mm TRS stereo jack, unbalanced
Maximum output level: +21.5 dBu @ 0 dBFS
Output impedance: 0.1 Ohm
Signal to Noise ratio (SNR) @ +19 dBu (Hi-Power): 120 dB (AES17), 122 dBA
Signal to Noise ratio (SNR) @ +7 dBu (Lo-Power): 119 dB (AES17), 121 dBA
Signal to Noise ratio (SNR) @ +1 dBu (IEM): 118 dB (AES17), 120 dBA
Output level at 0 dBFS, Ref Level +19 dBu, load 100 Ohm or up: +21.5 dBu (9.2 V)
Minimum load impedance IEM @ +1 dBu: 4 Ohm
Minimum load impedance Lo-Power @ +7 dBu: 8 Ohm
Minimum load impedance Hi-Power @ +19 dBu: 24 Ohm
THD @ +20 dBu, 32 Ohm load (1.9 Watts): -124 dB, 0.000057 %
THD+N @ + 20 dBu, 32 Ohm load (1.9 Watts): -118 dB, 0.00011 %
Max power @ 0.001% THD: 2.1 W per channel
Channel separation 20 Hz – 20 kHz: 80 dB @ 32 Ohm

Balanced Phones mode
As before, but:

Maximum output level: +27.5 dBu (18.4 V) @ 0 dBFS
Output levels switchable: IEM +7 dBu, Lo-Power +13 dBu, Hi-Power +25 dBu
Output levels switchable: IEM 1.73 V, Lo-Power 3.46 V, Hi-Power 13.8 V
Output impedance: 0.2 Ohm  Signal to Noise ratio (SNR) @ +25 dBu: 120 dB (AES17), 122 dBA
Signal to Noise ratio (SNR) @ +13 / +7 dBu: 119 dB (AES17), 121 dBA
Minimum load impedance IEM @ +7 dBu: 8 Ohm
Minimum load impedance Lo-Power @ +13 dBu: 12 Ohm
Minimum load impedance i Hi-Power @ +25 dBu: 40 Ohm
Max power @ 0.001% THD: 4 W per channel  Channel separation: > 130 dB @ 1 kHz

Digital Inputs

Lock Range: 28 kHz – 200 kHz
Jitter suppression: > 50 dB (> 1 Hz)
Accepts Consumer and Professional format

1 x XLR, transformer-balanced, galvanically isolated, according to AES3-1992
Input sensitivity 1.0 Vpp
SPDIF coaxial

1 x RCA, transformer-balanced, according to IEC 60958
High-sensitivity input stage (< 0.3 Vpp)
AES/EBU compatible (AES3-1992)
SPDIF optical

1 x optical, according to IEC 60958
ADAT compatible

Digital Outputs

1 x XLR, transformer-balanced, galvanically isolated, according to AES3-1992
Output level 2.7 Vpp
Format Professional according to AES3-1992 Amendment 4
Single Wire mode, sample rate 28 kHz up to 200 kHz
SPDIF coaxial

1 x RCA, according to IEC 60958
Output level 0.75 Vpp
Format Consumer SPDIF according to IEC 60958
Single Wire mode, sample rate 28 kHz up to 200 kHz
SPDIF optical

1 x optical, according to IEC 60958
Format Consumer (SPDIF) according to IEC 60958
Sample rate 28 kHz up to 200 kHz
Clocks: Internal, AES In, SPDIF In, ADAT In
Jitter suppression of external clocks: > 50 dB (> 1 Hz)
Effective clock jitter influence on AD and DA conversion: near zero
PLL ensures zero dropout, even at more than 100 ns jitter
Additional Digital Bitclock PLL for trouble-free varispeed ADAT operation
Supported sample rates for external clocks: 32 kHz up to 200
Internally supported sample rates: 44.1 kHz up to 768 kHz
Included power supply: external switching PSU, 100 - 240 V AC, 3.3 A, 40 Watts
Standby power consumption DC 12 V: 170 mW
Standby power consumption AC 230 V: 280 mW
Idle power consumption: 14 Watts, Max. power consumption: 30 Watts
Idle current at 12 V: 1.16 A
Dimensions (WxHxD): 215 x 44 x 160 mm (8.5" x 1.73" x 6.3")
Weight: 1.2 kg ( 2.2 lbs)
Temperature range: +5° up to +40° Celsius (41° F up to 112°F)
Relative humidity: < 75%, non condensing

ok's picture

..John Atkinson's negative attitude towards mojo's mystique. It measures reasonably well for a NOS dac and it seems to sound the part as any "alternative" component favorably reviewed here.

Chubby's picture

The DENAFRIPS Terminator does not use a NOS R2R

MatthewT's picture

Negative review. Not everything can be better than everything that came before it.