Mojo Audio Mystique EVO Pro D/A Converter
The vast majority of modern audio DACs use a Delta-Sigma DAC chip because such chips are now tiny, manufactured by companies that sell them in mass relatively cheaply, and have become outstanding in quality. Examples of such chips include the California USA Company’s ESS Technology 9038PRO SABRE chip used in many of the highest-end DACs. Even their slightly lesser chip, the ESS Technology 9028PRO SABRE has been used impressively in some high-end DACs with which I am familiar such as the Mytek Brooklyn DAC + at only $2195.
There are, however, a few companies that use other methods. One novel alternative is illustrated by PS Audio’s DirectStream (DSD) DAC at $6000 that stands out because of its use of a Field Programmable Gate Array (FPGA) which is a circuit board that is programmed with software (that can be upgraded on a regular basis) to perform the essential conversion, and does so using DSD. This FPGA method for use in Hi-Fi audio, however, requires sophisticated software to run which most audio designers do not have the knowledge to assure (PS Audio, for example, leads their effort with one incredibly talented man: Ted Smith). The DirectStream DAC has been my reference for a number of years now.
Yet another method is one that has been around for decades but is rarely seen in modern DACs: The R-2R Resistor Ladder DAC. It is a simple and clever variation of a so-called binary-weighted DAC that uses a repeating sequence (say, of length b) of only two resistor values R and 2R to convert a binary digital audio sample into its analog voltage. Each step (ladder) of the sequence handles 1 of the bits of the digital sample (for example from b = 16 bit samples). The bit in position i goes through b—i steps yielding its correct resistance value and hence correct voltage. The main point is that one does not have to have separate resistors for each of the b values of resistance needed (R, 2R, 4R, 8R, ..., 65536R); only the R and 2R values are needed. And it does not require any error correction, upsampling, oversampling or digital filtering/noise shaping.
Although the R-2R Resistor Ladder DAC is the simplest and easiest to understand (only very basic knowledge of electrical engineering is sufficient), it is not the easiest to construct and implement for creating the precision needed in a high-end audio DAC which is probably why there are so few companies manufacturing them. Notable companies that do so include Schiit Audio, MSB Technology, and in more recent years Mojo Audio—the focus of this review with its new version of its Mystique DAC: the Mystique EVO Digital-to-Analog Converter (EVO).
EVO refers to ‘Evolution’, meaning its chassis allows for ever-improving, retrofittable upgrades and updates from a ‘Basic’ model to ‘Pro’. Importantly, Evo still uses the original (manufactured over 30 years ago) NOS (Non-Over-Sampling) AD 1862 R-2R chipset, but is a significant improvement/upgrade to the 3rd generation of the Mystique, the v3, and offers a variety of higher end options. For example it still utilizes LC choke input technology power supplies (3 digital, 2 analog)—originally developed by Western Electric and Bell Labs almost 100 years ago—heavy and expensive but implemented with the best of modern components, but the EVO now offers the add-on option of using superior Lundahl Transformers. As on the v3, the EVO only plays PCM and does so up to 24/192 (no DSD or MQA support).
Mojo Audio’s new Mystique EVO has three models: Basic at $5999, B4B at $7999 and the model under review, the Pro at $9999.
The Pro I was sent for review was equipped with additional options raising the price to $11,349 (S/PDIF: Lundahl Amorphous Core, Analog Choke: Lundahl Amorphous Core, and .01% Vishay TX2575 ‘Nude’ resistors). Note: The Vishay Nude resistors are considered the best resistors made by Vishay and the most transparent resistor in the world. They are relatively expensive and the ‘Nude’ means that rather than a ceramic covering they use something similar to the polymer coating used on Litz wires (minimum non-conductive coating). Included onboard the Pro is an OCC UniCrystal V-Quad Cu24 wire between power supply and signal boards and double-Pi AC input filter.
Remaining options such as Stillpoints Ultra II or Ingress Audio roller-block anti-resonance feet and advanced Graphene and copper EMI/RFI shielding were not on the review unit; with them the Pro price jumps to a high of nearly 15K.
Many thanks to Mojo Audio’s founder Benjamin Zwickel for sending me the EVO Pro for review. As I have witnessed over the last 7 years, Zwickel marches to the beat of a different drummer and does so with addictive passion, meticulous engineering and down to earth, simple presentation. He continues in that tradition here.
Mojo Audio describes the Mystique EVO Pro DAC as follows:
Our Mystique EVO digital-to-analog converter is a no-compromise design incorporating ultra-purist non-oversampling direct-coupled R-2R topology. With no pre or post digital filtering, digital noise shaping, upsampling, oversampling, or error-correcting algorithms, our digital signal path is the purest possible.
The result is the most natural time and tune possible from a digital source. Effortless micro-dynamics and incredible micro-detail preserve the tone, timbre, texture, and spatial cues that bring you closer to the original musical performance.
The EVO has no monitor on the front panel, no built-in volume control/preamp and not even a remote control: Except for choosing what input you want by selecting 1 button on the front (out of three from among USB, Coax, AES), there is nothing else to do. Powered on in the back, you just keep it on. It has both single-ended (RCA) out and balanced (XLR) out to a preamp as well as the 3 digital input connections. As for the chassis, which is built like a tank, it is a unique low-resonance, advanced shielding, 2-piece, black, hammer-tone, polymerized aluminum composite chassis. According to Zwickel , it took over five years to design. Last but not least, the EVO is handmade, each unit requiring about 1 week of work to build. The unit is 17.5”W x 3.5”H x 12”D and 29 pounds in weight.
My Use
I used my reference Mojo Audio Deja Vu Music Server endowed with a Roon Core, and connected to the internet via an ethernet cable. I used Qobuz and Tidal (via Roon) streaming in addition to using files on the internal music server’s hard drive. I used a Wireworld Platinum Starlight 7 USB Cable to connect the EVO to the Deja Vu. To my reference preamp (PS Audio BHK) I used the single-ended (RCA) out from the EVO because I needed 1.5 meters and did not have an extra cable of that length at hand in XLR; but I did experiment later with XLR after borrowing one of sufficient length (see later section). Then balanced out from the BHK to my reference Audio by Van Alstine DVA SET 600 Mono Block Solid State Amplifiers and onto my reference speakers (Alta Audio Celesta FRM-2M). Reference cabling throughout included the highest level of Audio Art Cable, Anticables, Waveform Fidelity (review forthcoming of their newest version of speaker cables) and Clarus.
Sound
When I first set up the DAC, I was very busy with other work for about two days. During that time I let my wife and kids play whatever their hearts desired (as a way of ‘burning’ in the DAC). Finally, I had time to seriously listen alone for several hours. What I heard was a sound quality so unusually compelling that I began to question what had caused this in my listening space. Was it the new rug? Not possible since the rug was not new anymore (purchased months ago). Was there some new furniture changes in the setup space? No, nothing of that sort. The only differences were perhaps a tiny plant or two that my wife had placed way out in one corner and a new water filter for our ever growing pet turtle, Tootle.
Although so many high-end DACs sound super these days, and at times admittedly I have had difficulty distinguishing the sound quality between them, this DAC seriously challenged me: It demonstrated a rare cohesion of all the best individual audiophile qualities such as timbral accuracy, timing, imaging, an ample 3D soundstage, clarity, spatial cues, textures, and so on. So, I decided to challenge it by going back to play many of my favorite recordings to gain a better understanding of what I was witnessing; I offer a sample of such here. Often, this only required 16/44.1 resolution (Red Book).
Alexandra Leaving, from the album The Essential Leonard Cohen, (2002), 16/44.1 FLAC, Qobuz or Tidal. This track was part of a serious collaboration with Sharon Robinson and in this piece she was the backup vocalist. In past listening on other DACs in my system, Cohen and Robinson sound so close together in timing that often visitors not in the know would ask me, ‘who is the singer’ as opposed to ‘who are the singers’. With the EVO, it is obvious right away that there are two singers, and Leonard’s voice is presented with an abundance of texture that I do not recall hearing so evident before. The best presentation of this piece I have heard on my system. Poetry in motion. Ditto for (same album), the track ‘Waiting for the Miracle’: the textures of Cohen’s voice and subtle lovely spacial queues I heard from the EVO yielded the best presentation I have ever heard on my system from any DAC I have tried. I found myself listening to the entire double album and appreciating way more some of Cohen’s other tracks; ‘Famous Blue Raincoat’ for example was expressed so powerfully with emotion by the EVO.
J.S. Bach: Goldberg Variations, Kenneth Gilbert (using a Hubert Bedard harpsichord), 1987, Harmonia Mundi, 16/44.1 FLAC on Qobuz. Gilbert, a Canadian, was recommended to me by Anthony Kershaw when I commiserated with him how I was finding some harpsichord music recordings tedious and boring and thin sounding. (He pointed out that long ago conductor Sir Thomas Beecham described the sound of the harpsichord as ‘two skeletons copulating on a tin roof’!) Anyhow, the Goldberg Variations have always been one of my favorite pieces of music, but performed on a piano. Although I still prefer the piano, Gilbert’s erudite but at times even playful performance with superb recording quality are on full display with the EVO. No thin sounds here. Rich and full presentation that extended all the way to the back of my system’s sound stage; a realism with spacial cues from the recording that surpassed anything I had heard before.
‘Piano Sonata: The Moon’, from the album Trilogy, Chick Corea Trio, (2013), a live performance, 24/96 FLAC on my internal hard drive. This piece has always kept me riveted and mesmerized, especially at the 10 minute mark; the extraordinary jazz drummer Brian Blade is at his best with unbeatable intricate cymbal work along with Corea on piano and Christian McBride on double bass. The cohesion of the three musicians displayed by the EVO is something to behold. A live performance that the EVO turns to gold.
Perhaps as a fine match with the Nude resistors in the EVO, I must include some pieces by Musica Nuda, such as ‘Ninna Nanna’ from the album Quam Dilecta (16/44.1 FLAC on Tidal) and ‘I Giorni di Festa’ (Live) from the album Verso sud (24/44.1 FLAC on Qobuz). Magnificent sounding from the EVO, so evocative and intimate; just vocals and bass. I have certainly enjoyed these pieces before, but the EVO’s presentation kept me so much more enamored with, and engaged in, the performance.
Balanced or single-ended output to the preamp, how do they compare?
As mentioned earlier, I started out by using the RCA output because I did not have a long enough XLR cable on hand, but within days an audiophile friend living nearby kindly lent me a pair of Clarus Crimson XLRs. Besides the obvious significant difference in volume level (XLR are louder), the sound quality was extremely close; at best I decided that the RCA offered a dash more of a rounded sound, maybe, while the XLR offered a dash more of a shaper sound, maybe; but I would take either one at the end of the day if the cable lengths needed are short as in my case.
An interview with Benjamin Zwickel
I asked Zwickel questions via email.
1. For as long as I can remember (at least 7 years now) you have never been into ‘trendy’ things in the audio arena (DSD, MQA, PCM higher than 24/192 for example, among others), but instead have focused on what you have learned from experience as an engineer/designer/audiophile and believe to be important in yielding the very best sound quality in the designing of your products. Could you give us some insight into your thinking/philosophy about this?
I’m a music lover first and an engineer second. So, to me all of this technology is a necessary evil to allow me to enjoy recorded music in my home. If you were to take all the recording mediums and formats that exist, from Edison transcription cylinders, to DSD1024, and everything in between, over 90% of all recordings available today are only available in one format: 16-bit 44.1KHz Red Book. Aside from that, I’ve seen formats come and go, so I knew that the recording industry was never going to switch to MQA or DSD when most of the world was still happy with Red Book. And lastly there’s the math: unless you want to make a fool of yourself and argue with a mathematical legend like Dr. Nyquist, the indisputable fact is that sampling at double the frequency of the bandwidth yields 100% of the resolution. The only advantage of sampling at higher frequencies would be to avoid digital artifacts in the audible spectrum caused by aliasing. According to recording engineering legend Dan Lavry, artifacts from aliasing can be avoided by sampling at 60KHz. So 88.2KHz is more than high enough to capture 100% of the music being recorded. Anyone who believes otherwise is not comparing apples to apples: the HD or other format recordings they are listening to have been remastered better, or the playback hardware they are listening on has been engineered better, rather than the format itself being the reason for the higher resolution they are experiencing. For those of you who disagree, I suggest you read this.
2. When and how did you come to the realization that a R-2R Resistor Ladder DAC was the way to go and why?
I was first introduced to R-2R by one of my audio mentors a couple of decades ago. At that time there seemed to be a bit of a mystique about the vintage R- 2R DACs. This is where I got the name for my Mystique DAC. As I studied the engineering differences between R-2R and Delta-Sigma DACs I learned that Delta-Sigma was based on error correction algorithms where as R-2R was the only technology that didn’t assume there were errors in the musical bit stream. As I made further listening comparisons it became quite clear that Delta-Sigma always seemed to sound overly refined and smoothed over where as R-2R was more articulate and revealed subtlety and nuance that seemed to be filtered out by the Delta-Sigma process. From then my goal was error prevention rather than error correction, which is why my DACs have insane power supplies, integrated anti-resonance, and advanced shielding.
3. You clearly are not interested in designing ‘integrated’ devices such as a DAC/Preamp/Music Server/ —all in one. Why is that?
The short answer is that if I built an integrated digital source with power supplies, anti-resonance, and shielding equal to my Mystique DAC, it would be too large and heavy for most people to lift. The more complex answer is similar to why the best phono stages are not integrated into preamps: noise. Computer- based servers/streamers and optical transports are extremely noisy so ideally they should be in a separate chassis with a separate power supply.
4. Now that you have designed superb power supplies, music servers, and DACs, what do you think will be your next desired challenge in audio?
People ask me all the time when am I going to manufacture amplifiers and phono stages, but I simply can’t get past the digital source. Our new EVO line of components will include a Deja Vu EVO music server/streamer/NAS/router and an Illuminati EVO power supply. You can expect upgrade options for our new EVO line for years to come. The only new direction Mojo Audio is moving into will be the recording side: we’ll soon be manufacturing pro audio ADCs, as well as DACs, recording computers, and power supplies. And I hope to personally be doing some ultra-purist acoustic recordings over the next few years. To me it all comes down to the source, and the recording is the source of all sources.
Summary
Many high-end DACs available today no doubt offer extraordinary transparency, fine imaging, a large soundstage and impressive timbral accuracy—and most can even boast of playing anything that you throw at them (PCM, DSD, MQA). But how many can take the basic ingredients of fine imaging, extraordinary transparency, textures, spatial cues and impressive timbral accuracy and combine them in a 3D soundstage in an uncanny way into something that sounds so beautiful, natural sounding, musical and addictive as to demand that one say (often attributed to Aristotle), ‘The whole is larger than the sum of its parts’? Very, very few. The Mojo Audio Mystique EVO Digital-to-Analog Converter is certainly one of them. A must hear. The best sounding DAC I have so far ever heard in my own system. Very highly recommended.
Further information: Mojo Audio
