• AURALiC ARK MX+ DAC Review

    The AURALiC ARK MX+ digital to analog converter is a well engineered, uniquely constructed, and very good sounding DAC. Right out of the box it's visually and physically evident the ARK MX+ was built to a very high standard. This 12 lbs. DAC made of iron, nickel, and silicon looks and feels like it was built to military specifications. Internally the AURALiC ARK MX+ houses some unique design elements including the main attraction AURALiC Sanctuary Audio Processor. This multi-core ARM9 processor and the ARK MX+ unique ActiveUSB™ asynchronous USB interface is a combination not found in any other converter. The aforementioned building blocks play a major role in solidifying this DAC's foundation and most importantly contribute to the final result that is very good sound quality.








     

    ARK MX+ DAC - Well Engineered, Uniquely Constructed

    The $1,999 AURALiC ARK MX+ is not simply another version of the same industry standard parts assembled under a unique name. This Digital to Analog Converter DAC was designed by AURALiC, in combination with Switzerland based ArchwaveAG, using a distinctly different approach from any DAC I've yet reviewed. AURALiC and ArchwaveAG designed the unique AURALiC Sanctuary Audio Processor that comprises the core of the ARK MX+. The Sanctuary Audio Processor is built on an ARM9 multi-core processor. This processor is capable of handling 32 bit / 192 kHz multi-channel audio without breaking of sweat. However, the rest of the DAC is not as capable as this powerhouse processor. The ARK MX+ handles two audio channels up through 24 bit / 192 kHz. Nonetheless it's nice to know the hardware is capable of much more than anyone could ever ask of it. On a side note, many CA readers probably recognize the name ARM. ARM processors are in close to 90% of the world's mobile devices such as phones, tablets. and portable music players. Another unique design element of the ARK MX+ is its USB implementation. According to the USB Prober application available for Mac OS X the ARK MX+ is an Isochronous asynchronous data endpoint. That in itself is not incredibly unique. Wavelength and dCS have had Isochronous asynchronous USB DACs for years. What is unique to the ARK MX+ is what AURALiC calls ActiveUSB. ActiveUSB uses a large data buffer, up to two seconds of audio, and a multi-stage high resolution PLL circuit to severely limit the digital disaster known as jitter. Astute and purist Computer Audiophile readers likely noticed that the ARK MX+ is asynchronous but also uses a PLL. The purist asynchronous USB group of people have held strongly to the view that no PLL can be used in a truly asynchronous USB implementation. Instead of a PLL to generate a synthesized clock signal purists use two crystal oscillators. One for frequency multiples of 44.1 kHz (88.2, 176.4) and the other for multiples of 48 kHz (96, 192). This is where things get interesting. The AURALiC ARK MX+ also contains two crystal oscillators, one for each same rate family. In one way the DAC is like other asynchronous USB DACs and in another way it's like adaptive USB DACs that use PLLs. I haven't been able to clearly understand how this specific data path works as it's not like anything I've ever used. When consulting with an industry veteran and colleague who has designed high end components for decades I was told AURALiC is using asynchronous sample rate conversion (ASRC) within frequency families (44.1 kHz or 48 kHz) so the PLL is run in slave mode but with a new master clock for the output. This DAC run in asynchronous USB transfer mode and uses asynchronous sample rate conversion (ASRC). Most DACs use one or the other for of async. In addition to the clocking scheme with fixed oscillators and a PLL, the ARK MX+ uses a self-adapted up sampling circuit. AURALiC, and many other manufacturers, are believers in the sonic benefits of upsampling when its done right. This specific upsampling implementation uses more of a purist method of converting from one frequency to another. The self-adapted upsampling automatically selects the final frequency based on the original incoming frequency. Music with a base rate of 44.1 kHz is upsampled 176.4 while music based on a 48 kHz clock is upsampled to 192 kHz. According to AURALiC and many audio purists this method of upsampling avoids distortion created when converting frequencies using a non-integer algorithm such as converting 44.1 kHz to 192 kHz. There is no simple answer or best method of upsampling. Asking ten engineers for the best method would likely produce twenty different answers. Potential customers are encouraged to listen for themselves if at all possible.


    Additional somewhat geeky information about the AURALiC ARK MX+ includes the use of AURALiC's Pure-Power AC filter that isolated common-mode and differential-mode interference, a customized Plitron low noise transformer, an AKM AK4399 32-bit DAC chip and a TI SRC4392 sample rate conversion chip that's limited to 24 bits. The ARK MX+ does not contain a digital or analog volume control. The DAC requires one to use a preamp for best sonic results. Analog outputs are both single-ended (RCA 2Vrms) and balanced (XLR 4Vrms). I used the balanced outputs exclusively for this review. Digital inputs on the ARK MX+ are limited to USB and Coaxial S/PDIF (electrical RCA). In lieu of an additional digital input AURALiC opted for an AES/EBU digital output capable of the same sample rates as the inputs up through 24/192. This digital output allows one to use the ARK MX+ as a USB to AES converter if desired.

    External design elements of the AURALiC ARK MX+ include material to block electromagnetic interference and dampen resonances. AURALiC's AFN402 material is an alloy consisting of iron, nickel, silicon and other rare metals. Lifting the ARK MX+ or wrapping on the external housing with one's knuckles is a good way to experience what AFN402 is all about. AURALiC is also using its Alire resonance damper material on the ARK MX+. Alire, according to AURALiC, "[Is] made by arranging and pressing multiple raw materials of diverse resonance frequency from layer to layer. [S]ticking Alire to the inner side of cabinet so as to make the 'sandwich' structure, resonance could be effectively restrained and its resonance frequency could be changed." The ARK MX+ is a very solid unit to say the least. My one minor complaint about the unit has to do with its feet. The ARK MX+ has three feet on the bottom made of plastic cylinders with rubber in the center. The rubber pads of the feet don't extend beyond the plastic cylinder/housing. Thus, the DAC's hard plastic cylinders touch the component, rack, or desk underneath the unit. A fairly heavy DAC such as the ARK MX+ can easily scratch most surfaces it rests upon if the rubber feet don't protrude beyond the plastic cylinders.


     


    Using The AURALiC ARK MX+ DAC

    I used the AURALiC ARK MX+ for a few months in a few different configurations. One configuration didn't work, one worked well, and one worked wonderfully. The AURALiC's asynchronous USB implementation requires either a Class 2.0 USB compliant operating system such as Linux and OS X 10.7 or installation of the included software for proper operation under Windows. I attempted to use the ARK MX+ with the Aurender S10 music server from Widealab. The ARK MX+ was recognized as an async 192 kHz DAC by the Aurender but that's where the fun stopped. For the most part the ARK MX+ does not work with the version of Linux used in the Aurender. Redbook 16/44.1 content played fairly well but suffered from pops and ticks every thirty seconds. Music above 44.1 suffered from more pops and ticks as the sample rates increased. As a control for my testing I used the Berkeley Audio Design Alpha USB converter with the Aurender and it functioned perfectly at all sample rates via USB. Thus, I know the Aurender can operate just fine with high resolution USB devices just not the ARK MX+. The second configuration I used was a MacBook Pro running OS X 10.7.x, Pure Music as recommended in the ARK MX+ manual, and iTunes 10.5.x. The ARK MX+ worked well without any software installation. Sample rates switched on the fly even up to 24/192 without a hitch. The third configuration I used was a C.A.P.S. v2.0link server, with SOtM tX-USB card, running Windows 7 and JRiver Media Center 17. It was this configuration I used 95% of the time during the review period as it worked wonderfully.

    Using the ARK MX+ with Windows required installation of the AURALiC software. This software installs an ASIO driver and an AURALiC USB control panel. This control panel is unavailable when using the ARK MX+ with another operating system. The two main features I used in the control panel were Sample Rate Selection Mode and the Isochronous Streaming settings. The Sample Rate Selection Mode can be set to Automatic or Manual. Readers guessing that Automatic Mode will adjust the sample rate automatically are incorrect. The required setting that allows applications such as JRiver Media Center to switch the sample rate on the fly is Manual Mode. This is a bit counterintuitive but once it's set a user should never have to return to the setting. The Isochronous Streaming settings page offers users the chance to tweak playback more so than on Mac OS X or Linux. On this page users can adjust stream buffer, ASIO buffer, and WDM buffer depth. According to the ARK MX+ manual, "Buffer depth settings will affect jitter performance of ARK MX+ which results in different sound quality. It is always recommended to set the value as low as possible. A modern PC with a ‘clean’ Windows system should be able to replay music without interrupt at all three buffer depths set to lowest." In my experience changing buffers to an extremely low number of samples or low ms value has only increased instability, not sound quality. Many CA readers report opposite results thus it's recommended user fiddle with these settings until their heart's content. Taking the advice of the manual I set the ms value to 1, the lowest setting. Playback was a bit choppy with drop outs, pops, and ticks. I doubled the value to 2ms and had success for the rest of the review period. (I also set the JRiver ASIO buffer settings to 0.02s). In fact, after setting all buffers to 2ms I tried to make something bad happen such as a drop out or halted playback etc. I changed sample rates excessively and even let JRiver run for a couple days to see if I could at least cause the ASIO Statistics display to show a dropout. I was unsuccessful. 2ms was the lowest setting I could use and it was rock solid from start to finish.

    I attempted to use the AR MX+ with other audio output modes such as WASAPI, WASAPI - Event Style, and Kernel Streaming but wasn't able to output stable sound. This was more of a test than anything else as the DAC ships with a special ASIO driver.




     

    Listening To Music Through The AURALiC ARK MX+ DAC

    Listening to music through the ARK MX+ was a pleasurable experience. The sound quality coming from my system with the AURALiC DAC in place was very good. Like all audio components the ARK MX+ has its own sonic signature that some listeners will love while others prefer something different. I don't think many audiophiles could listen to the ARK MX+ and completely dislike the sound. Again, it's very good. My overall impression of the sonic character was that similar to the equalizer smiley face curvelink with bass and treble up and midrange at the bottom of the smile. There is a reason this EQ curve is so popular. People really like it. The sound was a bit thicker or heavier than I prefer at the bottom end and a touch congested in the midrange at the bottom of the smile. This thickness is not to be confused with slowness as the ARK MX+ was not slow at all (for the most part) even to the lowest bass notes. I used the qualifier "for the most part" in the previous sentence because I experienced a strange slowness over the whole frequency spectrum while playing a couple 24/88.2 kHz tracks. I couldn't quite grasp what was going on because this slowness was absent throughout all my other listening sessions. Who knows, maybe this had more to do with me than the DAC. Perhaps the best thing I heard while using this DAC was blasting a few Pearl Jam albums at 90+ dB! The AURALiC ARK MX+ is a terrific rock and roll DAC.

    Comparing the AURALiC ARK MX+ to my Berkeley Audio Design Alpha DAC Series 2 with Alpha USB converter I concluded a major difference between these two DACs was air and space around the instruments throughout the soundstage. The Alpha DAC ($5,000) reproduces music with an incredible air surrounding the musicians and an openness at the top that is, pardon the pun, hard to top. Listening to Antonio Forcione & Sabina Sciubba's Meet Me In Londonlink (192kHz Super Hi Def Edition) on the Naim Label was really good through the ARK MX+ and great through the Alpha combination, at over triple the price. Partly because the Ark MX+ is a really good sounding DAC and partly because this recording is stellar, the whole listening experience can take one's mind out of the real world for 45 minutes. Track number two Take Five is a great take on the Dave Brubeck classic. Highly recommended and downloadable at 24/192 directly from the Naim Label. Listening to Gary Karr's Bass Virtuouso (Classic Records HDAD 2013link) at 24/192 gave me a different perspective of the ARK MX+. If sound could be captured in a photograph the finish would be matte as opposed to glossy. There's no right or wrong, better or worse, just personal preference that may change based on the type of music or subject of a photograph. Maybe the ARK MX+ would be great in combination with a system that's a touch too bright or glossy. Lastly I listened to the new album Chimes of Freedom: The Songs of Bob Dylanlink. Disc three track 19 is a cover of Don't Think Twice, It's Alright by pop star and Auto-Tune queen Ke$ha. Surprisingly this track has none of the effects that made her famouslink. Maybe that's why I like it quite a bit. It's simply Ke$ha's voice with a tiny instrumental in the background toward the end of the track. Listening through the AURALiC ARK MX+ the track sounded really good and a bit on the dark side. Listening to the same track through my Berkeley Audio Design setup I heard more openness at the top and got a better feel for the recording space. As I said a couple times earlier the AURALiC ARK MX+ produced very good sound with a character that many CA will enjoy. Especially at less than $2,000.

     

    Conclusion

    The AURALiC ARK MX+ digital to analog converter is extremely competent in this very competitive segment of the DAC market. The unique design of the ARK MX+ sets it apart from much of the competition. The Sanctuary Audio Processor controlling the entire unit with an ARM9 chip is distinctly AURALiC. In addition the well engineered asynchronous USB implementation is unlike any I've reviewed to this point. Using this DAC with a Windows computer enables listeners to tweak the ASIO driver if they so chose. Throughout my entire listening experience with the ARK MX+ the ASIO driver was rock solid and stable. The also solid iron and nickel chassis is as heavy as the rock music reproduced so well by this DAC. I had a blast listening my favorite band Pearl Jam through the very good AURALiC ARK MX+.


     



     



     


    Product Information:


     


    Associated Equipment:





     



     
    Comments 10 Comments
    1. Audio_ELF's Avatar
      Audio_ELF -
      @firedog<br />
      <br />
      I actually think that DACs need reviewing (in a computer context) as a complete package: that is comparing Berkeley USB + Alpha DAC to a USB connected DAC.<br />
      <br />
      Eloise
    1. One and a half's Avatar
      One and a half -
      read the reasons behind, yes, OK, but you can also get great results without upsampling, Ayre QB9 which the auralic competes in the price range is a fine example. <br />
      <br />
      It's OK to compare to a reference DAC, but also more to the point a device in the similar ball park in price and technology should also be included in the review. If the reference can be upstaged, you have a winner. <br />
      <br />
      Adjustments may to fun to play with, a DAC should play, and, as a priority play well out of the box sans trickery. Surely such distractions hide something else.<br />
      <br />
      I would need to listen to this DAC, before a purchase, however there are other tempting devices well ahead of the Auralic. To each to their own.
    1. firedog's Avatar
      firedog -
      Eloise-<br />
      <br />
      As both Berkeley units are sold independently of one another and clearly designed to work with non-Berkeley equipment, I don't really see how they are a "package."<br />
      <br />
      Magazine reviewers and other posters have remarked that a quality USB>SPDIF converter has improved the sound of even high-end USB DACs - not to mention bringing the performance of lesser USB DACs to close to state of the art performance. Apparently even in DACs claiming to have high quality USB (and even asynchronus USB) implementations, a good converter does a better job than most USB solutions built into DACs.<br />
      <br />
      In light of that, I'd want to know how much - if any - of the superior sound produced by the Berkeley units is due to the presence of the converter. If it turns out that most of the difference was due to the high quality signal provided by the converter, then that would be a clear sign to lots of users that buying the converter is a cost effective way of getting high end sound with a less expensive DAC(say for $3500 - $4000 for both boxes), without spending $5000 on a DAC alone. Or almost $7000 for the Berkeley "package".<br />
      <br />
      On the other hand, if it was shown that the converter did not account for most of the difference, then the opposite conclusion would suggest itself. That the converter is an add-on for those wanting that last bit of missing SQ, but that the bang for the buck is achieved by investing in the DAC.<br />
      <br />
      My view is that in light of the amounts of money these devices cost, that would be useful information to lots of readers here.
    1. Audio_ELF's Avatar
      Audio_ELF -
      Firedog...<br />
      <br />
      I disagree... I can't see how you can compare a "legacy" (SPDIF) input DAC to a USB / FireWire DAC without including a relevant converter...<br />
      <br />
      Perhaps you could see it as a three way comparison AURALiC vs AURALiC + B-USB vs B-ADAC + B-USB. <br />
      <br />
      If the review was about the B-USB I might agree with you more...<br />
      <br />
      Eloise<br />
      <br />
      PS... Is it me or does the name of the DAC bring thoughts of ... "You want me to listen to music?" ... "No I want you to die Mr Bond!"
    1. dallasjustice's Avatar
      dallasjustice -
      This is great music. I love this album. I am glad you are using it to evaluate gear. I think the better the quality the music, the easier it is to hear differences in all components including DACs.
    1. vortecjr's Avatar
      vortecjr -
      is it fair to say that the Sanctuary chip is to USB what the Dice chip is to Firewire? My understanding is both use an Arm processor in their core....<br />
      <br />
      Jesus R
    1. sq225917's Avatar
      sq225917 -
      Surely it's the embedded code that really counts, not the chip which is just executing instructions?
    1. The Computer Audiophile's Avatar
      The Computer Audiophile -
      Hi awayalways - Thanks for the good questions. I'm sure many others wonder the same things. Here is a bit about my methodology.<br />
      <br />
      I use the Berkeley Alpha DAC on a daily basis unless I have a review DAC in my system. Thus, I spend many many hours with the Alpha every month. I am more familiar with the Alpha than any other piece of equipment in the world. Despite my familiarity with the Alpha I still "reset" my brain before listening to a new DAC like the ARK MX+. I do this by listening to the Alpha for several days before placing the new DAC in my system. Several days can equate to between 10 and 25 hours. <br />
      <br />
      After this, I place the new DAC in the system. (Keep in mind the new DAC has been running in another system without speakers connected so I can follow a manufacturers break-in instructions). Upon first listen I can immediately identify major difference between the DAC and the Alpha. Familiarity with one component and imprinting its sonic signature in one's brain can make it easy to identify difference when something changes. <br />
      <br />
      Over the next few weeks I usually switch between DACs every couple days to get a sense of the sound differences. While I listen to the DAC in for review I write down impressions as I am listening. This allows me to come back to suspected differences at later dates to make sure I still hear what I heard at the time of my writing. <br />
      <br />
      99% of the time I shy away from quick A/B comparisons because they usually hurt rather than help me evaluate differences. This is because human brains use auditory restoration to fill in gaps when sounds change or are not present when they once were. Thus, quick A/B tests for me seem to meld the two component sounds together and produce a sonic signature that is not present in either component by itself. <br />
      <br />
      <br />
      Most of the time I am visually aware of the product I am listening through. It's very hard to device a system that would allow me to not visually know what product I using. So far the only way I've managed to do this is to press the input buttons on the preamp remote several times in random order and hope I don't know what input is actually in use when I stop. Very unscientific but can work in a pinch. This is impossible with DAC that connect directly to a power amp.<br />
      <br />
      <br />
      Thanks again for your questions. Let me know if I can clear anything up.<br />
      <br />
      <br />
      <br />
      <br />
      <br />
      <br />
      <br />
      <br />
    1. Jsmith's Avatar
      Jsmith -
      Just a comment about my experience with the Auralic DAC. I had to update it to a new firmware version for OS 10.8. To do this you actually have to hook up the ARK MX+ to a windows computer, load the windows driver so the PC sees it, and then run the firmware updater to OS 10.8. After that my iMac detected it just fine.