This review page is supported in part by the sponsors whose ad banners are displayed below

Making it play beautiful music
: Yes, after the initial set-up, music will most likely emerge from the speakers but your system will sound much better with careful attention to settings in the operating system, the media player software and the DAC. All these elements can interact in insidious and invidious ways. Making computer digital media foolproof always seems to be somewhat of a work in progress. When one looks carefully into the details—and one must to obtain audiophile-grade playback from a computer source, PC or Mac—there are always issues. For starters, streaming bit-perfect data to a DAC usually requires avoiding automatic sample rate and depth conversion within the operating system or media player. Don’t do it on the computer. Usually. And another more obvious point is to be sure to archive media with a lossless compression audio codec (Apple Lossless, Windows Media Audio Lossless or FLAC). You can also simply rip CDs to uncompressed WAV. Each format has its peculiarities but they can deliver bit-perfect data.


In the case of Windows, the built-in audio 'stack' has elaborate digital signal processing for home theater applications; it works quite well in this context but the S/PDIF stream is not bit-perfect in most cases. For pristine data, a third-party media player is required. There are several products that will accomplish this under Windows Vista and Windows 7. One popular example is the $49.98 J. River Media Center now at version 15. This program taps into Microsoft’s WASAPI (Windows Audio Session Application Programming Interface) which, among other things, allows Media Center to bypass the internal DSP and digital volume control dithering used by Windows Media Player. The J. River Media Center player supports almost every known digital audio file format; a 30-day fully functional free trial version with updates is available.


For Apple Mac users, choices are fewer. Pre-eminent is the $995 Amarra Computer Music Player which integrates tightly—and exclusively—with Apple iTunes. The latest version (1.1 as of this writing) now supports Apple Lossless. An entry-level version (no playback above 96kHz) is still a rather lofty $395. Free trial versions are also available but according to reports, playback is restricted in major ways. Consider that for roughly the price of the Amarra software alone, one could buy a high-end digital media-centric Windows 7 laptop with say an 18.4” screen, BluRay, 500+ GB of storage and the J. River Media Center license. Or one could purchase a $200 network-attached storage (NAS) device with 1.0 TB capacity for media files and pocket the difference.


Amarra is however tested against a long list of audiophile-grade computer DACs. This is a major consideration and should make system integration easier. Device firmware changes from time to time and such qualification is especially relevant for FireWire-based DACs, which have complex drivers. For USB-based devices and their simple native (to the OS) drivers, it probably isn’t. Is the sound quality any better? Whether the audio compression/decompression (codec) software inside Amarra is better than those in the other media players is a subtle question. Reviews are somewhat equivocal on this point so far. The results are certainly very good overall. On the other hand, a major anachronism is Amarra’s forcing the user to install a hardware dongle on a USB port to protect the product against copying. This is a major and unnecessary imposition. Requiring the user to dedicate—that is waste—a USB port for this purpose can be a problem. On laptops in general and Apple MacBooks in particular, there are never enough USB ports and most USB DACs really don’t like hubs. Delusional fan boy enthusiasms aside, keep in mind that all software has issues.


Listening
: How does the Lavry DA-11 perform on a reasonably high-quality playback system? It depends. This DAC can sound extremely good, revealing its recording-industry DNA in some cases but not so well in others. Each of its digital inputs has advantages and disadvantages for the computer audiophile and a specific sonic character to be shortly revealed. The differences relate not only to the essential characteristics of each format but also to the DA-11’s engineering implementation - and of course interconnects.


Once bit-perfect data streams are set up properly, proper cabling is the next critical factor. Bits are most definitely not all the same. After all, they’re simply funny looking analog waveforms that enjoy spewing transients, reflections and radio-frequency (RF) artifacts in all directions.



To the DAC via USB: To illustrate this last point, when driving the DA-11 through Wireworld Cable’s high-end Starlight USB interconnect (shockingly bright fuchsia in color), length in fact did matter. Shorter was considerably better. Ultimately, the best results were obtained with a specially made cable whose  length might best be described as itsy-bitsy. Wireworld’s CEO and chief designer David Salz has even been known to brute-force solder amplifiers and preamplifiers together in order to achieve a zero length cable. This one was close.


When directly connected to the Qosmio via USB, the Lavry had a sonic character that to me typified first-generation attempts at mating DACs with computers in this way. There’s a grainy, gritty and grayish adulteration of the music and an overall feeling of dryness. There’s also limited gradation or nuance in the dynamics. The defects are strongly subtractive in nature. What survives loses most of its underlying substance and is brittle and skeletal in nature. Instrumental timbres lack richness or color. Soundstage width is considerably reduced. From front to back, the presentation collapses into a single plane. Images of artists and instruments are also flattened and two-dimensional. Backup singers congeal in a single mass and reverb in large spaces is crudely quantized to decay in a couple of short steps rather than continuously fade away over several seconds.



Via TOSLINK: The Wireworld Supernova 6 glass fiber optic cable with its 3.5mm jack plugged right into the Qosmio gave some improvement. Soundstage increased in depth and width. There was more body and dimensionality to the images, with better realistic separation of vocalists. Still, the overall rendering was not what it could be.


Via S/PDIF, first pass: At this point and wondering what to do next, my eye fell upon a couple of trusty M-Audio interfaces parked on a shelf. These devices—staples of the prosumer audio world—handily convert FireWire to S/PDIF. Saved! My initial exploration of the S/PDIF inputs used Windows Media Player with its obligatory internal DSP stack and digitally dithered volume control. However, now the sound was approaching that of real music. Instruments of all kinds developed realistic timbres, acoustical spaces expanded considerably and soundstaging was starting to reflect the recording venue. The colorations inherent with Windows Media Player are highly euphonic and easy to take especially compared to the dehydration observed over USB and TOSLINK. The sound is smooth, coherent, somewhat forward but quite acceptable for casual listening. KRK Rokit 8 self-powered studio monitors would be very happy with such a feed. Progress was being made but we weren't there yet.


Once again via S/PDIF
: About the time of the M-Audio revelation, Norbert Lindemann, CEO and lead designer for Lindemann Audiotechnik, sent over an updated version of his $650 USB-DDC 24-96 D/D converter module. This tiny (really tiny) "little machine" in his terminology takes USB and transforms the bitstream into S/PDIF over RCA coaxial or TOSLINK interconnect. Mirabile dictum! Everything was immediately transformed and immeasurably improved. Initially, transients were hyper realistic with enormously exaggerated over-the-top edginess but the rest—the timbres, soundstaging and instrumental images—were robust, coherent and colorful.


After quickly running through the available lengths of USB cable—thank goodness Wireworld had sent me several—it became clear that longer, about 2.0m, was best in this case. A 1.0m length of Audioquest Eagle Eye connected the USB-DDC to the DA-11’s RCA S/PDIF input.


The soundstage now extended way outside the speakers, apparently limited only by the lateral room dimensions (my Usher Be-718 monitors are about 6 feet apart in a 12' wide and 18' deep room). Performing artists poured out their hearts from the middle of my room as though through a phantom center speaker. In the case of female vocalists, nuances of sound radiating from the mouth, throat and chest seemed to be almost visible. That fundamental element of musical enjoyment—emotional involvement—came through in abundance. Music was made by human beings, not avatars. The NCA Biber Missa Salisburgensis, a monumental and magnificent piece written in 1682 specifically for the Salzburg Cathedral, celebrates the 1,100th anniversary of the archdiocese’s founding in 582 A.D. Exploiting the architectural features of the building, Biber placed arrays of opposing choruses throughout the structure. It’s a fascinating piece—medieval and modern at the same time—that showed off the timbre fidelity and soundstaging excellence of the DA-11. Similarly the Naxos Samuel Sebastian Wesley Anthems is one of the best-recorded choral performances available. The clarity, purity and naturalness of real choral music in a real venue is astonishing.


In contrast, the XRCD remastering of the famous Proprius Now the Green Blade Riseth showed off all kinds of 'improvements' that would be invisible on DACs less transparent to the source. This mix seems designed for car radios, with way too much highlighting of soloists, chorus and instruments, each of which seems to exist by itself  outside of a coherent acoustical surround. Most of these XRCDs sound rather strange to me. They grind and drag their way along and are seldom improvements over the originals.


Spatially and on the best recordings however, if one walked towards and through the plane of the speakers, it was as though sonic images were real freestanding entities right there in the room. The sense of air between performers imposed itself on my listening space. Soundstage width and depth were limited by the recording engineering and venue. The number of voices in choruses doubled or trebled and reverberation in large halls faded away gradually and smoothly, with subtle changes in timbre.


In the Reference Recordings Mozart Piano Concerto No. 24, the piano radiated—perhaps coruscated would be a better word—showers of acoustical color which, rather than being congealed into a uniform sphere, varied considerably depending upon what part of the instrument created them. Massed strings of diverse kinds came across as the fragile wood-and-varnish-and-horsehair contraptions they are. What was missing as far as I could tell was that final refinement of spatial ambiance, namely the sparkle that one feels or—speaking of coruscations—almost sees in the concert hall or club. There’s a charge in the air which some of the best gear can capture. What the Lavry DA-11 does in exchange is to loft images against an extremely quiet background. The fundamental rendering of acoustic space is excellent and the only thing missing is making the air come alive. Bass is very deep, well-defined, taut and controlled; transient impacts can rattle the fillings in one’s teeth. High-end treble is delicate, life-like, spatially detailed, convincing and almost magical. Everything in-between seems equally authentic.


James Taylor on Hourglass is somewhat nasal and occasionally flat but each track reveals subtle details of miking, subtle instrumental effects and the unusually complex vocal backups. You’ll hear more musical reality than you ever thought possible. Take a look at the FTT noise spectrum from the DA-11 especially in the 1kHz to 4kHz band. This kind of performance is certainly part of the picture as to why the Lavry sounds as well as it does. Note that the blue trace shows the noise at full-scale recording-industry standard, i.e. high output levels. The green traces represents full-scale for consumer gear and the data are even better.

Back to the TOSLINK: Almost as an afterthought, it occurred to me to try out an Audioquest OptiLink-5 fiber optic cable between the Lindemann USB-DDC and DA-11. What astonished me was how closely the TOSLINK could duplicate the qualities of S/PDIF. Only on a high-end system and with careful listening would it be possible to differentiate the two. So far about the only way to describe the difference is that there was a slight feeling of deflation.

It was as though the soundstage and vitality of the performance were just a little bit reduced, less life-like, not as energetic. However just as the Lindemann demonstrated unequivocally that the Lavry DA-11 can make beautiful music with USB technology in the loop, it also proved that glass optical fiber is now in the running for serious use for the most demanding computer-audiophile applications.


Live and learn!

Summary
: The Lavry DA-11 is clearly a work in progress with respect to computer-driven audiophile applications. It does many things extremely well and with a refresh of the USB circuitry, could be a highly competitive and possibly category-leading product. It’s already very good. And what might happen if some of Lavry’s state-of-the-art DAC technology from their ultra-high-end recording-industry gear were added to the mix? The other day, Dan and Priscilla Lavry came over to my house to hear how their creation sounded "in the wild” (of Maui). After listening to the system for a while and going over my observations, Dan said with a twinkle in his eye, “But have you heard it through the AES-EBU inputs?” Well, not yet. I took the hint and will be trying it out shortly, this time using a high-end FireWire to S/PDIF and AES-EBU converter along with state-of-the-art optical fiber that comes with its own transmitter and receiver modules. The results should be quite interesting.


Strong points
• Continuous, smooth, colorful and nuanced sound over S/PDIF and TOSLINK.
• Accurate and rich full-bodied timbres; natural, convincing imaging.
• Transparency that reveals every detail of recording and mastering.
• Absolute polarity control.
• Soundstage wide as the Pacific Ocean, almost as deep.
• Universal power supply built-in, no wall warts.
• Likely to last a very long time.
  Weak points
• Doesn’t realize the potential of current USB technology.
• Unusual user interface design and operation.
• Lack of RCA unbalanced analog outputs; middling RCA S/PDIF connector.
• No custom remote control.


Addendum: And finally via FireWire and AES-EBU with feeling.
Shortly after the Lavrys' departure, the latest version of the Weiss INT202 FireWire Interface ($1300 or $1600 with remote) arrived. This device converts the FireWire data stream to AES-EBU or S/PDIF. The INT 202 (review upcoming) includes a hybrid analog/digital attenuator and an absolute polarity switch both on the minimalist remote control. In this case the Lavry DA-11 was set to no attenuation (56 on its digital volume control) so that the INT 202 was in charge of playback levels.


Although I had contentedly reflected on how well everything had sounded thus far, the first moments of listening with the INT 202 made it clear that more was to come. If my listening experience now could be summed up in a word, it would be effortless, meaning essentially an across-the-board minimization of strain or restriction in any of the listening parameters we use to evaluate recorded music. One could relax into the music and let go rather than being hung up on shortcomings.


For starters, via FireWire and the INT 202, noise levels and artifacts of all stripes were fundamentally lower. I used Audioquest Raven and Eagle Eye (very similar) digital interconnects on the AES-EBU and S/PDIF outputs respectively. As one might expect, the AES output was fresher, a bit more dynamic and slightly quieter than the S/PDIF. Its sound was an extension of the goodness conveyed by S/PDIF. Yet another thin layer of digital haze was lifted and now things began to get really interesting.


The soundstage of the recording venue imposed itself even more tangibly on my listening room. Voices and instruments arose from a dark or rather invisible and neutral sonic background. The volume of the hall, studio or isolation booth hung in space between and beyond the speakers with strong center fill. On multi-track extravaganzas such as the Sheffield Pop Experience, separately recorded individual instruments or groups of performers dropped into the mix each had its own soundstage and timbre within the overall acoustical scene created by the engineers. 'Fix it in the mix' doesn’t work in this age of high-resolution DACs.


Overall spaciousness, while maintaining clarity and focus, increased considerably. Reverberations were even more refined than before and continuously connected to the performers. They faded away smoothly and convincingly without any unnatural gradations. New details appeared out of nowhere but without calling attention to themselves and the timbre was even more colorful and organic. The gentle tapping of a foot, a sotto voce “One, two, three…” and turning the pages of the score helped made it all very real - including bring out a few details that one suspects the recording engineers would prefer home audiences didn’t hear. Resonances between plucked harp strings suggested gaiety, frivolity and otherworldliness—perhaps a glimpse into the intersection of the secular and celestial?
Lavry Engineering website
Enlarge!