Most-Read Reviews (Last 5 Years)
- 2013-04-15 - KEF LS50 Loudspeakers
- 2014-12-01 - Sonus Faber Olympica III Loudspeakers
- 2012-08-01 - KEF R500 Loudspeakers
- 2011-02-01 - Bowers & Wilkins 803 Diamond Loudspeakers
- 2014-12-15 - KEF Reference 1 Loudspeakers
- 2010-10-01 - Bowers & Wilkins CM5 Loudspeakers
- 2011-03-01 - Hegel Music Systems H20 Stereo Amplifier
- 2013-09-01 - Tannoy Definition DC10A Loudspeakers
- 2012-03-01 - Monitor Audio Gold GX100 Loudspeakers
- 2011-09-15 - Paradigm Atom Monitor v.7 Loudspeakers
Most-Read Reviews (Last 365 Days)
- 2017-04-01 - KEF Reference 3 Loudspeakers
- 2017-04-15 - MartinLogan Masterpiece Classic ESL 9 Loudspeakers
- 2017-03-01 - Audio Research Foundation LS28 Preamplifier
- 2017-01-01 - Hegel Music Systems Mohican CD Player
- 2016-12-15 - PS Audio BHK Signature Preamplifier
- 2017-02-01 - Audio Research G Series GS150 Stereo Amplifier
- 2017-03-15 - PS Audio BHK Signature 300 Mono Amplifiers
- 2016-12-01 - Audio Research G Series GSPre Preamplifier
- 2017-02-15 - Exogal Ion PowerDAC
- 2017-05-15 - Devialet Expert 130 Pro DAC-Integrated Amplifier
Most-Read Reviews (Last 90 Days)
- 2017-10-15 - Devialet Gold Phantom Loudspeakers
- 2017-09-01 - Hegel Music Systems Röst DAC-Integrated Amplifier
- 2017-10-01 - NAD C 338 Hybrid Digital Wireless Streaming DAC-Integrated Amplifier
- 2017-09-15 - Bryston BCD-3 CD Player
- 2017-11-01 - GoldenEar Technology Triton Reference Loudspeakers
- 2017-11-15 - Markaudio-Sota Cesti T Loudspeakers
- Written by S. Andrea Sundaram S. Andrea Sundaram
- Category: Full-Length Equipment Reviews Full-Length Equipment Reviews
- Created: 01 July 2012 01 July 2012
These days, you can’t swing a boutique USB cable without hitting a multi-thousand-dollar digital-to-analog converter, each of which is based on one of a relatively small number of DAC chips. Some people point to that fact to suggest that there aren’t any differences among DACs, but the implementation of a particular chip matters just as much as -- or more than -- the chip itself. The Invicta DAC ($3995 USD), the first product from Resonessence Labs, is based on the well-regarded ESS Technologies Sabre 9018 Reference DAC chip. Resonessence Labs’ website notes that some of the engineers who worked on the Invicta were also involved in developing the Sabre chips. The clear implication is that they should know best how to maximize the Sabres’ performance.
The technical discussion of the Invicta’s design and specifications on the company’s website and in the user manual is the most comprehensive I’ve seen for any product. I’ll only summarize some of the main points here:
The digital volume control operates in the 32-bit domain, so there is no resolution loss with 24-bit sources until you reach 48dB of attenuation. The eight-channel ES9018 chip is run in quad-differential mode to feed both the XLR and RCA outputs, while a separate ES9016 chip feeds the headphone outputs. This arrangement allows independent volume levels to be set for the line and headphones. A shielded toroidal transformer supplies the Invicta’s power, with separate regulation for each part of the circuit. The analog, digital, computer-interface, and control sections are galvanically isolated from each other to prevent any interference. The Invicta employs no phase-locked loops (PLLs). Instead, the incoming digital data are buffered and reclocked at the DAC chip. That should make the Invicta largely immune to jitter on the incoming datastream.
Form and function
The Invicta is a black-anodized aluminum box measuring 8.75"W x 1.75"H x 10.5"D. It’s not flashy audio jewelry, but it’s beautifully put together and feels rock solid. From left to right across the front panel are: the Invicta logo, two 1/4” headphone jacks (with an activation button above each), the central display, four buttons below it, a group of six lights to indicate the incoming sample rate (44.1, 48, 88.2, 96, 176.4, and 192kHz), an SD card slot, and a rotary Volume/Function Select knob that can also be pressed to toggle between Play and Pause. The back bristles with inputs and outputs, including: an IEC power inlet, analog out on both XLR and RCA connectors, AES/EBU in, TosLink in/out, USB, HDMI, and three BNCs. These last two items are configurable via firmware. As supplied, two BNC connectors are for S/PDIF inputs, and the third is for word-clock synchronization. Using the Invicta with the coaxial outputs typically found on consumer equipment will require BNC-to-RCA converters. The HDMI port currently functions as an output to display navigation and track information when playing files from an SD card. Alternatively, it could be configured to accept I2S or HDMI audio.
All inputs support 24-bit PCM data with sample rates up to 192kHz, and the USB connection operates asynchronously (when using Windows, installation of a driver is required for sample rates higher than 96kHz). The desired input is selected from a menu, and the input to which the Invicta will default on startup can be programmed. The volume for all outputs is controlled digitally in 0.5dB steps. Ideally, you’ll use the Invicta’s maximum output of 4.6V balanced or 2.3V unbalanced directly into a power amplifier. If using a preamplifier, you can save a power-on volume level that is most appropriate for your equipment. The volume levels move together when the control knob is turned, but you can assign offset values separately for each of the headphone jacks. Plugging in a set of ’phones doesn’t mute the line outputs, nor is there a control for doing so. Menu navigation and volume adjustment can be accomplished either through the front-panel buttons and rotary dial or with the included remote control. The small print of the menus is unlikely to be legible from the listening seat, but once music starts playing, the display is almost entirely taken up by a clear, bright volume indicator (dimmable in ten steps from the settings menu). When the Invicta is in USB mode, the remote and front-panel buttons will also control play/pause and track forward/reverse with most media players.
The SD card option is less common than the other forms of input, and so bears some explanation. The slot supports SD cards of up to 2GB capacity, and SDHC cards up to 32GB. Files can have sample rates of up to 192kHz, but must be in the WAV or AIFF formats. No compression is allowed -- not even lossless. That’s no problem for audio professionals who might want to quickly check a field recording, but audiophiles who store their music in FLAC or ALAC can’t simply drag and drop. In the Invicta’s current firmware, you navigate the SD card by its directory structure. Since you won’t be trying to sort through many hundreds of gigabytes’ worth of music, this is still a workable option. If you’ve included ID tags and cover art in your files, the song titles and artwork will be displayed on a connected monitor once the track begins playing. (I wasn’t previously aware that this tagging was possible with WAV, but a few programs allow for it.) You’ll also see a bar that tracks your progress through the file. Once a folder has begun playing, the system will move sequentially through the tracks in that folder, and stop when it reaches the end. Resonessence has plans to expand the Invicta’s sorting and navigation functionality, and to include support of SDXC cards -- both of which may make the SD feature more compelling. In the meantime, one can hardly complain that they’ve skimped on the other connections.
If you’re using a Mac, the Invicta should recognize that and automatically switch to USB 2.0 mode, thus allowing sample rates of up to 192kHz. Those who, like me, use Windows need to manually choose USB 2.0 in the Invicta’s settings and install the provided driver in order to enable sample rates above 96kHz. The procedure took me all of two minutes. I used foobar2000 and JRiver Media Center in ASIO, WASAPI Exclusive, and Kernel Streaming modes -- all of which bypass the Windows mixer -- and successfully played files at all standard sample rates with no problems. I also fed the Invicta CD data over an AES connection from my Ayre Acoustics C-5xeMP universal disc player, and experimented with playing files from an SD card. Any differences between the three input options were very small -- with, perhaps, a slightly drier ambience from SD. I also tried feeding digital data from my Harman Kardon CDR 25 CD player/recorder through a 12’ length of generic plastic optical cable. That last setup produced a rather loose, unfocused sound, suggesting that the Invicta is not completely indifferent to the quality of the incoming datastream.
The Invicta’s two digital reconstruction filters -- switchable with a single button push on the remote -- offered markedly different sounds. Miles Davis’s Harmon-muted trumpet in the title track of his ’Round About Midnight (CD, Columbia CK 40610) had more bite with the Fast-rolloff filter setting, and was a bit darker, with more of the fundamental pitch of each note, when set to Slow rolloff. Similarly, I heard a bit more body in the tone of Itzhak Perlman’s violin -- on his live performance of sonatas by Beethoven and Franck, with pianist Martha Argerich, from the Saratoga Music Festival (CD, EMI 5 56815 2) -- when I selected Slow, and a little more emphasis on the strings with the filter set to Fast. Since these filters should affect frequencies only near the top of the passband, I was curious whether I would hear any difference with high-resolution recordings. Ella Fitzgerald and Louis Armstrong’s Ella and Louis (24/96 FLAC, Verve/HDtracks) has only very little frequency content above 20kHz, so I didn’t expect a Fast or Slow rolloff to make any difference in the timbre of Armstrong’s trumpet. But not only did his instrument gain a little weight, his voice was also a shade deeper with Slow. The slightly deeper, darker sound of the Slow filter setting was consistent in both direction and degree across all of the music I played through the Invicta.
I was surprised to find that filter selection also changed the character of the bass. The Fast filter was a little drier, tighter, and leaner, while Slow was broader, with seemingly greater extension. Ray Brown’s double bass on the Oscar Peterson Trio’s Night Train (24/96 FLAC, Verve/HDtracks) had a more articulated attack with the Fast setting, and seemed a little larger with Slow. The kick drum in “Run like Hell,” from Pink Floyd’s The Wall (CD, EMI 8 81243 2), exhibited a bit more punch with Fast, and more weight with Slow. With jazz and rock, either filter resulted in perfectly acceptable sound; I didn’t have a strong preference. With orchestral music, on the other hand, I appreciated the grander sound the Slow filter gave to timpani -- as in the Minnesota Orchestra’s recording of Sibelius’s Symphonies Nos. 2 and 5 under the baton of Osmo Vänskä (24/96 FLAC, BIS/e classical). Whichever filter I selected, the bass was always tuneful and firmly in control.
As great as the Invicta sounded with high-quality recordings, it still made less-than-stellar recordings sound good. The Slow filter removed some of the sheen from overly hot modern rock, such as David Gilmour’s On an Island (CD, Columbia 80280), and the harsh grating from bad digital transfers of analog recordings -- such as Sade’s Diamond Life (CD, Epic EK 85240). It also reduced the stridency of the brass and violins in some older orchestral recordings. Here, I’m not talking about the rolloff of some soft-dome tweeters or the warmth characteristic of a single-ended tube amplifier. The effect of the Invicta’s Slow filter was much more subtle. Instrumental and vocal timbres were entirely believable, but the sound was more relaxed than that of many other high-end DACs.
The Invicta’s Slow rolloff filter took a step back, smoothing out the rough edges present on a great many recordings. Even the wider-bandwidth Fast filter managed to avoid the edginess that many listeners have come to accept as “digital sound” over the past several decades. But my descriptors smooth and relaxed shouldn’t be taken to mean sloppy or imprecise. The Invicta’s sound was anything but. Notes started and stopped in perfect time, and transients were quick and clean.
The Invicta’s precision was best exemplified in its presentation of spatial information. Both filters produced tightly focused images -- instruments and singers were placed precisely and unwaveringly from left to right across the soundstage with the Fast filter, those images beginning just behind the speaker plane but varying only moderately in depth. With Slow, the front of the soundstage was only slightly farther back than with Fast, but the depth was greatly expanded -- I could more easily discriminate each sound source’s position from front to back. The reverberant field around each instrument was better defined, and with recordings made in natural environments, those reverberant fields blended seamlessly into each other to create a coherent whole.
For example, on Bucky Pizzarelli’s Swing Live (24/96 FLAC, Chesky/HDtracks), the space between and just to the sides of my speakers was transformed into the slightly dry acoustic of a small jazz club. In this setting, the distance between performers isn’t huge, but I still heard the vibraphone upfront and left, the guitar on the right, the clarinet centered and slightly farther back, and the drums and bass behind all three. The Norwegian Armed Forces Staff Band required a much larger concert hall for their recording of Eugène Bozza’s Children’s Overture (24/176.4 FLAC, 2L/SoundStageRecordings.com), and the Invicta made the difference obvious. The trumpets were far back, and the sound of the French horns reverberated off walls that were still farther off, yet clearly defined.
With the Consortium Vocale Oslo’s performance of Crux Fidelis (24/176.4 FLAC, 2L/SoundStageRecordings.com), the Invicta enrobed the voices in the reverberant wash of the church acoustic. But as well as it conveyed the blending of those voices with each other and that large space, the Invicta still portrayed the choir as a collection of individual voices rather than an amorphous blob.
Perhaps the greatest surprise was the amount of spatial information the Invicta was able to recover from CDs. Though they couldn’t quite compete with the best high-resolution files, the better CDs in my collection came much closer to the sound of instruments playing in real spaces than I’d previously thought possible from the medium.
Using the DAC in my Grace Design m902 headphone amplifier, I do some computer-based listening through headphones at my desk, but the primary digital source in my main system is the Ayre Acoustics C-5xeMP. I connected the balanced outputs of the Ayre and Invicta to the balanced inputs of my GRAAF GM 50 integrated amplifier with matching pairs of Nordost Red Dawn LS interconnects. Matching levels with a 1kHz test tone put the Resonessence’s digital volume between -0.5 and -1dB, so I alternated to average out any level-based bias. For comparing recordings on CD, the Ayre provided data to the Invicta through its AES connection. For higher-rez material, the C-5xeMP played DVD-Audio discs that I’d burned, while I used the Invicta’s USB or SD card options. (My Ayre was refusing to output sample rates higher than 48kHz, regardless of the setting of its rear-panel DIP switch.) The differences between high-quality digital front ends are small, but it’s just those small differences that drive this hobby/obsession of ours.
The C-5xeMP casts a wider soundstage than the Invicta, but the outlines of instruments aren’t as clearly drawn -- each sound source is spread over a larger area. Those images float more freely in space through the C-5xeMP, but are less integrated into the specific ambience of the recording venue. That difference makes the Ayre better at placing the performance in my listening room, but the Invicta was better at delivering a window on the space in which the recording was made, whether studio or concert hall. The Ayre conveys a greater sense of depth and relative positions along that axis than the Invicta’s Fast filter, but can’t match the Invicta’s Slow filter for overall depth, specific placement of instruments, or the way the reverberations of those instruments trail off into that space.
The C-5xeMP isn’t edgy -- some even call its sound laid-back -- but it’s not as smooth as the Invicta. If a recording sounds harsh or brittle through most systems, it’s likely to sound that way through the Ayre; the Invicta, and particularly its Slow filter, treated such material more kindly. In many cases -- maybe most -- that smoother sound made music more enjoyable. Sometimes, though, I felt the Invicta’s sound was a little too smooth.
Wynton Marsalis and Willie Nelson’s Two Men with the Blues (CD, EMI 5 04454 2) has great musical merit, but I find it a bit too forward through the C-5xeMP, which keeps me from listening to the album straight through. I had no such concerns with the Invicta. But while Nelson’s voice was more present, and displayed more of its familiar nasal quality through the Ayre, that character was slightly muted by the Invicta’s Slow filter, as if Nelson had taken a step back from the microphone or was coming down with a slight cold. Arto Noras and Bruno Rigutto’s recording of cello sonatas by Fauré, Franck, and Debussy (CD, Apex 0927 40599 2) is a more natural recording than the Marsalis-Nelson. Both components produced a great cello sound, but the Ayre gave more of the scrape of bow against strings. Switching to the Invicta’s Fast filter restored some of that texture, but reduced the ambience around each performer.
The C-5xeMP also sounded a bit livelier than the Invicta. In the second movement of the Franck, the right-hand notes in the piano had a bell-like quality through the Ayre, and were a little more dense and damped through the Resonessence. I heard the same difference with classical pianist Hélène Grimaud’s Resonances (24/96 FLAC, Deutsche Grammophon/HDtracks). All of the mandolin plucking on Sam Bush and David Grisman’s Hold On, We’re Strummin’ (24/88 FLAC, Acoustic Disc/HDtracks) was fast, clean, and clear through the Invicta, but each note popped more through the C-5xeMP.
Most readers seem to want reviewers to pick winners and losers. In terms of soundstage focus and the resolution of low-level spatial information, the prize goes to the Invicta. But in other aspects of performance there was no obvious victor. Some listeners will prefer the more ebullient sound of the C-5xeMP; others will be drawn to the smooth, nonfatiguing character of the Invicta.
The headphone jacks on most components are usually afterthoughts, when they’re included at all, but Resonessence Labs has given them as much attention as the Invicta’s line outputs. One reason I was assigned this review is that I do a lot of listening through high-quality headphones, and so was best positioned among our writers to evaluate this feature.
With a maximum voltage output of greater than 5V and power output of greater than 250mW, the Invicta should be capable of driving any reasonable pair of headphones to levels that will cause hearing damage. In practice, it had no trouble making the rather obstinate HiFiMAN HE-500s come alive. The Invicta’s essentially nonexistent noise floor meant that it was equally compatible with highly efficient ’phones and even in-ear monitors. Bass was weighty, highs were extended, and the midrange was clean and even. Resolution of low-level detail was truly astonishing. On the Perlman-Argerich recording, I heard not only every inflection of Perlman’s bow, but also very low-level audience noises, such as stifled coughs, that I didn’t recall having heard before. Returning to my Grace Design m902 headphone amplifier-DAC, the noises were there, just not as startlingly clear. While the Grace ceded resolution to the Invicta, it countered with instrumental and vocal timbres that were a bit more fleshed out. You may wish a different flavor to your headphone sound than the Invicta provides, but by objective measures of quality, its headphone outputs are first rate.
I began this review by remarking on the many choices now available in a high-end DAC. One way a product can differentiate itself from the pack is by its feature set, and the Resonessence Labs Invicta has all the inputs we’ve come to expect to see in such a device. It exceeds those expectations with the inclusion of the SD card option, which may appeal to those who don’t want a computer in the listening room. The 32-bit volume control means that, in a digital-only system, the added noise and cost of a separate preamplifier can be dispensed with. For headphone listeners, the Invicta’s high-quality outputs can allow you to sidestep both the expense of a dedicated headphone amplifier and the dilemma of how best to incorporate it into the rest of the system. The Invicta’s build quality, inside and out, is at least commensurate with its price of $3995, if not above.
The most important measure of any audio component is the way it sounds. Depending on the filter selected, the Invicta sounds either dead neutral or very slightly dark. The Fast filter creates very credible images, but the Slow filter conveys the acoustic environment of high-quality recordings to a shocking degree. What these filters share is an absolute freedom from grit, grain, and glare, which they accomplish by smoothing out rather than covering up the sharp edges. The result is a sound that is very clean and hi-rez without ever feeling forced. More than any of its other virtues, it is this smooth, relaxed character that sets the Invicta apart from its competition, and that I believe will win it a large following.
. . . S. Andrea Sundaram
- Digital source -- Ayre Acoustics C-5xeMP universal disc player
- Integrated amplifier -- Graaf GM-50
- Speakers -- Esoteric MG-10
- Computer -- custom laptop running Windows Vista, foobar2000, JRiver Media Center
- Headphones -- Stax SR-507, Ultrasone Pro 2900, HiFiMan HE-500, Beyerdynamic T70, Etymotic ER-4PT
- Headphone amplifiers -- Woo Audio GES, Grace Design m902
- Interconnects -- Nordost Red Dawn LS, QED Silver Spiral
- Speaker cables -- DH Labs Q-10
- Power conditioner -- Equi=Tech Son of Q
Resonessence Labs Invicta Digital-to-Analog Converter
Price: $3995 USD.
Warranty: Three years parts and labor.
863 Coronado Crescent
Kelowna, British Columbia V1W 2K3
Phone: (778) 477-5536