EMM Labs DV2 D/A processor Measurements

Sidebar 3: Measurements

I measured the EMM Labs DV2 using my Audio Precision SYS2722 system (see the January 2008 "As We See It"), using the Audio Precision's AES/EBU digital output—unlike JVS, I had no problems with this input—and USB data sourced from my MacBook Pro running on battery power, with Pure Music 3.0 playing WAV and AIFF test-tone files.

Apple's USB Prober utility identified the EMM processor as "DV2 Reference DAC with MQA Decoder" from "EMM Labs," with the serial number "LAB0012." The DV2's USB port operated in the optimal isochronous asynchronous mode. Apple's AudioMIDI utility revealed that, via USB, the DV2 accepted 16- and 24-bit integer data sampled at all rates from 32 to 384kHz. The AES/EBU and coaxial and TosLink S/PDIF inputs all locked to datastreams with sample rates of up to 192kHz.

The volume control operated in accurate 0.5dB steps. With the volume control set to its maximum of "100," a 1kHz digital signal at 0dBFS resulted in a balanced output level of 4.26V into 100k ohms with the output level set to Low, and 7.25V with it set to High. (Except where indicated, all measurements were taken at the High setting.) The maximum unbalanced output levels were half the balanced levels, as expected. The output impedance was the same from both sets of analog outputs, at a moderately low 297 ohms from 20Hz to 20kHz, and both outputs preserved absolute polarity (ie, were non-inverting).

The DV2's impulse response with 44.1kHz data (fig.1) indicates that the reconstruction filter with this signal is a very short linear-phase type. With 44.1kHz-sampled white noise (footnote 1, fig.2, red and magenta traces), the DV2's response rolled off gently above 20kHz and didn't reach full stop-band suppression until 44.1kHz. Peculiarly, given this slow rolloff, the aliased image at 25kHz of a full-scale tone at 19.1kHz (blue and cyan traces) was still suppressed by more than 110dB. (Perhaps this was due to the DV2's reconstruction filter changing its behavior with this signal.) The second and third harmonics of the 19.1kHz tone are visible above the ultrasonic noise floor, though both are low in level.


Fig.1 EMM DV2, impulse response (one sample at 0dBFS, 44.1kHz sampling, 4ms time window).


Fig.2 EMM DV2, wideband spectrum of white noise at –4dBFS (left channel red, right magenta) and 19.1kHz tone at 0dBFS (left blue, right cyan), with data sampled at 44.1kHz (20dB/vertical div.).

When I examined the DV2's frequency response with data at 44.1, 96, 192, and 384kHz, its 44.1kHz output rolled off sharply above 20kHz (fig.3, gray and green traces). The responses at the three higher sample rates followed the same shape, flat to 20kHz, with then a slow rolloff, disturbed by a steeper drop at each Nyquist frequency. With data sampled at 384kHz, the DV2's output was down by 11dB at 120kHz. Channel separation was superb, at >120dB in both directions below 3kHz, and still 112dB at 20kHz (fig.4). The low-frequency noise floor was both low in level and commendably free from any power-supply–related artifacts (fig.5).


Fig.3 EMM DV2, frequency response at –12dBFS into 100k ohms with data sampled at: 44.1kHz (left channel green, right gray), 96kHz (left cyan, right magenta), 192kHz (left blue, right red), 384kHz (left blue, right green) (1dB/vertical div.).


Fig.4 EMM DV2, Channel separation (5dB/vertical div.)


Fig.5 EMM DV2, spectrum (0Hz–1kHz) of dithered 1kHz tone at 0dBFS (20dB/vertical div.).

When I increased the bit depth from 16 to 24 with a dithered 1kHz tone at –90dBFS (fig.6), the noise floor dropped by 25dB, meaning that the DV2 offers 20 bits' worth of resolution, which is close to the state of the art. With undithered data representing a tone at exactly –90.31dBFS (fig.7), the three DC voltage levels described by the data were well resolved and the waveform was perfectly symmetrical. With undithered 24-bit data, the result was a superbly clean sinewave (fig.8).


Fig.6 EMM DV2, spectrum with noise and spuriae of dithered 1kHz tone at –90dBFS with: 16-bit data (left channel cyan, right magenta), 24-bit data (left blue, right red) (20dB/vertical div.).


Fig.7 EMM DV2, waveform of undithered 1kHz sinewave at –90.31dBFS, 16-bit data (left channel blue, right red).


Fig.8 EMM DV2, waveform of undithered 1kHz sinewave at –90.31dBFS, 24-bit data (left channel blue, right red).

Harmonic distortion was extremely low in level even into 600 ohms (fig.9), with the second harmonic the highest in level in the right channel (red trace), at –106dB (0.005%), and the third harmonic the highest in the left channel (blue), at –112dB (0.0025%). Intermodulation distortion was very low, with the second-order difference product produced by equal-level tones at 19 and 20kHz with the combined waveform peaking at 0dBFS, lying just below –90dB (0.003%) (fig.10). However, contradicting the spectra shown in fig.2 are several aliasing products visible in fig.10. Reducing the level of the test signal to –3dBFS didn't change this behavior.


Fig.9 EMM DV2, spectrum of 50Hz sinewave, DC–1kHz, at 0dBFS into 600 ohms (left channel blue, right red; linear frequency scale).


Fig.10 EMM DV2, HF intermodulation spectrum, DC–30kHz, 19+20kHz at 0dBFS into 100k ohms, 44.1kHz data (left channel blue, right red; linear frequency scale).

Tested for its rejection of word-clock jitter with 16-bit AES/EBU data, the DV2 turned in superb performance: all odd-order harmonics of the LSB-level, low-frequency squarewave were at the correct levels (fig.11, sloping green line), though with some broadening of the peak that represents the high-level tone at one-quarter the sample rate. This is presumably due to the presence of low-level, low-frequency random jitter components. There is also a pair of low-level, power-supply–related sidebands at ±120Hz. These sidebands and the spectral broadening were also present with 24-bit J-Test data (fig.12).


Fig.11 EMM DV2, high-resolution jitter spectrum of analog output signal, 11.025kHz at –6dBFS, sampled at 44.1kHz with LSB toggled at 229Hz: 16-bit AES/EBU data (left channel blue, right red). Center frequency of trace, 11.025kHz; frequency range, ±3.5kHz.


Fig.12 EMM DV2, high-resolution jitter spectrum of analog output signal, 11.025kHz at –6dBFS, sampled at 44.1kHz with LSB toggled at 229Hz: 24-bit AES/EBU data (left channel blue, right red). Center frequency of trace, 11.025kHz; frequency range, ±3.5kHz.

Overall, EMM Labs' DV2 offers state-of-the-art measured performance.—John Atkinson

Footnote 1: My thanks to Jürgen Reis of MBL for suggesting this test to me.
EMM Labs Inc.
115-5065 13th Street SE
Calgary, Alberta T2G 5M8,
(403) 225-4161

tonykaz's picture

"If the DV2 is music to your ears" & "if it's within your price range" then don't hesitate to audition.

Wouldn't an Audition be the first thing needed to determine if it's Music to your ears?, of course it's within owning, why else bother looking at it?

EMM Labs needs a John Darko type of explainer. ( of course they fully understand that fact )

Tony in Michigan

Ortofan's picture

... spend $30,000 on this device when a Benchmark DAC3 HGC provides "state-of-the-art" performance for a mere $2,195.

tonykaz's picture

Every DAC out there is State of the Art,
every gear out there should be auditioned,
every gear can take your breath away,
every gear is transparent
every gear delivers only slight variations in performance,
every gear should be considered if budget....

Every review seems to contain the same Conditional Endorsements.

Well, Mr.Ortofan, does your Benchmark DAC3 HGC do SACD?

Tony in Michigan

Ortofan's picture

... convert DSD data, then the answer is yes.

tonykaz's picture

I suppose there are some folks that invested in SACD & DSD technology enhanced Format Music recordings.

Being an "Early Adopter" is a pricy experience, isn't it?

I auditioned DACs back in 2015 without noticing any significant musical differences. I used Sennheiser HD600 Headphones which are finer transducers that any Loudspeaker I've every experienced.

For my RedBook needs, DACs sound rather similar with AB comparisons.

Tube Rolling is far more interesting and musically fulfilling.

Matching Loudspeaker Transducers & Amps can bring scintillating thrills.

I certainly will "Audition "more DACs "IF" I can hear some important difference, I just haven't heard anything wonderful, yet!

Tony in Michigan

ps. I'm still having a hard time understanding or justifying any gear beyond Schiit or PS Audio price levels. Is there anyone out there that can justify $100,000 Plus price levels for a Home Audio System?

MZKM's picture

Every DAC is not transparent, especially many that do MQA, as every one I’ve seen uses the same filter for PCM, which degrades PCM performance, which may be one reason why people say they hear an improvement with MQA, as the DAC they are using to test likely isn’t doing PCM justice.

tonykaz's picture

Is MQA still being promoted ?

It seemed like Format Attrition had already set in.

So, I hadn't considered a possible degradation from that "intentional" cause.

Thanks for straightening it out.

Tony in Michigan

andy yuen's picture


Thank you for your great write-up. I owned an EMM Labs DA2 and an EMM Labs TSDX transport and I found the DA2 sounded best thru the EMM OptiLink, playing SACDs. Can you test that option and let us know the results?

Jason Victor Serinus's picture

Hi Andy,

Thanks for the strokes.

Stereophile's policy is to review only piece of equipment at a time, and to keep everything else constant in our reference systems. Otherwise, we can never tell for certain what is creating the sound we're hearing. Without the EMM OptiLink or an EMM transport, I'm afraid I cannot perform this comparison.


ejlif's picture

but how can the sound of the EMM DAC here even be considered when you have so many variables and factors involved? I have been following your reviews of hi end DACs and considering one for myself partly on your reviews. You add a server you aren't even familiar with just for the review? The server alone was mentioned by JA as having a major influence on the sound and that is just added as an unfamiliar component to you as well as the Bridge. I mean we aren't getting a review of the DAC when you say that it takes the 8K server and DCS bridge to get it sounding great. I am very interested in your opinion of how these upper tier DACs compare to each other but when one is being used with one server and another with another I don't see how any kind of objective comparison can be made. Am I missing something here? I mean no disrespect this DAC/server/renderer/ROON/network stuff has definitely got me intrigued. I just don't understand why you would jump from ROON NuC to the Aurdender doesn't that alone play a major role in the sound?

Thanks, Ryan

Jason Victor Serinus's picture

I sense no disrespect, and appreciate your comment. Your concerns are valid. Please note:

1. The dCS Network Bridge has been reviewed by me—the review is online—and is an integral part of my system. There is nothing unfamiliar about it at all to me. It is the most transparent way to get files into a DAC that I have available to me. It's a really wonderful product that I use virtually every time I'm in my music room (which is almost every day).

2. Roon has been my preferred music software for quite some time. I've compared its sound on computer to Audirvana Plus - I've given up on Amarra because its search engine stinks - and Roon trumps it easily. You can't play back files through Roon and feed them directly into the EMM Labs DV2 without using a computer, music server, or bridge such as the dCS Network Bridge.

3. I have reviewed the Aurender A10, and was thus familiar with Aurender's software and basic sound. The N10 has better sound than my MacBook Pro. It has been reviewed by us. It made absolute sense to use it. In fact, I have written a follow-up about the N10's sound.

4. I experimented with three different ways of getting files into the DV2: computer, music server, and bridge. Trust me (or don't trust me), there was a lot of listening involved in this review. I ended up with pages and pages of notes, and hauled lots of equipment back and forth. Contemplating the cable changes alone might have been enough to send someone into a tailspin.

I spent a huge amount of time with the DV2, which is now with me on loan to use in future reviews because I think it sounds so good. Because my best source for feeding it files is the Network Bridge, which doesn't have a USB out, I limit my files to 192 PCM or DSD64, and forego playing MQA. I can listen to / review the rest using the Rossini v2.0 DAC. (My follow-up that discusses Rossini v2.0 will appear in a later issue of the magazine.)

If you ever get to the PNW, search me out. I've had members of the Pacific Northwest Audio Society here on multiple occasions, and have now become audiophile buddies with several PNWAS members. It's a good thing.

Thanks, Ryan.

andy yuen's picture


Many thanks for responding. Can you request for an EMM transport or an EMM streamer to test the EMM DAC? Using another manufacturer network streamer to test the DV2 may not be the best option because of synergy issue, etc.

Jason Victor Serinus's picture

Thanks for reading my reviews and valuing my opinion.

All my reviews are assigned by Stereophile's new editor, Jim Austin. He can certainly contemplate a review of EMM's CD/SACD transport, which I'd be happy to undertake.

EMM Labs does not have a streamer available, as far as I can tell.

Yes, each company uses a different protocol. CH Precision and dCS (amongst others) use or are contemplating using dual AES for hi-rez files, and base their products on that protocol; others use USB. There are also two different ethernet protocols. Each protocol has its advantages. Some products are Roon Endpoints, while others are not. Some products decode MQA, but not all. And some companies use proprietary links which only work within their family of products. Life is not simple.

We all do what we can do, in the best way we know how. Well, at least some of us. I certainly strive to.

andy yuen's picture


I think you have not heard the full capabilities of the DAC section of the DV2 until you used the optilink connected to an EMM transport (playing well recorded SACDs) and a EMM Pre Reference preamp or any world class preamp!