The Entry Level #40 Page 2

While the AudioQuest DragonFly has a black zinc-alloy case with subtly rounded edges and a smooth, almost rubbery texture, the Audioengine D3 has a case of injection-molded aluminum with very clean, flat edges. Small as both models are, the AQ easily feels more solid, more pleasant to touch. By contrast, the Audioengine feels somewhat tinny and hollow. The AQ is almost impervious to fingerprints; the Audioengine wears them with apparent pride. Through regular listening sessions, I noticed that the D3's case was consistently warm to the touch—never alarmingly so, but warm enough to warrant mention.

Unlike the DragonFly, which offers no clear or easy way beneath its surface, the D3's case is held together with four tiny hex screws. If you open that case (I did not), you should find an Asahi Kasei Microdevices AK4396 D/A converter (perhaps most famously employed by the late, lamented Logitech Transporter network music player), a Texas Instruments LM49726 op-amp chip, and a Texas Instruments TAS1020B USB controller chip operated in asynchronous mode. The D3, which is capable of handling bit depths of up to 24 and sampling rates of up to 96kHz, upsamples incoming data, has a specified maximum output voltage of 2V RMS, and is claimed by Audioengine to successfully drive headphones with impedances as low as 10 ohms. As with the DragonFly, the D3's volume is controlled by the host computer's operating system. When I plugged the D3 into one of my laptop's USB ports, the computer took almost a full minute to recognize and identify it as "USB Composite Device Audioengine D3."

Audioengine states that the D3 requires 40 to 50 hours of break-in before it reaches peak performance. Right out of the box, it delivered a big, bold, pleasantly forward sound, with remarkably brilliant high frequencies and fast transient attacks. Early in my listening I thought the D3 was, in fact, too sharp on top, its images too explicitly drawn, its definition of leading edges too intense. It's important to note that I drew these conclusions while listening through the Skullcandy Aviator over-the-ear headphones, which themselves tend to emphasize the highs. Just as any speaker must be considered within the context of its partnering amplification, any headphone amp must be considered within the context of the headphones it's driving. Nevertheless, the Audioengine's brightness mellowed out over time, though not at the expense of that intoxicating speed and vibrant high-frequency color. And while it never exhibited the DragonFly's exceptional rhythmic control or matched the AQ's tight-fisted grip on musical notes, the D3 threw a large soundstage with excellent image focus and, most impressively, had a way with snare drums and cymbals that was magnificent to behold.

Despite the hundreds of times I've enjoyed the breakneck, nearly reckless performance of "Epistrophy (Incomplete)" from At Carnegie Hall, by the Thelonious Monk Quartet with John Coltrane (908kbps ALAC rip from CD), I've never been so tantalized by the drum and cymbal work as when I played it through the Audioengine D3. I was so taken by the speed, clarity, delicacy, and grace of each masterful stroke against that brilliant ride cymbal that, after a few consecutive listens, I had to pause to research the drummer, of whom I'd previously known nothing. Now, thanks in large part to the Audioengine D3, I've totally fallen in love with the fiery, high-stepping swing of Rossiere "Shadow" Wilson (1919–1959).

Speaking of informed appreciation of something beautiful . . .

Cambridge Audio DacMagic XS USB DAC–headphone amplifier
A few years back, Cambridge Audio strengthened its position as a leader in the world of digital-to-analog converters with its successful DacMagic, a relatively small (8.6" H by 2" W by 7.6" D), affordable ($449), versatile model that could be placed vertically or horizontally, upsampled all incoming data to 24-bit/192kHz, and offered three digital inputs and three analog filters. Reviewing the original DacMagic in March 2009, Sam Tellig praised its well-defined bass, extended highs, natural midrange, and wide soundstage. "In a word," he wrote, "it sounded glorious."


Now, with the DacMagic XS ($199), Cambridge Audio has entered the fast-growing market of portable USB DAC–headphone amplifiers. While the AudioQuest DragonFly and Audioengine D3 have the approximate shape and size of a USB thumb drive, the DacMagic XS measures 2.1" L by 1.2" W by 0.4" D and more or less resembles a large domino—or, perhaps more precisely, a very small hi-fi component. Its attractive, black-finished case of brushed aluminum has cleanly beveled top and bottom edges and smoothly rounded sides.

Rather than plugging directly into a computer's USB port, the DacMagic XS connects to its host via a 6"-long micro-USB extension cable. "By plugging in via a cable," Cambridge Audio's website explains, "the DacMagic XS avoids adding a bulky and easily-bashed lump to the side of your laptop, unlike some inferior USB DACs." (Wow. Wonder who they're talking about?) Many users will appreciate this intelligent feature, but I found the extension cable somewhat inelegant; plugging the very small connector into the DacMagic XS's equally tiny USB jack was almost always awkward and never quite as simple as it should have been.

Overall, I guess I prefer adding a bulky, easily bashed lump to the side of my laptop. To be fair, AudioQuest currently provides the DragonTail ($16.95), an attractive and apparently well-designed extension cable; and, while Audioengine has considered providing an extension-cable option, the company says it has received no such demand from its customers.

At the end opposite its micro-USB jack, the DacMagic XS, like the AQ and Audioengine devices, has a ¼" jack intended to accept the lead from a set of headphones or to connect, via an aftermarket cable, to a component hi-fi system. (Have I mentioned that the USB DAC–headphone amplifier is an absolutely brilliant component category?)

The DacMagic XS uses a high-quality, asynchronous-mode ESS Sabre ES9023 DAC—the same chip found in Peachtree Audio's popular decco65 D/A integrated amplifier. Connected to a USB port in is default USB Class 1 mode, the DacMagic XS will handle 16- or 24-bit files with sampling rates up to 96kHz; switched into a USB 2.0 port, it handles sampling rates up to 192kHz. The claimed output voltage is 2V RMS, the minimum recommended headphone impedance 12 ohms. Finally, unlike the AQ DragonFly or Audioengine D3, both of which use the host computer's volume control, the DacMagic XS provides its own onboard, 53-step volume control via two round pushbuttons labeled "+" and "–." After only a few seconds, my laptop recognized the DacMagic XS as "USB Composite Device CA DacMagicXS 1.0" and automatically set the operating-system volume control to maximum, thus allowing me to use the DacMagic XS's own volume control for fine-tuning.

The sound? In a word, glorious. Whereas the AudioQuest DragonFly traded some speed for a more relaxed, full-bodied sound, and the Audioengine D3 sacrificed some grip and control for transient clarity and overall presence, the Cambridge Audio DacMagic XS fell somewhere in between, offering greater top-end clarity and extension than the AQ, but with a richer, more relaxed overall sound than the Audioengine. In addition, the DacMagic XS was more revealing of recording artifacts and low-level detail than either the DragonFly or the D3—a characteristic whose advantages varied with the quality of the recording.

With "Nebula Ball Rests in a Fantasy Claw," from the Fucking Champs' thrilling but heavily compressed V (1070kbps ALAC rip from CD), the Cambridge was less forgiving than the Audioengine and far less forgiving than the AQ, shining light on the recording edits that screw up the rhythm near the song's midpoint. But in "Epistrophy (Incomplete)," from Monk and Coltrane's At Carnegie Hall, the Cambridge did a wonderful job of both preserving the recording's live, timeless quality and illuminating the single-note runs from bassist Ahmed Abdul-Malik that underlie and elevate Coltrane's soaring tenor sax.

I wish Robert McGinley Myers could have heard it. The effect was magic—and it was real.


dalethorn's picture

The ear/brain can hear very tiny differences, and even subliminal things are sensed. But at different times of day the ambient noise varies, AC line noise varies, the human body may have very different active energy levels - all adding up to very different perceptions.

Utopianemo's picture

Thanks, Stephen.  

Some very astute observations.  It kind of makes me a little happier with my entry-level Grados and my Hifiman Waterlines.  Maybe I don't need more expensive gear, just 3 or 4 decent-sounding headphones that I can rotate between for that new-sound feel. :)

Music_Guy's picture

I enjoyed the referenced articles/blogs (almost) as much as your piece here.  The concept that even though I may know better, things are actually enjoyed better with context and external reinforcment stimuli makes perfect sense to me.

My music sounds better when I sit down and savor it.  When I hear a piece of music that I haven't listened to in a long time, I hear things I did not hear (or don't remember having heard) before.  I absolutely enjoy my stuff more because I know how I searced for it and chose it and set it up than if I just came upon it complete.

As the technology improves, the differences between high-end DACs and entry-levels ones is shrinking. (IMO)  It takes careful comparison with short intervals between listens to detect these differences.  My less than golden ears find it nearly impossible to hear the differences when I hear gear in different places and at differnent times. But when I do crtically listen, I invariably hear differences and for the most part enjoy each new sound and usually think it sounds "better" especially when I am told that there is some improvement in there soemwhere.

You know what car runs better after I wash it.  I know it can't be...but it sure seems that way to me.

I am going to polish my tonearm, re-seat the cables, re-route the speaker wire and enjoy the improvements....

Tiger Al's picture

I'm gonna throw in a cd and enjoy the music.

Kettch's picture

Hi Stephen, I just read that you are leaving Stereophile. Is this true and will you be continuing to write through a personal blog or other platform? I always appreciate your insights from your column. 

John Atkinson's picture

Kettch wrote:
I just read that you are leaving Stereophile. Is this true and will you be continuing to write through a personal blog or other platform?

Sadly, it is true. Stephen left Stereophile at the end of March after more than 13 years with the magazine. He was made an offer to join cable manufacturer AudioQuest that he would have been crazy to turn down. His new position is VP Communications for that company.

His final "Entry Level" column appears in our June issue. He will no longer be writing about hardware for Stereophile or blogging for our website, but it is possible that he will still be contributing record reviews to the magazine.

We are not replacing "The Entry Level" as such but editor Michael Lavorgna will start contributing a column with the July issue.

John Atkinson

Editor, Stereophile

Utopianemo's picture

Mr. Atkinson,

I was immediately sorry to hear that Stephen has left the building.  When you say "The Entry Level" will not be replaced, do you mean that Michael Lavorgna will be continuing "The Entry Level"?  I sure hope so.  I am not familiar with Mr. Lavorgna's writing, but I know you understand how important this specific column is to your magazine(and to us).

Stephen, if you can hear me, good luck and thanks.  Em, good luck and thanks in any case.


Nathan Daniels

John Atkinson's picture

Utopianemo wrote:
I was immediately sorry to hear that Stephen has left the building.

Me too :-)

Utopianemo wrote:
When you say "The Entry Level" will not be replaced, do you mean that Michael Lavorgna will be continuing "The Entry Level"?  I sure hope so.

No, Michael's column will explore the kind of products he currently writes about for in greater depth. We will continue to cover affordable components in our reviews and regular columns, but "The Entry Level" was so strongly identified with Stephen Mejias that I thought it best to cease publishing the column as it currently stood. I hope to be able to launch a replacement column devoted to entry-level components when I have discovered a writer who offers the unique combination of experience, listening ability, a love of music, the ability to communicate what he and she thinks, and a love for uncovering affordable great-sounding gear.

John Atkinson

Editor, Stereophile

Kettch's picture

Hi John, thank you for the reply and I had hoped it was just a rumour but at the same time I am happy for Stephen. His column will be greatly missed.


Utopianemo's picture

I'm sorry, I just don't buy the wine thing.  I agree with Felix Salmon's assertion in the above referenced Reuters piece that knowing the price one paid for wine corrolates to one's enjoyment of said wine in part.  I know that effect is true for some people, but by no means all. My own particular bias is oriented around getting the best value; If I wasn't paying specific attention to the taste, I'd be the most impressed with the one that had the highest cost that I paid the least for.  That's why I get all excited when I hear the phrase "it sounds as good as speakers retailing for twice the price" in reviews.  That's the placebo that activates my pleasure center.

dalethorn's picture

It's even more likely to sound "As good as items at twice the price" if the user is willing to do a little tweaking, but it seems so many audiophiles today say "If it isn't perfect out of the box I'll just send it back".

tnargs's picture

Of course our receptivity changes from time to time, with the same equipment. I take that as a given.

And of course it means the Stereophile-style 'listening' 'test' is a joke. I take that as a given too.

And of course the placebo effect is real. I take the fact that audiophiles even want to discuss this point, with reference to our hobby, as a simple matter of re-education (away from everything audio journalists have been writing and towards simple realities).

But the Myers solution of selling good gear and going back to worse gear is not one I commend, except for him personally if he was truly unable to simply enjoy the music through high performance gear. But that is simply a *personal* psychological issue for him.

It is still okay to obtain high performance equipment and thoroughly enjoy our music through it, comfortable in the knowledge that there is less interference between us and the music that way.

luvmusic1945's picture

When i drank, i would listen to music and never thought the volume was high enough for me to hear everything; about 6 months after i quit, i put on a record and was amazed at the instruments i could now hear on same  record. I was now paying attention to the music. Going back to the wet years and the subject of price vs hearing, i'd like to add that i once went to a high gear store and heard a pair of speakers that sounded so real, that i went home and would not play my gear for a couple of weeks. I knew it would only make me feel sick, knowing, what i could be enjoying if i invested in better stuff. It's between the ears, some say, but as a V8's posess a distinctive sound, you know, when you hear one.


In my experience the biggest changes in audio response in transducers come from changes in humidity.