Esoteric N-05 Network Audio Player


Equipment report
Music servers and computer audio
Esoteric N-05
Esoteric N-05 Network Audio Player

I started using no filter, then tried a couple of the available filters. I also tried an oversampling option. For most of the review I used the internal renderer, but to test the external inputs, I used an external server connected via USB cable.

In heavy rotation recently at Casa Forrester: Mari Kodama’s recording of all the Beethoven Piano Sonatas, a DSD64 Pentatone recording purchased from I’ve enjoyed Kodama’s performances on CD and SACD (ripped to my hard drive, of course), but welcomed the opportunity to acquire all 32 sonatas in the superior-sounding DSD format. On Sonata No. 32 the piano sound was very powerful, yet detailed. I have always admired the way DSD piano recordings depicted hammer action on piano strings, and this recording is a good example. The N-05 projected the lower registers of Kodama’s Steinway D with considerable power, and the well-defined microdynamics revealed Kodama’s sensitive phrasing. Through the N-05, I could hear complete, accurate note production, beginning with the initial transient that occurred when a key was pressed and a hammer struck a string, followed by the sustain part of a note, where complete harmonics were portrayed, to the decay part, where the note dwindled off into silence. An amazing recording, reproduced in realistic detail.

Next up was old favorite “Miserere” from The Tallis Scholars’ album Allegri’s Miserere & Palestrina’s Missa Papae Marcelli, a 96/24 FLAC recording downloaded from This a cappella choral work, a setting of Psalm 51, was originally reserved for performance in the Sistine Chapel, and is performed here by a small choral group recorded in a church. The main group is located at the front of the soundstage, while a small solo group is located some distance behind the main group. The N-05 laid out the soundstage before me with considerable realism: The solo tenor appeared centered at the front of the soundstage, the rest of the main group was spread out between the speakers at the front of the soundstage. The solo group sounded appropriately distant, yet I was able to hear them in considerable detail as the soprano soared to a high “C.” Some digital components impose a smear of distortion when the distant solo group enters, but not the N-05. Also, the solo tenor’s voice was free from any edginess or distortion, which can be a problem with some components. Although he’s located front and center, there was a feeling of spaciousness around the tenor, as though he was singing in a large space, which, in fact, he was.

Another old fave, “Folia Rodrigo Martinez,” was ripped as an AIFF file from the CD La Folia 1490-1701 (Alia Vox AFA 9805). On it, Jordi Savall and his band re-create a historically informed performance of a musical work dating from the year 1490. If that sounds like stodgy, boring music, it’s actually one of the more rollicking fun pieces of any vintage I’ve heard. Through the N-05, the piece sounded quite detailed, beginning with the three opening whacks on the cascabels (sleigh bells), each of which sound slightly different, as they do on better components. The baroque guitar plays a tune that is echoed by a harp, and although they’re playing the same tune, the two instruments sound somewhat different. I’ve heard the difference sound more pronounced with some other, more expensive gear, however. The main tune is played by leader Savall on his viola da gamba, a lively melody that constantly varies in loudness and speed. Percussion instruments consisting of castanets, a wood block, and a drum accompany the melody—the castanets sound somewhat distant, and have a slight tendency to get blurred into the background. The drum extends surprisingly deep—I think it goes into the mid-20Hz range—and provides a foundation for the piece. The N-05 reproduced the drum with power and impact, though perhaps a smidgen less of both than I’ve heard from a few other components. Still, it was a good workout for the subwoofer. The wood block was audible throughout the piece, and the N-05 accurately portrayed the transients of strikes on the block. The viola da gamba’s harmonics sounded accurate, although I noted a slight brightness.

So I’ve used three different musical samples: a solo piano recorded in DSD format, a choral group recorded in high-resolution PCM format, and a small instrumental band, recorded from a CD: a variety of recordings. Now let’s see how those selections sound when we switch in a couple of the N-05’s filters. (The filters can be set through the use of the Menu button on the front panel.)

First, the FIR1 filter, described in the manual as “a steep roll-off…used to sharply cut signals outside the audio band.” This type of filter is sometimes known as a “brickwall” filter because its action outside the audio band is very drastic. It only works on PCM recordings, so I didn’t use the DSD recording here. On the “Miserere” track, I thought I heard two effects, both very subtle and hard to detect. First, I thought I detected a very slight hardness with the FIR1 filter switched in. Second, I thought the sound of distant solo group behind the main choral group was very slightly more diffuse, less defined. With the “Folia Rodrigo Martinez,” the effect was also slight; I thought the slight brightness or edginess increased a small bit, and I thought the castanets sounded a little smeared. (In case you’re wondering, yes, I did fully break in the FIR1 filter.)

OK, now let’s switch to the SDLY2 filter, which the manual describes as a “short delay filter with a slow roll-off…used to gently cut signals outside the audio band.” I picked this filter because I thought it would be the most different from the FIR1 filter. And it did sound different. On “Miserere,” the slight hardness seemed gone, while the sound of the solo group sounded “groupy-er,” or better defined. On “Folia Rodrigo Martinez,” the edginess was reduced, though was not totally gone, and the castanets seemed better defined. Of all the filter settings, I preferred the SDLY2 filter. There’s a DSD filter, too, which rolls off the response over 50kHz, to eliminate the extremely high frequency noise that DSD produces. I must confess I heard absolutely no change when I switched in the DSD filter; maybe those with more extended hearing and speakers with a super-tweeter could hear a difference.

The N-05 gives you the capability to upconvert PCM signals to higher sampling rates. One such option converts them to DSD256 signals, which I decided to try. I figured that if any upconversion were obvious, it would be the DSD upconversion, and I was right, it was obvious. On “Miserere,” I wasn’t smitten by the effect; the sound seemed somewhat homogenized, a bit rounded off. The sense of depth was good, but overall, the sound seemed a bit congested. If the N-05 were mine, I’d leave the upconversion turned off—a personal choice.

So there you have it. I didn’t want to spend a lot of time with filters and especially not with upconversion, since the effects are subtle and subject to personal preference. But since the options are there, I felt I should sample some of them and tell you what I found. Your reaction to them might be different than mine.

To evaluate the N-05’s external digital inputs, I suppose most people would plug in a computer, and that’s what I did; however, rather than use a PC, I used an SOtM sMS-1000SQ network music player that performs most of the functions the N-05 does. It plays DSD and PCM music files and is controlled by an app running on an iPad. It also streams Tidal, although not as straightforwardly as ESS. It’s even housed in an attractive chassis that looks nothing like a computer. It differs from the N-05 in that it doesn’t have an internal DAC, so it’s just what I needed to plug into an external digital input on the N-05 and use its internal DAC. I connected the SOtM network music player to the USB input on the back of the N-05, moved the network cable from the N-05 to the SOtM, turned it on, turned on the iPENG 9 app that controlled its functions, selected the USB input on the N-05’s front panel, and was ready to play music. I confess my expectations were that the SOtM playing into the N-05 DAC would sound indistinguishable from the N-05 player/DAC combination. Ha!

Beethoven’s Piano Sonata No. 32 had slightly weightier sound, with stronger upper bass. I heard a tad more detail and better definition of the overall formation of notes, with especially well-defined leading-edge transients. Great microdynamics made Kodama’s interpretation more exciting. “Folia Rodrigo Martinez” sounded rather different—most noticeable was the more extended and powerful bass drum, which had considerably more extension and impact. The slight brightness I had heard from the N-05 was no longer present. Castanets were noticeably better defined and didn’t merge into the background noise. Harp and guitar were more distinct and sounded clearly different—and more like a harp and guitar. “Miserere” also sounded different, though perhaps not as much. The main choral group was portrayed with more detail, so the individual voices were more distinctive. The solo tenor’s voice sounded more expressive; I could better hear how he phrased the words. The distant solo group sounded further behind the main group, yet I heard more detail in their individual parts. No hint of brightness was evident, although the high frequencies were quite well defined.

“How can that be?” the bits-is-bits crowd will ask. “All the player does is produce a bitstream that is sent to the DAC.” Doggone if I know why—all I do is report what I hear.