ong time ago when I tested the DSS 30 back in 2008, a product of the Italian company Bluenote (now Gold Note), I did not know how to name it. Company's material referred to the Digital Static Source (DSS), a solid-state memory player. Today I would say that it was an audio files player (playing music from thumbdrives and HDDs connected to USB inputs). In spite of the nine years that have passed since then, this type of product still has no one common name. Manufacturers are usually disoriented by it, but from time to time someone is making a use of it and calling their product in a really interesting way. One of the latest examples is the NETWORK BRIDGE, a product of the British company Data Conversion Systems, known as dCS.

The most commonly used names for such device today are: "streamer", "renderer", "server", etc. These names have been transferred from the computer industry since music file players are in fact specialized computers, and computer users are the biggest proponents of this kind of music source. In the audio industry we use other names, but - I think - there is nothing to argue about, ultimately it is people who decide about such things, simply by using one or another name. But one should know what these names mean.

For the benefit of this test, following "High Fidelity's" practice, I suggest that we agree that the Network Bridge is an audio files transport. It performs an analog function to the optical discs (CD, DVD, Blu-ray) transports, ie it reads the signal from the storage medium and decodes it into a form understood by the digital-to-analog converter. On one end music files go in, on the other PCM or DSD audio goes out.

A complete source of digital signal based on music files consists of:

• An external storage – a thumb drive, USB disk or NAS or / and a network connection that allows streaming signal in real-time,
• Audio files transport,
• Digital-to-analog converter.

Such system might use separate devices or can be integrated into a single chassis. The most common are audio files players, that is, transport and DAC in a single enclosure to which we attach external storage media or stream files over network (see Lumin). This system can be further integrated by including memory storage device (hard drives, see Aurender A10, in Polish) or even add an amplifier, thus offering a complete audio source (See Lumin M1).

NETWORK BRIDGE

Since the DSSS 30 Tube, that I mentioned at the beginning, a lot has changed, because now - for example – an Ethernet port is an obligatory feature of any player, as it allows user to stream files from NAS (via router) plus allows him to use a control app installed on smartphone or tablet. The most important change, however, is how people listen to music from files. For a long time, it seemed that music was going to be downloaded on a hard drive, either after buying it in online stores or ripping it from CDs. Last year, however, proved that constantly improving download speed (bandwidth of network connections) and the emergence of several important streaming services led to the situation where only hard-core high-enders keep music locally, ie music on their disks, while most listeners move completely into the virtual world of streamed music.

The Network Bridge is supposed to be a "bridge" that connects both streaming and local data carriers to our digital-to-analog converter, and thus the audio system. Although I use the name "transport", there is no “mechanism” on board and yet, it is "transporting" WAV, FLAC, AIFF, DFF and DSF files as well as those using lossy codecs and decoding them to PCM or DSD respectively. The bridge is UPnP compliant (the company calls it "UPnP renderer") and it works in asynchronous mode. It comes with Wi-Fi card allowing user to connect his iPod, iPhone or iPad via Apple AirPlay. But it is also compatible with ROON, which opens up new possibilities for users.

The Network Bridge is a small silver or black device with a micro LED on the front panel. One controls it using an appropriate mobile app. From the signal source side, we connect it using a RJ-45 (Registered Jack-type 45) cable to the router. We can also connect a USB drive to it, but in the future this port will be used as an audio output. dCS supports files with music signals up to 24 bits and 384 kHz, and DSD64 and DSD128 in DoP format. If you use a DAC that does not support high sampling rates, dCS converts PCM and DSD signals down to 176.4 / 192 kHz or 88.2 / 96 kHz PCM.

There are three output options. The most recommended is a link based on two AES / EBU (Audio Engineering Society / European Broadcasting Union) sockets, a solution also used by Chord. Dual AES allows transmission of PCM signals up to 24 bits at 384 kHz plus DSD64 & DSD128 in DoP format. However, there are only few digital-to-analog converters on the market able to receive signal in this way. The second option is to use the SDIF-2 interface using two BNC sockets, for left and right channels, and a separate one for the clock, but - see above. So I think that until the USB output is implemented, most user will use the RCA output or a single AES / EBU. In both cases, it will allow the device to output PCM signal up to 24 bits / 192 kHz and DSD64 in DoP.

In the near future, a USB port that we can use to connect an external USB drive will work as a USB output with the same parameters as dual AES, ie PCM up to 24/384 and DSD64&128.

The audio files transport's (renderer, streamer, bridge etc.) only job is to download signal from an external storage (NAS, USB, Internet), convert WAV, ALAC, FLAC, dsf and dff files to PCM or DSD respectively and send it to an external DAC. That is why I use the term "transport" - it is analogous to CD transport, it is an intermediary between the signal carrier and an external D / A converter.

So it is important both how the PCM and DSD signal is decoded, and what source delivers that signal. In the latter case, it would be analogous to the way of preparing different types of CDs - XRCD, UHQCD, Platinum SHM-CD, Blu-spec CD2, all of which are equivalent to different types of external drives. During the test I used both the Synology DS410j / 8 TB NAS disk and the Liksys WAG320N router as well as the Fidata HFAS1-S10U server / file player. I separate part of the test was performed using files stored on the USB drive connected directly to the dCS.

The hierarchy was clear: the best sound I achieved with Fidata, then Synology, and finally the USB drive. It is therefore worth devoting more than usual time to the proper configuration of the NAS. The dCS distributor, Audiofast, offers such configuration by equipping us with the ROON software.

The dCS transport was connected to the Chord DAVE digital-to-analog converter. Since the Bridge did not output the signal via the USB at the time of this test, I connected it with a DAC using a BNC digital cable with BNC-RCA adapter - dCS features RCA output and DAVE has BNC inputs only. This output allows transmission of PCM signals up to 24 bits and 192 kHz and DSD64.

I used Acoustic Revive LAN-1.0 PA cables with the RLI-1 filters on both ends connect dCS with the Fidata server and router. The Bridge was powered with the Acoustic Revive Power Reference Triple-C with the RSA-14 Triple-C filter and was placed on the top shelf of the Finite Elemente Pagode Edition rack. On top of it I put the Verictum X Block passive filter.

It would seem that - I will repeat what I already mentioned - the players (transports) of audio files should sound the same. After all, their task is to retrieve the signal from an external storage medium, convert the WAV, ALAC, FLAC, dsf and dff files to PCM and DSD respectively and send them to an external DAC. Theoretically, this kind of work should be done perfectly by any computer. In fact, however, the differences between these types of transports are greater than between Compact Disc transports - and these can be awful in some cases and ecstatic in others.

Network Bridge is no exception, it clearly has its own sonic character. And to no surprise, it largely resembles what I know from the Rossini and Vivaldi players, keeping proper proportions in mind, of course. Even using a completely different D / A converter, by simply comparing it to a CD transport and other file players, one could quickly and reliably determine the characteristics of the Bridge's performance.

The main feature is the intensity of the sound. This is something that, along with the development of this market segment and this type of product is getting better, but in most cases it is still the Achilles heel of file players, not excluding the module that is part of the Rossini. The Network Bridge outperforms Rossini and by a large margin. This is why it surprisingly well presented vocals, both from CD rips and from hi-res files. If the vocal is recorded with a near microphone set, if it isn't later altered during mastering (or remastering), the dCS presents it it in a large, intense, tangible way. It never “shrinks” it, which is often done by other audio file players.

DSD files with Frank Sinatra and Billie Holiday, but also classic PCM 16 / 44.1, like David Sylvian from Sleepwalkers and Clannad from their last album Vellum sounded very convincing. Even Leonard Cohen from his last but one album Popular Problems sounded in a strong, good way. It was still a poor recording in terms of voice quality - I do not quite understand why some people are so thrilled about it - even though I used the hi-res 24/96 file, but all things considered, it sounded surprisingly good.

All the more so, that with the energy and volume (general, subjective magnitude of the phantom images) there was also a quite low set tonal balance. No emphasis in treble not upper midrange, no “audiophile” tricks. The sound was “meaty”, there was commitment, power. To be clear: it's still not the same quality level we are dealing with in recording or remastering studios. Also, in my opinion, it's still not the same level that physical media have to offer. This stage is still ahead of us. But first of all, I do not know any audio file player that would do that, and secondly the Network Bridge performs well enough for me to completely accept and understand those who for some reasons put everything on one card and move to the dark side of power, ie to "network."

The feature of this device that distinguishes it among all others is the distinct differentiation of the tone and position of the instruments. The soundstage along with imaging of the instruments – that's what I mean by that - is fantastic, better than delivered by twice as expensive Fidata and incomparably better than when I played the same tracks from my laptop (with J-Play software player). I would say that in the so-called "real" life everything is more connected, even "blended", but ultimately we are talking about a reproduction, a recording, and it is a completely different kind of art than live music.

This might be something that will surprise most people at the beginning. Volume, intensity, richness - everything that I mentioned before - will come later. First, we will be struck by the scale of a soundstage and distinct separation of the individual components. This is part of the hi-res files heritage, primarily the PCM ones, something that they bring to audio. CD to some extent flattens this element, at least compared to vinyl. Although transitions between sound sources are clearer with the digital disc, and therefore the differentiation is also better. But Vinyl can make it all more cohesive thus combining all elements in a more natural way into a coherent story.

PCM files, but played at this level we talk about in the context of the Bridge, go even farther than a CD, but often too far away, ie the instruments lose connection between each other. The dCS shows them very clearly, but also points to the different tone of the individual elements, which leaves listener dealing not with color stains, but with impressionistic painting viewed from the distance, from which it translates into something more than single sounds, in a coherent message. Which is even more addictive when we play music from DSD files.

A slogan displayed on one of the windows at this year's High End show in Munich, in Lindemann's room: "DSD is the new analog" is quite reasonable. dCS clearly showed the differences between these two coding methods, i.e. PCM and DSD. It did not try prove one of them is much worse, did not emphasize their downsides, but also did not blur the differences. And yes: the PCM files sounded in a very precise, clear way with breathtaking panorama, while the DSD sounded smoothly, darker, with layers of the stage combined in a more cohesive way, with naturally presented 3D effect. So I agree that the high-end has finally got its digital "niche" - DSD files - in which it can cherish the "analog" ethos.

Finally, the feature of the Network Bridge, which one has to confront with one's expectations, because not always and not for everyone it will be something that one is looking for. The sound of this transport is characterized by far-reaching differentiation between instruments and plans, but within these plans resolution is no longer so excellent, it is simply good. This is something that no audio file player I know - aside from workstations in recording and mastering studios - can do as well as physical media, including CDs. Let's also add that dCS significantly smooths the edges, rounds the attack, giving the impression of "flow", but does not reflect all the nuances that are available with the LP, and even more so with an analogue mother-tape. Despite its high dynamics, the presentation as a whole is rather controlled than "live-like."

Summary

The dCS Network Bridge is a better audio file transport than the one implemented in the Rossini. This is one of the best transports I know, and only two and three times more expensive ones are able to better reproduce tone and are even more resolving. Everything the dCS does, it does well and some things even exceptionally well, so it can actually be a high-end audio source in a high quality audio system. I have to admit that even on me, a physical formats fanatic with particular appreciation the CD, the Network Bridge made quite an impression. A high quality NAS as a partner will allow you to achieve a great sound - let's not forget this element of the puzzle, because the Bridge clearly shows differences between different types of discs!

Although inconspicuous, the Network Bridge is very well built, and it hides a surprise inside. The chassis consists of a steel bottom, aluminum back plate and robust, reinforced with ribs side and upper walls. The latter is damped using a piece of bituminous mat. Such element damping vibration found in a product of people known for their "hard" engineering made me smile, because it turns out that the claim that has been discussed in the audio industry for a long time, that digital devices are extremely sensitive to micro-vibration, has finally been recognized by the mainstream engineering.

Only element on the front is a blue LED. If the device is connected to the router, it will be on permanently, when contact is lost – it flashes. On the back there are number of sockets: 2 x AES / EBU, RCA, 3 x SDIF-2, Wi-Fi antenna socket, two BNCs for external clock, RJ-45, USB and power inlet with mechanical switch.

The whole circuit is assembled on a single PCB. The player layout comes from Stream Unlimited. The Stream 820 board was chosen because it offers wireless Wi-Fi and cable network connections. This is a newer model than the one used for Rossini – the latter feature Stream 800. The dCS's created their own software for the Field Programmable Gate Array (FPGA) Spartan-6.

The Stream Unlimited module is plugged in the main PCB prepared by dCS. This arrangement provides several additional possibilities: the signal is additionally clocked, there is also a multi-stage linear power supply, isolating the digital section and the oscillators. This large PCB features the above mentioned FPGA, as well as two beautiful oscillators, mechanically compensated and temperature controlled (they have their own "heating" that stabilizes their temperature).

Specifications (according to manufacturer)

Dimensions (WxDxH): 360 x 245 x 67 mm
Weight: 4,6 kg

Digital inputs:
- RJ45 | FLAC, WAV & AIFF do 24 bit, 384 kS/s, DSD64 and DSD128 in DFF/DSF, other formats: WMA, ALAC, MP3, AAC & OGG
- USB 2.0 in asynchronous mode | PCM up to 24 bits, 384 kS/s, DSD64 and DSD128 in DoP
- USB type A in asynchronous mode | PCM up to 24 bits, 384 kS/s, DSD64 and DSD128 in DoP

Digital outputs:
- 2 x AES/EBU | PCM up to 24 bits, 192 kS/s, DSD64 in DoP. Used as a Dual AES pair, the interface outputs PCM at up to 384kS/s, DSD/64 & DSD/128 in DoP format.
- 2 x S/PDIF: 1x RCA, 1x BNC | PCM up to 24 bit, 192 kS/s, DSD64 in DoP
- TOSLink | PCM up to 24 bits, 96 kS/s.

Analogue outputs
Output signal: 2 V or 6V rms, selectable
Output impedance (XLR): 3 Ω, minimum loading – 600 Ω (recommended 10 kΩ – 100 kΩ)
Output impedance (RCA): 52 Ω, minimum loading – 600 Ω ( recommended 10 kΩ – 100 kΩ)

Word Clock I/O on 2 x BNC | 44,1, 48, 88,2, 96, 176,4 or 192 kHz

Noise:
- 16 bit > -96 dB0, 20 Hz-20 kHz unweighted
- 24 bit > -113 dB0, 20 Hz-20 kHz unweighted (6 V output).
Cross-talk: > -115 dB0, 20 Hz-20 kHz.
Power consumption: 26 W typical | 35 W max.