We are often asked about what the best connection is to use between a music server and a DAC – Ethernet, USB, S/PDIF, AES3 or I2S. But this is not a question about different connection types. It is about different solution architectures. This is because Ethernet, USB and the synchronous interfaces (S/PDIF, AES3 and I2S) occur at quite different stages of a computer audio solution, and so are not direct alternatives.
The image below illustrates how we use a multi-stage approach to delivering a computer audio solution. The purpose of the stages is to take a stream that has poor clock timing accuracy, and progressively improve the timing before the data enters the DAC stage, where the digital signal is converted into an analog signal. You can certainly make the music play with a simple one or two stage process, but to get a high-end audio result, more stages are necessary.
Remote Server – An internet streaming service like TIDAL or QOBUZ performs a Server role, but its delivery is compromised by the chaotic nature of internet delivery, and so sound quality is enhanced by re-serving the stream with a powerful local server. The remote server sends the music file to the local server using a streaming protocol, which means that the file packets are sent in the sequence that they are to be played, but they are not sent with the precise timing that a DAC chip requires.
Local Server – A Local Server organises your music sources, whether they be internet streams or locally stored files, and this role is performed using server apps like Roon Server, Squeeze Server, HQPlayer Server, Minimserver, PLEX, etc. The local server uses a streaming protocol to send files in the correct sequence to the player. A local server has a lot of functions to perform, and more power means it is more able to perform all of its functions, plus send the file packets to the player with some degree of timing accuracy.
Player – The Player runs Player apps that are compatible with your chosen Server app, such as Roon Player, Squeeze Player, MPD, HQPlayer NAA, etc. The Player takes the streamed file and turns it into a digital audio signal, using PCM packets or a DSD stream. The Player needs medium power in order to perform its functions and send the digital audio signal with good timing to the next stage. The better the quality of the stream from the server (in terms of timing and low noise), the easier it is for the player to do its job and output a quality signal. Typically the transport to the next stage is over asynchronous USB.
Re-Clocker – The Re-clocker provides the last step in the computer audio chain, converting the asynchronous signals in the previous steps to the synchronous signal that the DAC chip requires. The sound quality of what is output by this stage is very much affected by the quality of the incoming signal from the player. Isolation from noise generated by the previous stages is important, and this is more effective when the previous stages are designed to add very little noise to the system and to the signal. Noise can enter not only on the signal but also on earth paths and power supply paths, or from nearby components. At this stage the quality of the clock is crucial, because for the first time in the process, the output signal is synchronous.
As you go from left to right in the diagram, the timing gets better, and the transmission method goes from asynchronous (eg. using a streaming protocol over a packet-switched network, or block-mode transfer from a local disk, to RAM), then through a less asynchronous USB connection, and finally to an entirely synchronous feed to the DAC chip. Asynchronous transmission between the stages is necessary early on because with less than perfect timing accuracy the receiver needs some level of control of the arrival of data to avoid its memory buffer filling up or emptying out, to allow the buffer and re-clock stages to perform their roles without drop-outs or skips.
Therefore, on the left, the clock used is less important than on the right where it is crucial.
On the left the greater priority is powerful processors, but higher power comes with higher levels of electronic noise. As we move to the right, we need progressively less powerful processors. Each stage requires a different, but equally smart design: to provide the right level of uncluttered processor power to improve the signal timing; to reject as much of the noise on the incoming signal as possible; and to add as little of its own noise as possible, while maintaining very high bandwidth signal transmission.
It is the complete end to end process that matters and so you cannot skip, or short-change, any individual stage, without compromising the total result.
In the early years of computer audio, people used basic computers for the earlier stages. This placed a heavy burden on DACs to do the rest, so DACs sprouted Asynchronous Ethernet and USB inputs. In the absence of quality music servers, DACs had to include more of the computer functions in order to improve signal timing before the DAC chip.
Over the years audiophile music servers emerged, they got progressively better at the job, and audiophiles began to realise that using a standard computer did not get the job done nearly as well as using a good music server to feed the DAC, regardless of the quality of the DAC.
For even better sound, The Antipodes K50 has taken the additional step of performing the asynchronous-to-synchronous Re-clock stage before the DAC. In our view, when the Player and Re-clock stages are performed inside the DAC, it is placing powerful processors and their attendant noise too close to the DAC stage. This results in DAC design having to trade off processor capability to keep noise very low. By placing the Re-clock stage in the K50 we can give it all the power and parts quality required to do the job to the highest possible level of accuracy.
Some of the top DAC manufacturers clearly agree with us, producing multi-box solutions where the Re-clock stage and sometimes also the Player stage are in a separate case or cases from the DAC stage.
Of the synchronous connections, I2S is better than AES3, is better than S/PDIF. But the differences are not so large that S/PDIF, using a high-quality S/PDIF cable, cannot out-perform I2S using a basic cable. But I2S has the advantage of being able to handle much higher bit-rate transmission, and a clear channel for transmitting the clock data.
An ideal cost-no-object design would place each stage in its own separate case. When price is a consideration, this ideal can be traded-off by placing some stages together in the same case. Which ones you put together determines the type of connection you use between the music server and the DAC.
For example, with a K50, we recommend you use a synchronous connection rather than USB or Ethernet. But this is based on our contention that the Re-clock stage in the K50 is far better than the Re-clock stages in any of the DACs we are familiar with.
In the same way, a DAC manufacturer knowing that most of its customers are still using basic computers as their servers, will tell their customers that the DAC’s Ethernet connection is best.
Which one of us is right depends entirely on the music server and DAC that you use.
We hope this explains that your choice of connection between your music server and DAC is not so much about differences between the connection types, but a decision you should consider about the ideal composition and architecture of your computer audio solution. In other words, your choice of connection is largely determined by which stages you want done by your music server, and which stages you want done by your DAC.
- If you want the Server, Player and Re-clock stages to be done by your Music Server, thereby keeping computing tasks away from your DAC chip, then connect your music server to your DAC with a synchronous connection, S/PDIF, AES3 or I2S.
- If you want the Server and Player stages to be done by your Music Server, and have the DAC perform the Re-clock and DAC stages, then connect your music server to your DAC with a USB cable.
- If you want the Music Server to only do the Server stage, and leave the rest to your DAC, then connect your Music Server to your DAC with an Ethernet cable.