How can an Antipodes Music Server improve sound quality when all it is doing is sending ones and zeros to the DAC?
High-end digital audio is not just about the ones and zeros.
It is true that for many digital applications, it is just about the ones and zeros. The digital data can be sent in packets, in any sequence, and with an acceptable amount of delay. At the other end, the file can be re-assembled, and any packets missing or unreadable can be requested to be sent again. This can result in bit-perfect transmission of emails and files, because time is not tight in these examples. In this case it is indeed just about the ones and zeros.
In more time-sensitive digital applications like a voice call or streaming media, the packets are not sent chaotically, they are sent using some form of streaming protocol, and are sent over high-bandwidth backbones. If a packet is dropped or delayed unacceptably, then error correction is used. Telecommunications companies used to manage dedicated circuits for voice calls, getting efficiencies in the backbone by multiplexing. Now they use shared packet networks and solve the problems with sheer bandwidth and error correction, to achieve even greater efficiencies. In this case it is still about the ones and zeros, but it is also about the effectiveness of the transmission protocols to manage a time-sensitive service over a chaotic packet network. It also involves signal buffering and regeneration at the receiver end, as well as error correction. The customer experience is certainly not just about the ones and zeros in this case.
In high-end audio, the objective is loftier. We see the ideal as ensuring that the digital audio signal entering the DAC chip is a perfect square wave representing not just the ones and zeros, but also the clock data. The ones and zeros are only a concept and the transmission is actually an electrical or light signal that represents them, and it is subject to interference and bandwidth limitations in the same way as for an analogue signal. It is important to understand that any distortion of the square wave results in distortion of the clock data. If the DAC chip does not receive perfect clock data, then audio quality is impacted. To achieve the ideal of a perfect digital audio square wave at the DAC requires a perfectly clocked signal at the beginning of the final transmission, zero noise interference with the signal, and infinite bandwidth capability to completely square out the wave that is transmitted. Just as with any other part of high-end audio, this ideal, and therefore our desire for perfect sound quality, will never be fully met. High-end digital audio is absolutely not just about the ones and zeros.
This is why high-end music servers are designed to deliver as clean and precise a digital audio signal to the DAC as possible.
Can’t a poor digital audio signal be fixed in the DAC? If the DAC uses an asynchronous input like asynchronous USB or Ethernet, then isn’t the upstream timing irrelevant? The DAC re-generates the signal and uses its own clock to time the signal.
To be clear, the use of an asynchronous interface is to make it easier to buffer and re-clock a poor digital audio signal. This is because with an asynchronous stream the receiver can exert some control over the arrival of packets to avoid the buffer over-flowing or running empty. What the DAC chip needs is a clean synchronous signal.
Conceptually, a poor signal can be fixed by regeneration just at the point before the DAC chip. But in the real world, that is unrealistic. Instead, high fidelity is aided by handling the signal in stages. Asynchronous transmission occurs in the early stages when the data is in poor shape. Synchronous transmission is always used in the last stage to feed the signal to the DAC chip, and the method used is typically I2S.
In our experience, for high-end computer audio, you ideally need four stages – Server, Player, Regenerator and DAC. At each stage the signal is transitioning from one form to another, and its precision is being improved.
- Server. The Server stage receives internet streams and/or reads from storage and then sends improved streams to the Player using a streaming protocol over an Ethernet connection. To do this well requires a high-power computer (eg. Intel i7).
- Player. The Player receives the stream and sends a further improved digital audio signal to the Regenerator stage, typically, using Asynchronous USB. To do this well requires a medium-power computer (eg. Intel Celeron).
- Regenerator. The Regenerator stage sends a high-precision digital audio signal to the DAC using a synchronous method (I2S, AES3 or S/PDIF). To do this well typically requires a low-power processor (eg. FPGA).
- DAC. The DAC receives the digital audio signal and converts it to analogue audio.
Note that buffering and reclocking occurs at all four stages, but we have singled out the Regenerator stage that first turns the signal into a synchronous signal because of the significance of this step. Just because the signal is regenerated does absolutely not mean it is not affected by the quality of the incoming data stream.
In our view, all of the first three stages should be performed before and outside of the DAC. Having an Ethernet input, a Player computer and a USB input inside a DAC significantly increases noise interference levels in the DAC. Removing the first two stages is very important, but there is also significant benefit from removing the third stage too.
When so many audiophiles use a standard computer as their Server, or stream from a Server that is hundreds or even thousands of kilometres away over the internet, it is no surprise that DACs have had to evolve to try and do all or most of the four roles. But, in our view, this is simply unrealistic and, ultimately, it is bad architecture if your goal is high-end audio sound quality.
To be fair, it is expensive to do all of the four stages separately from each other. Our K50 completes stages 1, 2 and 3 in a single box, with each stage having its own fit-for-purpose circuitry, dedicated power supplies, galvanic isolation between all of the stages, etc. And performing these stages outside the DAC has big advantages for sound quality. We have converted many audiophiles from using the Ethernet and USB inputs on their DAC, to using I2S, AES3 or S/PDIF from the K50.
It is true that there have been plenty of marketing claims of DACs being able to completely eliminate jitter (timing error), but in our experience these DACs always sound better when supplied a better signal. Interestingly, we are seeing more DAC manufacturers acknowledge this and separating out their Player and Regenerator stages into a separate box from their DAC box. I suppose you could say that the Server stage is our preserve, and the DAC stage is their preserve, but that we are now competing with each other on the Player and the Regenerator stages. Where there seems to be growing agreement, at least at the high-end, is that the first three stages should occur outside the DAC.
In our S30, S40 and K30 we only perform the first two stages, because this is a sensible compromise at this price-point. To get the cost of a digital audio front end down even further, then the argument to just use a good streaming DAC gets more realistic. But in our experience, using a S30 and S60 with an inexpensive USB DAC will make better music than spending the same total amount of money on just a better DAC.
As always, listen with an open-mind and decide this for yourself.
So what makes an Antipodes Music Server any better than its competitors?
We cannot speak for every competitor, but what seems to set us apart from the bulk of the competition is that we design our music servers from the ground up to minimize noise interference with the digital audio signal, and that this enables us to do a better job of maximising bandwidth.
We see a lot of the competition using basic computer parts to do the heavy work, and then using power supply designs and noise filtering to reduce the noise. Many use add-on boards to re-do a job that was done poorly on their motherboards. The trouble with this approach is that over-use of slow linear power supply designs and noise filters constrains bandwidth, and a digital audio signal needs a lot of bandwidth to square out the wave and thereby define the timing data with precision.
One way of putting it is that Antipodes uses fewer parts, but higher quality parts. We also avoid taking quality short-cuts, with the parts we use, when trying to hit a price point. For example, our least expensive music server is the Antipodes S30, but if you feed it with an Antipodes S60 power supply (an upgrade option for the S30), you get the same hardware and parts quality that we use for the Player stage in the K50.
But it goes further than just higher parts quality. One of the things we believe we do uniquely is to tune residual noise that cannot be totally eliminated. Imagine that three different chipsets generate noise that each peak at a frequency that reflects their clock speed. If the peaks coincide then nodes are created and the noise floor goes up dramatically at that frequency. Shifting the clock speeds can avoid noise nodes and reduce the combined noise level, as illustrated in the image below. Managing this for an entire music server is very complex, but yields significant benefits for sound quality.
If you compare an Antipodes music server with many of those from our competitors, you will hear a clear difference. If you only listen for tonal qualities, you may not hear a fundamental difference. But if you let yourself engage emotionally with the music you will hear more life, urgency and drama with an Antipodes music server. Listening to music is about being moved emotionally – to make you want to smile, cry, dance … Then you will understand why we take a different approach.
Can you explain a bit more about the impacts of noise and bandwidth?
We mentioned earlier, that in order to get a perfect square wave digital audio signal, you need a perfect clock, zero noise interference on the signal, and infinite bandwidth. With these three factors you can precisely define the transition point in time between one bit and the next.
Here is a graphical representation of a perfect digital audio signal.
This signal represents the data ‘01010’. The ones are represented by a 1v level, the zeros are represented by the 0v level, and the clock data is defined by the vertical lines that define the transitions between them. For argument’s sake, let’s say that the digital receiver recognises a transition from a zero to a one when the signal rises through the 0.5v level, and recognises a transition from a one to a zero when the signal falls through the 0.5v level.
Here is what happens when we add noise at a frequency that is below the bitrate.
Here is what happens when we add noise that is above the bitrate.
Here is what happens when the transmission bandwidth only just matches the bitrate.
Here is what happens when the bandwidth is one harmonic above the bitrate.
Here is what happens when we combine the impacts of some noise and some bandwidth limitation.
These examples over-simplify and exaggerate the point, but they do serve to illustrate how a combination of noise and bandwidth limitation obscures the clock data. But the point cannot be trivialised either as the perfect digital audio square wave is always going to be impossible to achieve. For the last graph above, it is obvious that the digital receiver is not going to be able to discern the transitions between the bits with perfect precision.
This is what a high-end music server is all about, delivering the digital audio square wave with as much precision as possible, and no music server will ever achieve perfection given the real-world issues involved.
Can you explain how your technology is applied in your model line-up?
The key building blocks of the current models are the following hardware components:
- V5.6H Main Board. The V5.6H is used in the K50 and K40, and is a fundamental reason why the K50 and K40 are our best sounding models. In terms of power, this board can upsample and transcode a CD resolution music file to DSD512 using Roon DSP, using a single cpu core. It has another 5 real cores to handle other functions at the same time as playback, making it very versatile, and giving it the headroom to run the Server Apps outstandingly well.
- V5.2H Main Board. This board has similar single core performance as the V5.6H board, but has fewer cores. This board is a ‘first’ – the first Antipodes board that is very good for both Server apps and Player apps. The V5.2H board has two real cores and two virtual cores. If the playback solution is kept relatively simple then this board gives you 75% of the audio performance of the V5.6H, and can be run hard without any over-heating concerns, as long as it is well ventilated. The V5.2H board is used in the K30 and the S40.
- V5X Main Board. This is our premier player board, with very high levels of resolution and refinement. This board is also powerful enough to run Server apps and Player apps at the same time. But note that we do not recommend you use this as just a Server device, as its voicing is not right for this use case. Its voicing works very well if it is used as a Player only or as a Server and Player. In terms of computational power, it can upsample and transcode CD music files to DSD256 using Roon DSP, but requires Roon Parallelisation to be turned on to use all of its cores to achieve that.
- R1i Reclocker Board. This reclocker board isolates the input electronics from the reclocking section and output electronics, and uses femto second clocks renowned for their high-end audio performance to provide excellent synchronous output performance, making your DAC’s job a lot easier than if you fed it with USB or Ethernet. The R1i board is used in the S20.
- R1x Reclocker Board. This is an exceptional reclocker board, using the best available components throughout, including the world’s best sounding oven-controlled clocks in an ultra-high-end circuit. The R1x is used in the K50.
- HSL80 Power Supply. The HSL power supply scheme used in the HSL80 and the HSL50 are responsible for a large part of the the sound quality improvements over our previous generation of music servers. The HSL80 is highly over-specified and is used in the K50, which uses three HSL80, and in the K40 and K30 which each use a single HSL80.
- HSL50 Power Supply. This power supply is the same design as the HSL80, but has a lower current capability. The HSL50 is used in the S60, which has plenty of headroom for use in the S Series. It is important to note that the S20, S30 and S40 have been tuned for optimum performance with a very fast power supply. For low cost, an smps power supply is suitable. For much higher performance, an S60 is ideal. While you can power an S20, S30 or S40 with a third-party linear power supply, you will lose much of the speed and excitement compared to using the S60. The S60 is highly recommended as an upgrade for any Antipodes music server from previous generations that used external power supplies. We know you will be stunned at the huge lift in performance available by using a power supply that has been designed specifically for high performance music servers.
Which models use which building blocks?
All Antipodes music servers, except the K40, are capable of running both the Server App and the Player App, and capable of being used as just the Server App or just the Player App. The K40 is designed to only run the Server App.
- S30. The S30 employs the V5X board, offers USB and Analog outputs, and is powered by a standard smps power supply brick. The existence of the analog outputs does not mean that an audiophile DAC is included. The Analog outputs are provided as a convenience, enabling a low entry point for a user that does not have an audiophile DAC. The sound quality increases significantly when a quality USB DAC is added. The S30 can be used to run Server Apps and/or Player Apps, but its hardware is optimised for the Player App function.
- S40. The S40 employs the V5.2H board, offering Direct Streaming over Ethernet and USB Outputs, and is powered by a standard smps power supply brick. The S40 can be used to run Server Apps and/or Player Apps, but its hardware is optimised for the Server App function.
- S20. The S20 employs the R1i Reclocker Board, offering S/PDIF, AES3 and I2S outputs, and is powered by a standard smps power supply brick. The S20 gets its input over USB. The S20 can be used to upgrade the sound quality of a S30 or S40.
- S60. The S60 employs the HSL50 Power Supply, which can be used to dramatically lift the sound quality performance of the S20, S30 and S40. A single S60 can power any two of the S20, S30 and S40.
- K30 The K30 combines the V5.2H board running the Server Apps with the V5X board running the Player Apps, with both powered by a single HSL80 power supply, and offering USB Audio Output. It is similar to combining the S40, S30 and S60 in a single case but with superior interconnection, and powered by the larger HSL80 power supply.
- K40. The K40 employs the V5.6H board, powered by a single HSL80 power supply, to provide the ultimate device for running Server Apps. The only output is Direct Stream Ethernet, which can feed an incredibly clean signal direct to the Ethernet input on any Player or DAC.
- K50. The K50 employs the V5.6H board to run Server Apps, the V5X board to run Player Apps and the R1x Reclocker board. Each of the three boards is powered by a dedicated HSL80 power supply. Outputs are Direct Stream Ethernet, USB, S/PDIF, AES3 and I2S.
- K10. The K10 is a high precision ripper in a case carved from just two pieces of solid alloy, that can be attached to any Antipodes music server, and then used to auto-rip your CDs. It connects to your Antipodes music server by a double USB cable.