antipodesDX 600blk single

Antipodes DX Music Serve
 

Specifications

Fanless alloy case - Silver or Black
Auto CD Ripper to Uncompressed Flac
Paranoid-mode for bit-perfect rips
RAM-cached playback from internal & external sources, & streaming services
USB Audio 2.0 Output
Plus auxiliary S/PDIF & Analog Outputs
1TB, 2TB or 4TB SSD Internal Storage
Also available without internal storage
Internal Linear Regulated Power Supplies
110-120VAC or 220-240VAC
Power consumption 44VA; 15VA standby
85mm (h) x 430mm (w) x 273mm (d)
Shipping weight 7kgs
Warranty 2 Years With Registration

 

Description

The DX is our flagship music server, using purist design choices - all solid state storage, RAM-cached playback, and all parts are powered by state-of-the-art dedicated linear power supplies.

 Organic images, staggering detail retrieval and wide dynamic range, without any digital glare or grain, revealing the finest textural differences between voices and instruments, drawing you into the emotion in the music.

 The DX is now also available without internal storage for ripping to NAS and RAM-cached playback from NAS.
 

Reviews

Stereophile - October 2015
Recommended Component - Class A+

Positive Feedback - July 2015

Digital Audio Review - May 2015
 DAR-KO Award


Audiostream - April 2015
 Review Part 2


Audiostream - April 2015
 Greatest Bits Award


 

DX Rear Panel Connections

dx_backplate3

Brochure

 

Antipodes DS Music server

antipodes DS and power supply

Specifications

Fanless alloy case - Silver or Black
Auto CD Ripper to Uncompressed Flac
Paranoid-mode for bit-perfect rips
RAM-cached playback from internal & external sources, & streaming services
High-End USB Audio 2.0 Output
Plus auxiliary S/PDIF & Analog Outputs
1TB, 2TB or 4TB 2.5" HDD Storage, or
1TB, 2TB or 4TB SSD Storage
External Switching/Linear Power Supply
110-120VAC, or
220-240VAC
DS: 230mm (w) x 220mm (d) x 80mm (h)
PSU: 70mm (h) x 90mm (w) x 130mm (d)
Shipping weight 5kgs
Warranty 2 Years With Registration
 

Description

The DS uses the same core server technology as the DX, and so is similarly open, dynamic and musically expressive.

 The DS achieves a lower entry price point than the DX mainly by using customised 2.5" (laptop) spinning drives, and an innovative new switching/linear power supply (fully linear regulated output).

 The innovations in the new 3rd generation DS dramatically reduce the noise floor compared to all previous DS models, eliminating digital glare and grain, for a huge gain in musical expression.
 

Reviews

AudioStream Geatest Bits Award

 Previous models have been awarded:

6moons Blue Moon Award

Audiophilia Star Component Award

Witchdoctor 5 Stars


 


DS Rear Panel Connections

ds_backplate

Brochure


Feature Set Overview

                       load music

Ripping CDs
Auto-rip CDs to Uncompressed FLAC
Paranoid-mode for the best possible rip

Manage Music Files
Music files are 'Shared':
Copy/paste/delete files using your PC/Mac
Drag & drop music files to the server
Download directly to the music server

Manage Tags & Cover Art
Use any tag editor on your PC or Mac
Open the files with your tag editor (eg, JRiver)
Make the tag changes
Add Cover Art
Saving the changes updates the server
 

                  Play From

RAM-caching for the best possible latency

Internal Storage
 Best sound quality is from internal storage
 Range of upgradeable storage options

External Storage
 Second best sound quality is from a NAS
 Third best sound quality is from a USB HDD
 Internal & external files form one library
 But can also be accessed separately

Internet Streaming Services
 Access to internet radio stations
 Access to Premium streaming services
 Includes: Spotify, TIDAL, QOBUZ, etc
 

                Play To

USB DAC Input - For Best Sound
 Fully compatible with USB Audio 2.0
 Bit-Perfect PCM to 32bit/384kbit
 Bit-Perfect DSD64 & DSD128 (DoP)

SPDIF DAC Input - 2nd Best Sound
 Designed for use with older DACs
 Bit-perfect PCM to 24bit/96kbit
 Higher resolutions trans-coded on-the-fly

Ethernet DAC Input - 3rd Best Sound
 Stream any file to any Streamer DAC

Amplifier Input - 4th Best Sound
 Designed for use before selecting a DAC
 RCA Stereo Analog Connection


 

Remote Control

Playback can be controlled remotely from a PC or Mac, but for the best user experience, use any SqueezeBox-compliant app for tablet/smartphone - available for purchase/download from the relevant application store for your device. We recommend iPeng8 for iPad/iPhone, Squeeze Ctrl or Orange Squeeze for Android, and Soundicity or Squeeze Remote for Windows Tablet. These are solid, mature and well-supported applications that benefit from ongoing development.


tablets1

Technology Stack Overview

Precision-clocked USB Audio 2.0 Outputs Plus Auxiliary S/PDIF & Analog Outputs.

Custom Scripting To Optimize Sound Quality.

VortexBox Integrated Software Suite With Fedora Linux Operating System.

Custom Firmware To Manage Each Individual & The Combined Chipset Noise Spectra.

Fanless Intel-Based Industrial-Grade Main Board & Quiet 2.5" HDD or Silent SSD Storage.


Design Philosophy


A Digital Signal Is Not Just Bits

A digital signal sent from a digital source to a DAC is an electrical or optical wave. The DAC's receiver reads the wave to identify the bits. When the wave has electronic noise interference and/or timing errors, the bits can still be read but the bits are not recognized with perfect timing and this results in poor sound regardless of the DAC used.
 


Buffering Does Not Eliminate The Problem

A common fallacy is that the DAC can simply buffer and reclock to fix the problem. But a buffer that is written to and read from at the same time does not eliminate jitter, it just changes it. This is because the process of writing to the buffer interferes with the process that reads from the buffer because they are not able to be physically separated. They are described as separate processes in the theory when you learn it at University, but they are only logically separate, not physically separate. Faith in theory, without empirical testing in the intended application, is dangerous because all theories are gross simplifications of what really happens, so don't fall for the 'bits is bits' or reclocking rhetoric.

 The only buffer process that works the way it is claimed to is when you take a noisy source and write it to storage, and then you read the stored bits after the writing has finished. This is essentially what ripping to a hard drive achieves, and any noise and timing error from the CD reading process is eliminated. But this is not the same thing as a buffering and reclocking process - and to suggest that it is the same, or even as effective, is very misleading.
 

Antipodes' Solution Is To Minimize & Manage Noise

Antipodes Music Servers provide what we have discovered to be a much better solution - to feed the DAC with a precision-clocked ultra-low noise digital signal. Unlike most other servers on the market, we don't use Ethernet streaming, large noisy 3.5" hard drives or noisy power supplies; and then rely on caching, buffering/reclocking or noise filtering to try to fix the mess.

 Playback is cached in RAM (not in higher latency storage) for the best possible playback architecture, but sound quality will still be directly affected by any electronic noise interference with the signal between RAM and output, and by unnecessary processes being performed in RAM at the same time.

 It is important to emphasis that noise interference can come from any powered item in the server, such as any hard drives, displays and other peripherals connected to the system system they do not need to be in the signal path to affect the signal path. And noise filtering should be minimized or preferably avoided. In our experience the best sound quality comes from not generating high levels of noise in the first place, and avoiding the need to use noise filtering. Adding noise filters seems to always reduce the fidelity of a low-noise system.

 Antipodes Music Servers focus on eliminating anything unnecessary, by using a minimal Linux operating system, and by pushing all interaction to your tablet, smartphone or computer; and through minimization and management of electronic noise at every step of the process that generates the digital signal from the stored file.

 The first step is to keep RAM activity to a minimum and this is achieved by using a purpose-built VortexBox Linux software suite and customized scripting to fully optimist how the software is employed to play the music files. The second step is system tuning through chip selection and customized firmware to tune the chip speeds, which affects not just the amount of electronic noise generated but also its frequency spectrum. Electronic noise in different frequency ranges can have vastly different effects on the resulting sound quality. At a total system level, it means avoiding different chipset noise spectra coinciding to create peaks, and ensuring the residual noise is placed in frequencies that do minimal damage to the digital signal. We liken this to the process of vibration tuning. An equipment rack does not and can not eliminate vibration. But a well-designed equipment rack will smooth the energy peaks, release the energy quickly, and channel energy into benign frequency bands.

 The design of Antipodes music servers also involves using over-specified power supplies designed for the task and as part of the tuning process and using customized electronically quiet storage media. But the superior sound quality achieved is much less about the parts we use than it is about the proprietary tuning of the total system to minimize electronic noise interference with the digital signal sent to the DAC.

 


DAC Connections

In the context of Antipodes music servers, USB is the best connection. With a noisy server (most other servers), using an Ethernet input to the DAC makes sense because it provides the greatest isolation from noise interference with the digital signal. But Ethernet generates high levels of noise within the DAC. This means that Ethernet is a band-aid, not an optimal solution.

 SPDIF, i2s and AES/EBU suffer because the system clock is in the server, not in the DAC where it really should be, and this is typically heard as dirty and less coherent sound.

 USB is ideal for an Antipodes server because it is architecturally superior to SPDIF, i2s and AES/EBU, and generates much less noise in the DAC than Ethernet. But it needs to be emphasised that USB is the best solution only when the server is very low noise, and when the DAC manufacturer has done a competent job of isolating the USB receiver from the analog power supply and circuitry - fortunately most have. But some DACs may sound better via the SPDIF digital output of an Antipodes music server, for this reason. Using Ethernet will always sound softened with images being less distinct.

 There are no simple mantras for you to grab hold of here - everything comes down to implementation - and only your ears can tell you when a piece of equipment has been done right. If you listen to one of our servers, the most obvious improvement is the organic and solid nature of the instruments and voices, along with elimination of the glare and grain that all other digital sources cause. Or you might check out the rave reviews of the Antipodes DX which tell you the same thing. Either way, we hope you will begin to appreciate that there is a vast superiority to providing a clean and accurate digital signal, as opposed to just getting the bits right.

 


Storage Options

The key thing about storage is the electronic noise it adds to the system. Just having a spinning hard disk connected and powered generates electronic noise regardless of whether or not it is spinning at the time. Using an SSD for buffering has little benefit beyond reducing the heat and vibration of the hard drive spinning, and it really only becomes necessary if you are using the large 3.5" hard drives. For this reason Antipodes only uses 2.5" hard drives in the DS, which are physically much smaller than a 3.5" drive, eliminating the heat and vibration issues. On top of this the 5v interface on a 2.5" drive generates less troublesome electronic noise than the 12v interface on a 3.5" drive. Antipodes goes further and customizes the firmware for the 2.5" drives to improve their performance in a music server.

 Remember that playback is from RAM in an Antipodes server, not from disk storage. Files are transferred in block mode into RAM for playback and so jitter at this step is irrelevant. Low latency is beneficial but otherwise the key sound quality impact is the presence of a powered disk in the system.

 Hard drives add a bright hard edge to the sound and a coarse grain that flattens the sound dynamically and tonally. Using 2.5" hard drives means much less of this problem. But removing the hard drives and using internal SSDs (in the DX and as an option in the DS) increases tonal color, liveliness, naturalness and ease, but only when all hard drives are removed from the server and have no electrical connection with it.

 Playback from a NAS is a cost-saving measure to reduce storage cost. The best way to play from a NAS, from a sound quality perspective, is to mount it to treat it as if it was internal storage, rather than use DLNA, because mounting the NAS folder reduces Ethernet activity compared to DLNA. Ethernet activity generates noise in the server. The effect is to soften the sound and images get less distinct; less organic; less real. This may be more palatable than the effect of using internal hard drives, but using internal SSDs is decidedly superior.

 The key issue with storage is the same as with the rest of the design of the music server. It is about electronic noise interference in the server with the digital signal being sent to your DAC. And noise filtering just shifts the problem, often making the sound worse. Different storage options have different effects on the sound, and there are no easy short-cuts. Unfortunately the best sounding option is expensive - using internal SSDs.

 The best sound comes from using internal SSDs. The next best sound can be achieved with playback from a NAS, using mounting not DLNA, but the audio performance and the system-responsiveness will be affected by network quality. The next best sound performance is achieved with internal hard drives. The next best performance is achieved with external USB hard drives - because the USB hard drive electronics are generally electronically very noisy.

 One key point to add is that the storage choice is just one part of the solution. There is no point in spending on SSDs until the rest of the server has achieved low noise, as the benefits of the SSDs will not be heard. If anything, the storage choice is the last one to focus on. Getting the server technology right is the first priority, the output stage is second, power supply third and storage technology is fourth. For example, the DX playing from a NAS sounds a lot better than a DS playing from internal SSD because of the superior power supply in the DX.

 


Power Supplies

This is a complex area and we will only make some brief points. The key point is that it is not as simple as choosing between switch-mode, linear and battery supplies. As stated earlier, there are no simple mantras, and particularly not with power supplies.

 Some parts of a music server perform better if powered with a switch-mode supply and the key design issue is selecting a topology that minimizes the high frequency noise interference generated. We find battery supplies to be quite inferior to certain topologies and implementations of linear regulated supplies, so we do not use them. On the other hand, most linear topologies actually sound worse than the best switch-mode power supplies.

 And then there is the issue of external versus internal supplies. We find an internal arrangement is the best because the DC feed can be kept very short, but it requires a specially made and expensive power transformer. A conventional transformer located in the case with the server will sound terrible due to the leakage flux interference generated.

 One of the key issues is where the power supply noise is placed, in a frequency spectrum sense, and this can be more important than the total amount of noise. For example, batteries generate noise in a particularly bad place and the result is that the music loses its speed and life. Similarly, using a generic linear power supply, or even an audiophile grade supply on a server may result in really poor sound, depending on the topology used. What is best for an analog circuit and what is best for a digital circuit, or even for different parts of a music server, can be very different.

 Antipodes music servers are powered by linear circuits feeding a combination of linear circuits and a small number of localised switch-mode circuits. Batteries are not used. The DX uses multiple internal linear supplies with a special transformer design, and employing a particular linear power supply topology. It is expensive to do it properly and we have done it properly in the DX. The DX power supply, aside from the SSDs, is by far the most expensive part of the DX, and arguably its most important part.

 To achieve a much lower price point with the DS, we have to do the power supply differently. The first area of saving in the DS power supply arrangement is using an external power supply, and the second is that we use some switching elements in the power supply. However, the regulated output stage of the DS power supply is still fully linear.

 Just as you will find in the design of any other high-end audio equipment, like a DAC, amplifier or speaker, these design decisions are based on massive amounts of time building prototypes and listening to them, and then refining the design over a period of years. The design of a great sounding music server is more challenging than designing a great sounding CD transport, and is far from being a conventional computer.