UK Sales: 0191 418 1122         International: +44 191 418 1133

Menu

www.canford.co.uk

  • Next day delivery
    available
  • Free standard UK delivery
    on all orders over £100
  • Industry leading
    customer service
  • Free technical support
    from our team of experts
DIGITAL AUDIO OVER ETHERNET NETWORKS

The use of Ethernet to transport audio digitally offers several advantages over analogue cables:

Physical

Shielded, twisted-pair, multi-pair cables are heavy, expensive, and difficult to deploy, almost unmanageably so above, say, 24 channels. Digital audio networking can transport a much larger number of channels on a single Ethernet cable.

Costs

The purchase of Ethernet cable results in significant savings on the cable and connectors, and labour costs to deploy it are considerably lower, not least in that the cable may already be installed in the building.

Signal Integrity and Range

In analogue audio systems, high-frequency components are progressively attenuated with increasing cable-length. Depending on the quality of the sending and receiving circuitry and the cable itself, this effect , as well as signal-level reduction and reduced immunity to interference, can be noticeable above even a hundred metres. Specialised send and receive circuitry, including balance and frequency response adjustment, is not unknown in broadcast environments, where studio to control-room cables lengths in excess of 1.5km have been known.

In contrast, while each Ethernet copper segment (10BASE-T up to 1000BASE-T) is restricted to around 100 metres in length, a fibre segment can reach up to 550m quite commonly (1000BASE-SX), up to 2km without optimisation 1000BASE-LX) and up to 10km using several standards, with 100km available for long-haul applications. Ethernet can be bridged to other networking formats making longer range possible. The signal degradations mentioned in the previous paragraph do not apply to a properly implemented Ethernet solution.

As one would expect, there are disadvantages.

Latency

A single stereo, broadcast-quality .WAV file can quite easily be streamed over an Ethernet network without any audible problems, depending on other traffic, using, for example, Windows Media Player to play music stored on another PC.

Since both channels are embedded in the same file, both channels are subjected to the same delays, as the transmitting PC reads the file from storage, feeds it to the network, and converts the data stream into Ethernet packets, the packets take a finite time to traverse the network and the receiving PC then converts the packets to a data stream to play through the PC sound card.

These delays are variable, depending on other activity in either PC, and occur in a normal situation, without considering any further delays due to any packet collisions or other traffic on the network.

If instead a multi-channel audio stream is considered, which originates from one network host, the various channels to be received by several different hosts of different configuration and at different distances from the source, it is clear that the channels will be out of synchronisation by the time they have been processed and leave the different hosts either as analogue or digital signals or as acoustic waves.This issue is exacerbated as the channel count rises and as a result the throughput of the network goes up.

Thus any digital audio networking system will have delays or latency. What is worse, it will not be constant unless steps are taken to make it so. In contrast, the electrical audio signals in an analogue cable travel at the speed of light. Delay can be introduced by the analogue circuitry at both ends of the cable, but is generally insignificant. Several implementations of digital audio over Ethernet control latency, so that it is constant from device to device, and when this constant delay is small enough, depending on application, may also be considered insignificant.

Costs

The additional cost, in components, of converting the analogue audio to digital audio and vice versa, and of hardware to connect the digital equipment to Ethernet, is not insignificant. Cost of labour to fault-find or analise problems may well also be higher; significant training is not always required to check an analogue signal path through from beginning to end. In a system where the required end result is multi-channel, digital audio, in a system operating over long distances or to/from multiple locations or multiple different equipment, these disadvantages are mostly outweighed by the cost advantages of the digital solution.