Dante Network Audio Technology

Dante makes use of "off-the-shelf" network equipment. The integrator should employ standard VoIP-style QoS in order to integrate Dante into existing network infrastructures alongside other applications. Dante uses standard Voice over IP (VoIP) Quality of Service (QoS) switch features, to prioritize clock sync and audio traffic over other network traffic. QoS is available in many inexpensive and easily obtainable Ethernet switches. Any switch that supports Diffserv (DSCP) QoS with strict priority and 4 queues, and has Gigabit ports for inter-switch connections should be appropriate for use with Dante.

The appropriate switch settings regarding QoS are as follows.

• QoS mode: Basic
• Trust mode: DSCP
  Please beware that the bus names are limited to 31 characters and special characters such as &, % and @ are not allowed in the Bus Names.

Latency. is deterministic in Dante systems. That is, it is not dependent upon circumstance, but instead, well known and consistent within the system all of the time.

Each Dante receiver introduces a specific, user-adjustable amount of latency before playing out audio to account for any delay variation in the network or end device, and this guarantees that all devices play out in perfect sync. The Dante Controller sets this latency per device, and the value selected should be based on the size of the network. Some examples are shown below:

It is a "system-wide network latency setting". So, how much additional latency or skew is there for each switch hop? As it turns out the answer is none. As long as the (system-wide network) latency the user sets is long enough to cover all the switch hops, the packet will get to the end of the chain before the buffer expires. Then Dante uses its clock alignment mechanism (Precision Time Protocol) to exactly line up the outputs down to the sub-sample level. You can compare the word signals between any two Dante units and they should be accurate within 100 nanoseconds of each other, or less than 1% of a sample period (checked up to 10 units). So the outputs at multiple points in the daisy chain will all be sample aligned, each with the specified latency from the sender.

The latency chosen in Composer is not additive, and it only needs to be long enough for a packet to traverse the entire network path. The only practical guide to give a user for this (beyond extensive network path and latency analysis using dedicated software) is the number of switch hops, but it is not additive.

_Note: The actual Dante latency will be 3 samples or 0.06 mS longer than the time indicated in the Dante latency settings in Composer. This is due to the sample buffering our DSP does and is unavoidable. _

There is a self-discovery "Plug and Play" process on the Dante network that will remember previously configured and Dante equipped units as they come into and out of the network. It is recommended that you use Gigabit switches when creating the Dante network. The latency setting is network wide and is not based on the number of units, or buses, but the number of switch hops. Edge and Radius can receive Dante buses from any Dante enabled 3rd party audio device.

A list of licensees of Dante technology can be found here:

https://www.audinate.com/meet-dante/dante-manufacturers-list

A list of 3rd party products which will be capable of passing audio to Edge and Radius can be found here:

https://www.audinate.com/products/dante-enabled

The routing and control of the Dante Network is configured by Composer Tools/Network I/O Manager. It is the name of the Bus which is it's significant identifier. Dante digital audio is limited to the parameters of Ethernet technology and makes use of standard VoIP and QoS switch features.