Recording - June 2014 - 56

Most of us now find recording and mixing digitally a common occurrence compared to 10-15 years ago. Digital technology brings with it the promise of better connectivity and interactivity amongst pieces of
gear, computers, and software. A natural
extension of that idea is networking-
numerous devices all communicating about
their capabilities, eliminating some of the
restrictions posed by the point-to-point
wiring we're all accustomed to when using
analog. Just as with setting up a computer
network for an office, where multiple
devices can share hardware like printers
and servers, a network of audio devices
should be able to share its resources. Ah,
but that's easier to say than do.
With a computer network, there's a set
of industry standards that allow devices to
communicate; there's rules for Mac users
and PC users. Follow the rules and it
works (most of the time-you can tell I'm
not an IT guy, right?); you find all the
printers, you share files, etc..
But audio is a bit different. What works
for your typical printer will not work for your
mic preamp, digital audio workstation
(DAW), mixing console, digital snake, etc..
Your typical Ethernet-based computer network has far too much timing unpredictability and latency to be practical in a
live performance or recording environment
where timing is everything. It's totally unacceptable to have the drummer's timing and
feel expanding and contracting by even a
couple of milliseconds. For these reasons,
many audio guys have found networking to
be confusing and frustrating.
As a result, manufacturers have been
coming up with proprietary protocols for
distributing digital audio to their products
for years now, and it's for a simple reason-there really isn't one good way to network products together that's available to
everyone. Audio devices in a professional
environment require precise synchronization and sample-accurate clocking. So
Manufacturer A's products cannot necessarily connect to and communicate with
products from Manufacturer B, and so on.
That's where a protocol such as
Audinate's Dante comes in. Dante is an
audio over IP (Internet Protocol) networking technology that becomes the backbone for the communication between
56

RECORDING June 2014

devices, regardless of who manufactured
them. Dante communicates over simple,
readily-available Cat-5 (Ethernet) cabling
and has lower installation and transport
costs, while at the same time eliminating
some of the other nasty problems that
plague analog systems: heavy and costly
multi-core cables, RF interference and
noise, cumbersome routing, and difficulties when attempting to split the sources
for multiple users.
Of course, Dante is not the only audio
networking technology out there that has
been designed for the specific needs of
audio (and video) professionals: DigiGrid,
CobraNet, HiQNet, MADI, A-Net, and
EtherSound have all been introduced to
help solve some or all of the same set of
problems. Audio Video Bridging (AVB) has
been in development for about five years
now, and it too promises to become that
universal pipeline that finally does for audio
gear what Ethernet does for computer gear.
I'm focusing on Dante as an introduction to
this technology because it's widespread,
adopted by many companies that are not
all part of the same larger corporation, and
easy to understand; the basic principles will
carry to other networking standards in
many cases.
So how does it work?
Dante runs over standard, off-the-shelf 100
Mbps and 1 Gbps (Gigabit) networking gear
and switches, using readily available Cat-5
and Cat-6 cabling, and does not require any
special connectors. The protocol can flow
through standard Mac OS X or Windows PC
computers using the Ethernet ports they
already have; no special sound cards or
adapters are required. Dante allows devices
and their individual channels to be labeled,
and those labels flow throughout the network
to all connected devices. That makes it easy
to keep track of what you're doing as you
route your audio signals. The protocol supports various sample rates from 44.1 kHz to
192 kHz and can support mixed sample
rates in the same network.
Audinate also offers the inexpensive
Dante Virtual Soundcard (DVS) software, a
simple way to add your computer and
favorite digital audio workstation to a Dantebased audio network. With Dante Virtual
Soundcard activated, your computer and its

audio software see another source for input
and output routing. Dante Virtual Soundcard
can be configured in a variety of I/O sizes,
depending on your needs: 8 x 8, 16 x 16,
32 x 32, and 64 x 64 channels are available (at a 48 kHz sample rate). Figure 1
shows the basic control panel for DVS.

Setting up a Dante network
You need at least two Dante-enabled
devices connected with Cat-5 cables to create
a network. The devices discover each other
as they're added to the network, so there is
no need for individual device addressing. For
routing channels from one device to another,
you'll need the Dante Controller software,
which is available free from the Audinate
website (www.audinate.com). And, of
course, you'll need a computer connected to
run the Dante Controller software.
Dante Controller is used to set up the
clocking, switched versus redundant
modes, and names for devices and their
channels. Once your routing is completed, the Dante Controller software and its
computer can be removed; the devices
will remember their routing assignments
until they are changed. Of course, if
you'll be using the computer as a Dante
Virtual Soundcard, it should remain connected as part of the network.
Dante Controller is basically an X/Y tablestyle grid of inputs and outputs; devices with

http://www.audinate.com

Recording - June 2014

Table of Contents for the Digital Edition of Recording - June 2014