Don't we have June yet? ...you can see the headline as a representative of what I have been expecting from Audinate for a long time, updates, information and proximity to the end customer, especially when it comes to support. The latter is known to be part of the company's philosophy and strategic shielding practice. Should the licensee mess with the user Audinate is more of a supplier. Then came the invitation to the event “Audio-over-IP in Broadcast with Dante” in Cologne. ...of course I have to go there and knock off a few topics that so far have been rather speculation the official announcement.

AES67
Unlimited freedom for audio-over-IP

AES67 aims to enable interoperability between different audio-over-IP solutions. In short, patch a Dante source to an AVB sink or vice versa. The whole thing then works for the big top dogs Dante, AVB, RAVENNA, Livewire and Q-LAN, the healthy network world is ready. Let's take a look at the current situation. AES67 was published on 11.09.2013. The paper can be obtained here. http://www.aes.org/publications/standards/search.cfm?docID=96  It describes on 43 pages quite comprehensibly the AES67 NOT an independent protocol is what you simply implement and make the existing above solutions obsolete. It does not compete with the protocols, but rather builds on them by standardizing the lowest common denominators. i.e. the manufacturers of the etc. Protocols implement an AES67 compatible mode and then signals can be exchanged with each other. It defines minimum standards or lowest common denominators such as clocking with IEEE1588-PTP, 48kHz, uncompressed PCM, 1ms latency.   That sounds promising...when can I use it?

At the Tonmeistertagung in November 2014 ALC NetworX reported about a big Plugfest in October 2014 on which this interoperability was successfully tested. At the Audinate event in Cologne, NTP reported on their experiences with the implementation of AES67. But first, take a step back. When is the Audinate Update coming with AES67? Don't answer at all! The manufacturers/licensees implement AES67 and Audinate helps if necessary. That is to say, the manufacturers just have to do it, Audinate does not owe an ‘announced’ update here. This now leads us to the Danish manufacturer NTP, which has a broadcast routing matrix in its program with the Penta and wants to implement AES67 for Dante in order to be able to exchange signals with Ravenna and Co. Ravenna is a protocol of ALC NetworX, which in turn belongs to LAWO. Therefore, the distribution in the broadcast segment due to the market shares of LAWO. Ravenna is often presented as an open system that anyone can easily implement, but license costs are also due here. NTP hangs with its products fully in the middle and has to operate everything first. So Dante has to be in AES67 compatible mode. there you are right there and also reports from the Plugfest in October 2014, but it probably did not work as smoothly as shown by ALC. The first problem is the lowest common denominators...i.e. are 1ms and 48kHz sufficient? Which revision IEE1588-PTP is used and how is RTP implemented? This is not seen the same everywhere and is not necessarily compatible as a result. The second problem is that the implementation of the Session Description Protocol (SDP) in AES67 is not sufficiently detailed and requires revision. Until then, you have to think about who exactly you want to be compatible with via AES67 and then implement it concretely. For a manufacturer like NTP this is probably Ravenna/Lawo but this is speculation and it is also about 6 months away from “finished!”

CONCLUSION:

AES67 is cool, but it still takes time to become practical!

...you have to keep in mind, however, that this is really pulling the string. Release of the standard 2013, Plugfest exactly one year later....it remains exciting. The nice thing about Dante is that no matter which manufacturer of the approximately 200 licensees builds a product here, it ALWAYS PLAYS WITHONE! ...the research work AES67 Implementation seems to be driven forward by those where it is hottest, these are probably the manufacturers of routing scratches in the broadcast segment who have to route to LAWO/Ravenna. I wonder why you don't turn the tables and 200 licensees use their power and agree on an AES67 specification under the direction of Audinate, which then ends in an AES67 revision and the approximately 20 Ravenna licensees then have to do it like the big audio majority ⁇

Control room at bridge
AESX210 and the universal head-amp control for everyone.

Let's talk about compatibility. Not only should the audio signal be compatible, the control data should also be compatible. The manufacturers of the aforementioned routing chips struggle with about 200 different control protocols to automate broadcast studio environments. Even if you think an IP router is an IP router is wrong, just as with the STP there are many procedures that all do not play easily with each other and are always made briefly compatible in individual cases and done. The fact that the internet works is a miracle. No one benefits from this confusion and everyone wants standards but everyone also wants to have a say and not always there is ONE solution but several approaches because for individual non-negotiable borders are crossed and then new groups are formed. For Internet Routing, Open Flow and Software Defined Networks are first attempts and for Broadcast Matritzen Amber+ and OCA. However, it is usually enough to set here from patch points. What Amber+ does exactly I don't know but it probably comes from the Ravenna corner and is a kind of open source protocol. The OCA Alliance defines standards for parameters that ProAudio users all use such as phantom power and filters ON/OFF, gain, frequency, quality, etc. The AES project AESX210 is to convert the work result into an AES standard later this year. Hell Yeah!, which Yamaha just announced a cooperation with Focusrite and makes the RedNet converters mountable AND controllable in the CL series, on the other hand, seems like a children's birthday. This shows me who does not live the open network idea will be the first to die out!

You're not coming in here!
Security requirements and other updates

Protecting audio networks is an issue in itself. First of all, you have to distinguish between sabotage and accidental mishandling. In the end, there are only the following possibilities for sabotage:

• Layer 1: thick security type with fat gun in front of the engine room
• Layer 2: Bind Mac address to port
• Layer 3: Subnets and routers

A feasible way for hidden permanent lines is, for example, to switch off the discovery mechanism of the devices after a successful Dante patch and make them invisible to the rest of the network. You need the help of Audinate, but you can. ...only bringing it back is something “pain in the ass” (O-Ton Kieran Walsh/Audinate)

I also learned some interesting news about routing. It seems that many R&D colleagues with whom I speak so Ravenna have been introduced as the only solution with which routing is also possible via the Internet. I always wonder where they've done this before? Good in the radio house with routers in the IT infrastructure OK, but Cologne with Düsseldorf? With Dante, that's probably all, so the CBC gets e.g. audio signals from the White House via Dante via the Internet into the studio ...”it is” again something “pain in the ass” and not without solving Audinate but it's possible. Of course reduced to unicast connections especially the clocking has to be tinkered here.

Audinate is working on important security updates and wants to deliver within 2 years. You have to keep in mind that everyone from the department store to the BBC Studio wants to have a say here and first have to set boundaries.

First, a few things I want but won't work:

• Master/Slave Dante Controller: the first controller instance reports to the network and the opening of further distances is only possible upon confirmation from the master or not at all. Unfortunately, this is not possible because the Dante Controller is not a program in the true sense, but only visualizes the endpoints.
• Cross-manufacturer offline library: Of course, I have to plan and prepare for several manufacturers without always having the hardware at my desk. This will not be the case for Audinate and will always remain the responsibility of the manufacturer/licensee.  In practice, I have to prepare a Yamaha island and a Lake island and eventually marry manually via Dante Controller. Maybe that's not so bad with all the mergers.

What will happen, however, is the routing via several subnets and user level access to the Dante channel. This, in turn, already plays at Dante Via and was vividly demonstrated.

Dante Via
The nutcracker for all audio drivers

Soon it will be time ... for Dante Via until about August 2015!

The price? ...is not yet fixed but of course affordable as a mass product, which leaves less than 100$ hope.

With Dante Via, all audio interfaces and software applications can be networked. Without Brooklyn module or other Dante hardware in the network. Dante Via is a pure software product. You can route a signal from audio interface computer A to audio interface computer B or the software output of a DAW like Logic on computer A to audio interface or DAW or other software on computer B. Very exciting thing! The latency is deterministic depending on the interfaces and not sample-synchronous as in the Dante network and will be about 40ms on the network side, but this is not yet exactly clear. A possibly necessary sampler rate conversion makes the operating system or its sub-programs such as Quicktime at Apple in normal operation on a computer does it as well. The maximum number of channels is also not yet finalized and will be around 16 stereo channels. You want to listen to a Dante channel? ....Zack no problem simply route to the headphone output from the computer. The patching must always be confirmed by the owner of the source. User Level Access already works on a channel basis! ...now Dante Via only has to run on Linux and then you can build fun things like intercom and listening tools with Raspberry Pi.

left