AVB works over ethernet networks, but as Kevin Gross explains on the LinkedIn AVB board – only networks with compliant equipment in the network – and by extension (which was my mistake of hope over reality) never outside of an internet.
Looks good until I get to this bit, “AVB can be integrated into any existing Ethernet network through the installation of AVB certified devices.”
All network equipment between AVB endpoints must support the new 802.1 AVB standards. “Existing Ethernet networks” generally do not meet these requirements. Deploying AVB involves deploying new network equipment as well as new or upgraded endpoints.
What does this mean to a cinema exhibitor who might want to create clever “between rooms” games or other interesting alternative content that relies upon audio and video signals getting passed around in a timely manner?
Later in the board thread it also becomes clear that while the audio portion is working well with many manufacturers, some video AVB ideas are waiting for technology jumps.
In this conversation from the thread, Cindy Davis is the editor at TechDecisions Media while the others are experts in this field (we’re working on getting permissions to post these comments – don’t read them until all permissions are granted):
Warren J. Osse CTS-D, DMC-E • Interesting article. The video part of this equation is more complicated. You have mentioned “channels” of audio and video; the bandwidth for a single HD video signal would be difficult for most networks at present. Due to video bandwidth requirements, audio seems to be the focus at present. Networks will need to speed up for the video aspect to function well. The adoption of this transportation mechanism does require very current Ethernet switches; note that AVB is Layer 2 at present. It works well at what is is designed to do. AVnu is doing a great job of maintaining compatibility and helping this fledgling technology move in to the mainstream.
Cindy Davis • Thanks Warren, We are going to do a follow up article on, “Preparing Your Network for AVB.” Would you be available for an interview on this? We will be looking for a few people to interview for this. thank for your feedback. Cindy
Thomas Mullins • I have to agree with Warren, and this is further hampered by the lack of support from the video manufacturers; the vast majority of supporters are pro audio companies. On a 1G switch, with 25% reserved for normal network demands, that leaves 750M of space left for audio and video. Barely enough for two HD video signals with embedded audio. Even if you take advantage of H.264 compression, the channel count is limited. I haven’t found many of the streaming companies (such as VBrick) who are interested in AVB. They see no need.
Kevin Gross • HD video can be anywhere from 3 Gb to 3 Mb or less. It just depends on how much compression you want to do. The bandwidth problem is actively being worked from both ends – improved compression and increased network capabilities. If there is still a problem, it won’t remain unsolved for much longer. Networked video is already in widespread use in IPTV, Netflix streaming, DVR sharing, security camera systems etc. We haven’t seen these applications adopt AVB. What do they gain by doing so?
Bill Murphy • Hello Cindy, Thank you for your article and editorial interest Cindy. Your mix of products for AVB includes I believe some which are shipping and some near shipping. My company, Extreme Networks, makes a wide range of AVB ready switches up to 40 GB that will require a software bump to support AVB. We were the core switch for the AVnu booth demo and are well known to the AVB endpoint vendors. We also provided demo kit on several booths. If you look at the Audinate video from the show you will see our switch talked about there as well. We discussed the issues the network faces in supporting AVB at the show as a regular daily presentation on the AVnu booth.
Bill Murphy • Good to meet you at infocomm Kevin. I look after AVB and Physical Security for my company. For Security, most video is not associated with audio. The need for accurate sync for the video in that industry is mostly for video which could end up in court as evidence. On the Pro-Video front, there is interest for high-speed ports (40Gb). It is driven by the desire to reduce cabling requirements and move from the SDI cost curve to the ethernet cost curve. As has been said, it is moving more slowly than Pro-Audio.
Erich Friend • It is the ‘built-in network manager’ features that makes AVB attractive to many end-users. Most don’t have, or desire to get a BICSI certification just to make a network function. They want plug-n-play, not plug-n-pray. The reduced hassle-factor of the AVB system promises to make systems configuration on-the-fly in live entertainment scenarios much more palatable. The added benefit of control, audio, and video arriving in sync is just icing on the cake. When the end-users begin to understand the convenience of AVB, they will demand it from the manufacturers. On a similar note, I have posted an article about the recent NATEAC conference and the discussion there about the future design of stage lighting [control] systems (http://theatreface.ning.com/profiles/blogs/a-gathering-of-minds-nateac-report). The distribution of control signals is, I believe, best achieved by an AVB compliant control network, but it will take the ‘waking-up’ of the stage lighting industry before this can be achieved. Educating those that define the ACN and RDM-net protocols about the tweaks necessary to ride the AVB train would be a good prodding for AVnu to pursue. With apologies to Dire Straits, a parody of their ‘Money for Nothing’ song might go something like this. . . ”Now look at them yo-yo’s, that’s the way you do it You play your program on that AVB That ain’t workin’, that’s the way you do it Money for nothin’ and your bits for free Now that ain’t workin’, that’s the way you do it Lemme tell ya, them guys ain’t dumb I want my A V B I want my A V B . . . “
calvin williams • I wanted to start installing AVB gear for video, I was there for all presentations on AVB at INFOCOMM one of the presenters told me Iam looking at 3 to 4 years away. or may be longer, The presenter said most of the video companies do not want to spend money on R&D no ROI.
Finally, while we are on this topic of AVB, John Grant tosses an elegantly honed spanner into another AVB discussion:
There are lots of “forum” or “alliance” or “trade association” organisations, mostly headquartered in USA and charging a fee which big companies won’t notice but which discourages small companies from joining. A cynic might think they’re organised as a way of shutting out the smaller, more innovative, companies without actually violating the US anti-trust laws.
There are other ways to organise compliance testing; one example is the EBU work on audio over IP. No doubt a similar initiative could be organised within an appropriate group such as the AES.
This discussion link will be interesting for policy wonks for its replies that defend the hard work of usually sincerely interested engineers doing unglorified work, even while working and doing what seems right for their large corporations…or small. If someone has an idea they don’t want to see it railroaded over by what they might feel is an ill-formed standard that will restrict their concepts. Though, sometimes it is corporate greed in its worse form…something I’ve only read about… and certainly it has never been done in the entertainment technologies field.