The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Reviews  Search Gear Database  Search Gear for sale  Search Gearslutz Go Advanced
Tags: , ,

Dante to AVB bridge
Old 3 weeks ago
  #1
Gear Nut
 

Thread Starter
Dante to AVB bridge

Asking here because it seems to be where the AoIP folk hang out.

Is anyone making a Dante to AVB bridge? I'd like to bring some MOTU AVB gear into the Dante fold (obviously could use ADAT I/O instead but, just curious).
Old 3 weeks ago
  #2
Lives for gear
 
emrr's Avatar
I know there were some nods from the Dante camp about doing this, but I haven't heard anything concrete.
Old 3 weeks ago
  #3
Lives for gear
 
just.sounds's Avatar
Old 3 weeks ago
  #4
Gear Nut
 

Thread Starter
Unfortunately using AES67 with AVB for sync whilst cool isn't going to get interop with existing AVB endpoints that operate at Layer 2 only. But maybe in time vendors will converge on AES67 and MOTU/Avid will update their firmware to support it. One can dream.
Old 3 weeks ago
  #5
Gear Head
 
albinotuba's Avatar
 

Biamp Tesira and BSS Soundweb London are both platforms with this capability. Neither will be cheap. You could pick up a Tesira Server-IO with an AVB card and a few Dante cards, or you could pick up a London processor with AVB and Blu-link and then get a Blu-Dan Blu-link to Dante converter. Both will be several thousand dollars.

EDIT:

Forgot about QSC Q-Sys. You could use that too.
Old 3 weeks ago
  #6
Gear Nut
 

Thread Starter
Quote:
Originally Posted by albinotuba View Post
Biamp Tesira and BSS Soundweb London are both platforms with this capability. Neither will be cheap. You could pick up a Tesira Server-IO with an AVB card and a few Dante cards, or you could pick up a London processor with AVB and Blu-link and then get a Blu-Dan Blu-link to Dante converter. Both will be several thousand dollars.

EDIT:

Forgot about QSC Q-Sys. You could use that too.
Thank you! Great to know that it's possible. Obviously, cost-wise, when the bridging solution is more expensive than the bridged device itself, it doesn't make a lot of sense
Old 3 weeks ago
  #7
Gear Head
 
albinotuba's Avatar
 

Quote:
Originally Posted by lukehatpadl View Post
Thank you! Great to know that it's possible. Obviously, cost-wise, when the bridging solution is more expensive than the bridged device itself, it doesn't make a lot of sense
Yeah you’re right. The cool thing with these platforms is that they can all handle thousands of channels of audio without breaking a sweat, so if you had hundreds of Dante devices and hundreds of AVB devices and hundreds of old Cobranet devices and you wanted to route audio between all of them, it can be done and it’s not that expensive compared to the total cost of such a system. Like you said, the problem is if you only have 1 or 2 Dante devices and 1 or 2 AVB devices, it doesn’t scale down that small very well.
Old 3 weeks ago
  #8
Gear Nut
 

Thread Starter
Quote:
Originally Posted by albinotuba View Post
Yeah you’re right. The cool thing with these platforms is that they can all handle thousands of channels of audio without breaking a sweat, so if you had hundreds of Dante devices and hundreds of AVB devices and hundreds of old Cobranet devices and you wanted to route audio between all of them, it can be done and it’s not that expensive compared to the total cost of such a system. Like you said, the problem is if you only have 1 or 2 Dante devices and 1 or 2 AVB devices, it doesn’t scale down that small very well.
Indeed. With the London, doesn't look like there's a version that supports both AVB and Dante – you'd need one of each and use BLU link to combine the two, if I'm reading it correctly.
Old 3 weeks ago
  #9
Gear Head
 
albinotuba's Avatar
 

Quote:
Originally Posted by lukehatpadl View Post
Indeed. With the London, doesn't look like there's a version that supports both AVB and Dante – you'd need one of each and use BLU link to combine the two, if I'm reading it correctly.
Yeah pretty much. They make a box called the BLU-DAN that can convert Blu-link to Dante, so you’d need an AVB-equipped London device and a BLU-DAN. The BLU-DAN is much less expensive than the other London processors.
Old 3 weeks ago
  #10
Gear Nut
 

Thread Starter
Quote:
Originally Posted by albinotuba View Post
Yeah pretty much. They make a box called the BLU-DAN that can convert Blu-link to Dante, so you’d need an AVB-equipped London device and a BLU-DAN. The BLU-DAN is much less expensive than the other London processors.
Ah, interesting. Do you know if they work at 96k? (I realise I should just Google this myself...)

EDIT: appears they can.
Old 3 weeks ago
  #11
Gear Nut
 

Thread Starter
Quote:
Originally Posted by albinotuba View Post
Yeah pretty much. They make a box called the BLU-DAN that can convert Blu-link to Dante, so you’d need an AVB-equipped London device and a BLU-DAN. The BLU-DAN is much less expensive than the other London processors.
Looks promising. One caveat I found in researching this is that the AVB devices cannot be synced to word clock (or Dante via BLU Link). This isn't ideal in my setup.

It would be nice if DAD and/or Avid implemented AVB in the AX32 (and/or MTRX). I've emailed DAD to find out as it was ostensibly planned for the DX32.
Old 3 weeks ago
  #12
Gear Nut
 

Thread Starter
Quote:
Originally Posted by albinotuba View Post
Forgot about QSC Q-Sys. You could use that too.
Doesn't appear to support 96k unfortunately (nor does the Tesira stuff). So looks like BSS is the only option.
Old 3 weeks ago
  #13
Lives for gear
 
tourtelot's Avatar
FWIW, I don't think that AVB will be a long-lived solution. Presonus, and I know it's only one company, has abandoned AVB. It would be better, although unlikely, if MOTU got on the Dante band-wagon. Again, IMHO, a better solution.

D.
Old 3 weeks ago
  #14
Gear Head
 
albinotuba's Avatar
 

Quote:
Looks promising. One caveat I found in researching this is that the AVB devices cannot be synced to word clock (or Dante via BLU Link). This isn't ideal in my setup.
My understanding, based on what the BSS guys have said in the training seminars I've gone to, is that AVB cannot be sync'd externally the way that Dante can. The AVB devices will elect a master clock and that device cannot slave to anything else. So in a network with AVB and Dante, the Dante devices would have to get their clock from the AVB bridging device.

Quote:
FWIW, I don't think that AVB will be a long-lived solution.
You're probably right as far as the pro audio industry is concerned, but commercial audio is doubling down on AVB. I don't think it'll go away any time soon.
Old 3 weeks ago
  #15
Gear Nut
 

Thread Starter
Post

Quote:
Originally Posted by albinotuba View Post
My understanding, based on what the BSS guys have said in the training seminars I've gone to, is that AVB cannot be sync'd externally the way that Dante can. The AVB devices will elect a master clock and that device cannot slave to anything else. So in a network with AVB and Dante, the Dante devices would have to get their clock from the AVB bridging device.
It does seem to be a limitation of BSS' implementation. MOTU devices that support both WC and AVB can resolve the latter's clock to the former, see here:
Resolving the master clock device to an external clock source
The MOTU device you’ve specified as the AVB network clock master can itself be resolved to an external time base such as word clock or optical. Just choose the desired clock source from its Clock Mode menu (in the Device tab). Doing so effectively resolves the entire AVB network to the external clock source.
So let's see, I just need another MOTU interface, a BSS BLU325 and a BLU-DAN to make this work in my ideal configuration
Old 3 weeks ago
  #16
Gear Head
 
albinotuba's Avatar
 

Quote:
Originally Posted by lukehatpadl View Post
So let's see, I just need another MOTU interface, a BSS BLU325 and a BLU-DAN to make this work in my ideal configuration
Or just an AVB-ADAT converter and a Dante-ADAT converter, which would cost less than half of that
Old 3 weeks ago
  #17
Gear Nut
 

Thread Starter
Quote:
Originally Posted by albinotuba View Post
Or just an AVB-ADAT converter and a Dante-ADAT converter, which would cost less than half of that
Right, the MOTU 8D or 112D is probably the most cost-effective solution when one consider's one's time and the requirement to interoperate with other MOTU gear (it doesn't support Dante, but I have a AES/MADI to Dante interface). Down the track, surely MOTU will have to get on the AES67 bandwagon.

Other option: Pivitec e64i/o-MADI although they seem to be hard to find.

Anyway, at the very least this post may be useful for the next person that is curious about this.

Last edited by lukehatpadl; 2 weeks ago at 05:13 AM..
Old 5 days ago
  #18
Gear Nut
 

I was just at InfoComm in Orlando, and went to the AES67 session, presented by members of the committee/authors of the standard. The question about AVB was raised. They said there would be no simple switch in the software. To date, there is only one company creating a "bridge" for AVB to AES67. The problem will be that this bridge will involve "proprietary" (as in non-standard) technology, because AVB originates from Firewire, etc, etc. Manufacturers using Auadinate's Brooklyn II chipset have a simple option in software to change output from Dante, QLan, Ravenna, etc to AES67. It will be that easy.
Old 5 days ago
  #19
Gear Head
 
albinotuba's Avatar
 

Quote:
Originally Posted by bhandy View Post
I was just at InfoComm in Orlando, and went to the AES67 session, presented by members of the committee/authors of the standard.
Oh cool. I was there too! I didn’t go to that session, though.

Quote:
Originally Posted by bhandy View Post
The question about AVB was raised. They said there would be no simple switch in the software.
I’m not sure what this means. I wasn't at the session so I don't know what exactly the question was, but it sounds like they were talking about something else; there has been some talk about incorporating the IEEE 802.1Qav, 802.1AS, and 802.1Qat standards that were developed for AVB into protocols like Dante and AES67, but that would involve hardware changes for many Dante and AES67 devices - “No simple switch in software” sounds like a reference to that, and not a reference to conversion between AVB and other protocols. Like I said, I wasn't at that session so I don't know exactly what the question was or how they answered it.

Quote:
Originally Posted by bhandy View Post
To date, there is only one company creating a "bridge" for AVB to AES67.
False. There are several devices from several manufacturers with this capability.

Quote:
Originally Posted by bhandy View Post
The problem will be that this bridge will involve "proprietary" (as in non-standard) technology
That isn’t really a problem. Dante itself is a proprietary protocol, and any manufactured product will inevitably involve proprietary technology.

Quote:
Originally Posted by bhandy View Post
because AVB originates from Firewire, etc, etc.
Uhhhh….. what?

Quote:
Originally Posted by bhandy View Post
Manufacturers using Auadinate's Brooklyn II chipset have a simple option in software to change output from Dante, QLan, Ravenna, etc to AES67. It will be that easy.
I know what you’re trying to say, but this is worded in a very confusing way. The Brooklyn chips will not natively output Q-LAN or Revenna streams, they only do Dante and AES67. If you're trying to connect a Dante device to a Q-LAN device, you configure the Dante device to output an AES67 stream, and the Q-LAN device will be able to connect to the AES67 stream. That's not something that the manufacturer does, that's something that the installer or end-user does.
Old 5 days ago
  #20
Gear Nut
 

Quote:
Originally Posted by albinotuba View Post
Oh cool. I was there too! I didn’t go to that session, though.



I’m not sure what this means. I wasn't at the session so I don't know what exactly the question was, but it sounds like they were talking about something else; there has been some talk about incorporating the IEEE 802.1Qav, 802.1AS, and 802.1Qat standards that were developed for AVB into protocols like Dante and AES67, but that would involve hardware changes for many Dante and AES67 devices - “No simple switch in software” sounds like a reference to that, and not a reference to conversion between AVB and other protocols. Like I said, I wasn't at that session so I don't know exactly what the question was or how they answered it.



False. There are several devices from several manufacturers with this capability.



That isn’t really a problem. Dante itself is a proprietary protocol, and any manufactured product will inevitably involve proprietary technology.



Uhhhh….. what?



I know what you’re trying to say, but this is worded in a very confusing way. The Brooklyn chips will not natively output Q-LAN or Revenna streams, they only do Dante and AES67. If you're trying to connect a Dante device to a Q-LAN device, you configure the Dante device to output an AES67 stream, and the Q-LAN device will be able to connect to the AES67 stream. That's not something that the manufacturer does, that's something that the installer or end-user does.
"What is the difference between AES67 and AVB?
The AVB standards define improvements to the standard Ethernet layer 2 protocol to provide synchronization, Quality of Service (QoS) and an admission control behaviour that prevents the network from becoming overloaded by media traffic; an AVB network requires use of specific AVB Ethernet switches. AES67 utilizes protocols on layer 3 or above, so is able to function using standard Ethernet switches. While it is entirely possible for AES67 to function over an AVB network, AES67 will not receive any performance benefits from the capabilities of the AVB network, except if the AES67 device fully supports the AVB layer 2 protocol stack.

How does AES67 compare with AVB?
AES67 is a standard for high-performance audio over IP interoperability, fully building on IP / Layer 3 protocols and above. AVB is a set of protocols expanding Ethernet / Layer 2 capabilities to enable the time-sensitive transport of media data including video and audio. It requires all participating devices (senders, talkers and switches) to support these protocols (of which some require dedicated hardware support not available in most “legacy” gear) and is limited to operate on a single LAN (so it is not routable). AES67 can run on existing manageable network infrastructure and can also work across subnets (i.e. AES67 is routable)."

The two presenters said there is only one manufacturer they were aware of with an AVB + AES67 chipset. There may be other products, but internally, they are separate processors/processes...according to them. Remember, there is AVB certified gear and AVB compliant gear. Those two are not the same thing. There are no virtual sound cards for AES67 alone, but there are four, maybe five virtual sound cards supporting Ravenna and AES67.

Sorry I was not clear. The simple switch to output to AES67 instead of each manufacturer's "proprietary" protocol (Dante, Ravenna, QLan, Livewire, etc) is in software. Yes, it is selected by the integrator. Yes, it is possible to transmit the native protocol of the respective manufacturer and AES67. It will depend upon the manufacturer and network.

The reference to AVB having origins in Firewire, it has to do with the transport protocol taking Firewire/IEEE 1394 frames and putting them into Ethernet frames.
Old 5 days ago
  #21
Gear Nut
 

Thread Starter
Quote:
Originally Posted by albinotuba View Post
I know what you’re trying to say, but this is worded in a very confusing way. The Brooklyn chips will not natively output Q-LAN or Revenna streams, they only do Dante and AES67. If you're trying to connect a Dante device to a Q-LAN device, you configure the Dante device to output an AES67 stream, and the Q-LAN device will be able to connect to the AES67 stream. That's not something that the manufacturer does, that's something that the installer or end-user does.
Right, and because Dante AES67 interop presently only supports multicast streams, you can't set the Dante latency below 1ms. Also, only 48k is supported.
Old 1 day ago
  #22
Gear Nut
 

Thread Starter
In other news, I did a bit of AVB interoperability testing between a MOTU Ultralite AVB, Avid S3, and OS X running Sierra. (Switch was an Extreme Networks x460.) Suffice to say – it's possible to get them all talking, but thanks to a bunch of quirks of AVB (such as different stream channel widths per device), it ain't easy. More here.
Old 1 day ago
  #23
Gear Nut
 

Thread Starter
Quote:
Originally Posted by bhandy View Post
The reference to AVB having origins in Firewire, it has to do with the transport protocol taking Firewire/IEEE 1394 frames and putting them into Ethernet frames.
I too was surprised by this, but it's true The AVB data format is based on IEC 61883, although it looks like these was/will be dropped in IEEE P1722a (support for video).
Loading mentioned products ...
Topic:
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get instant access to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump