Article: A Conversation About Network Audio, AES67, Ravenna, and Merging Technologies

[The Computer Audiophile]

View full article

[jcbenten]

Will not play for me...Win10 Pro, Chrome....runs on Edge

[rikhav]

This could not come at the right time

Of course, without any prior knowledge that you will come up with this I was reading about dacs supporting AOIP. Of course, the name of merging dac comes right at the top 

But there is one more pro Audio company which called Burl audio which makes a dac called B2 bomber and has aoip as input but uses a different aoip implementation called DANTE

 

All of it was too difficult to understand without having prior knowledge about AOIP, so looking forward to hear the whole discussion of yours with the merging technologies

 

Thanks again to you and merging technologies

[rikhav]

Just finished listening to whole interview and am very happy that for a single setup it's as easy as using a USB dac

All the mis-conceptions about the audio ip protocol and how to implement it on the source PC are cleared

Now that we know all that maybe you can request merging to send you a review sample of a nadac so we know how it sounds 

Luckily we have a merging distributor in India who is selling a nadac but unfortunately it's way out of my league as of now 

[The Computer Audiophile]

Hi rikhav - I'm happy you got something out of the interview. That's why we did it. 

 

I plan on getting a Nadac for review. 

[The Computer Audiophile]

3 hours ago, jcbenten said:

Will not play for me...Win10 Pro, Chrome....runs on Edge

 

On my Win10 is works on Chrome, but not on Edge. 

 

I'll look at the SoundCloud site to see if others have reported similar behavior. 

[esldude]

Thanks for the interview.  A little early for audiophile consumers yet I think, though that could change pretty quickly.  Since hearing of Dante and Ravenna it seems like a dream for all sorts of audio uses in the home though oriented for pro studio use.  We just need more focus on products for audiophile use I think.

 

 

[gldgate]

I've been using Dante via RedNet for about a year. Consider me impressed. Initial setup was certainly not "plug n play". However, after that initial hurdle my system has been rock solid. Much more so than when I was using a USB set-up. Glad AOIP is moving to home audio. I think folks here will be pleasantly surprised.

[monteverdi]

Pretty amazing possibilities, Genelec is already implementing Ravenna in their speakers and I like the thought of having some extra speakers to add to my stereo setup.

Not much discussion about sound quality compared to USB, Spdif etc. Should be better because Ravenna allows for error correction and distributed clock.

[mourip]

I have also been using Dante for over a year and it easily surpassed my tricked out USB chain. A thread was started here many months ago but it has drawn little attention. Glad to see some interest.

 

Please note that you can implement AES67/AOIP/Dante right now using your DAC of choice. No need to buy a Burl or a or Ravenna if you already have a DAC that you are happy with. Focusrite makes two models of ethernet to SPDIF converter that work great. These are the Rednet 3 and the Rednet D16AES. You load Dante virtual sound card(software) on your audio server and then it sends audio over ethernet to the Rednet device which converts to SPDIF and on to your DAC.

[joelha]

mourip and gldgate are spot on with their comments.

 

The two current Focusrite options are the best digital interfaces I've heard and I think I've owned most of the best ones.

 

There's no way I could know for sure but AOIP seems like it will have major implications for our hobby.

 

Kudos to Chris for bringing this important discussion online.

 

Joel

[input username here]

Thank you Chis, this was, as always, a great service to the CA community.  I am pretty sure I get it, but I have a few questions, if you don't mind.

 

1. Mr. Brulhart mentioned that "many" other manufacturers had already implemented systems with similar capabilities to Ravenna.  I assume he was talking about other pro applications, like Dante, but doesn't, say, Sonos do exactly the same thing too?  Of course the BIG difference being that Sonos's protocol will only mesh Sonos products, whereas Ravenna will (or at least promises) to link devices from many different companies.  Are there any other advantages to Ravenna (I know that Sonos also currently does not run high sample rates)?

 

2. Similar to monteverdi's observation, I wish that sound quality would have been discussed more.  It seems like LAN-based devices should sound better, but do they in practice?  Would a (old school) CD transport sound better if it sent its data to the DAC over (Ravenna) LAN, rather than AES or S/PDIF or even I2S?

 

3. And finally,while the allure of multi-channel audio never caught on for me, multi-channel video remains interesting.  Would you be able to process, say, Dolby Atmos, DTS:X or Auro-3D on the Zman module (or on whatever server is sending data to the Zman) and remove the need to buy an expensive (and, soon-to-be obsolete) hardware surround processor?  That would be something cool....

 

Thanks again!

[esldude]

@gldgate and @mourip

As this is new, and interesting to others, and likely to increase in the future, it might be instructive to the rest of us for you two guys to either do a thread or blogs about how you have arranged your systems, software you use for playback etc. etc. 

 

For instance, I would be interested in using the Rednet 3 for multi-channel use.  It seems like I could have the virtual soundcard software, and use the optical outputs to 8 DACs (7.1 surround) or maybe use 4 stereo DACs to accomplish this.  Yet is this something like either of you have done?

 

Also my apologies if you have threads or blogs already.  I missed them so links to them might be good here.

 

Okay here is one I even posted in last year.  Personal issues let it slip from my radar screen so to speak.

[mourip]

I would be glad to wrote up something. As a start, the best source with lots of good questions and annecdotal information is this thread on Head-Fi. Hopefully we can now begin our own good discourse.

 

AOIP/Dante

 

Just like many threads here one has to sift through a bit of personal eccentricity and contentiousness but there is a goldmine of good information there. The OP was famous for a another long thread that explored advanced USB chains in detail, comparing dozens of devices. He then discovered AES67 implemented by Focusrite as Dante. Many of us were then converted through his enthusiasm and relentless experimentation. 

 

Like all cutting edge audiophile endeavors we found one could improve the sound of these Focusrite devices by adding other supporting devices like using a Mutec 3+ USB to re-clock or using a high quality external wordclock. The good news is that we also found that with Dante tweaks of the PC made far less difference.

 

I will update with more of my experiences and describe my chain in the thread I started here.

 

Rednet on CA

 

I am really glad to finally see interest by DAC manufacturers. It certainly makes more sense to integrate this capability into a DAC rather than have a bunch of external boxes. Having said that using the Rednet D16 with my Yggy DAC have really upped my system to a place I would have never thought was possible. 

 

I will use the original CA Rednet thread to add updated info since it is a different flavor of AES67 than Ravenna described here.

 

Thanks for the interest.

 

 

 

 

[dbrulhart]

8 hours ago, input username here said:

Thank you Chis, this was, as always, a great service to the CA community.  I am pretty sure I get it, but I have a few questions, if you don't mind.

 

1. Mr. Brulhart mentioned that "many" other manufacturers had already implemented systems with similar capabilities to Ravenna.  I assume he was talking about other pro applications, like Dante, but doesn't, say, Sonos do exactly the same thing too?  Of course the BIG difference being that Sonos's protocol will only mesh Sonos products, whereas Ravenna will (or at least promises) to link devices from many different companies.  Are there any other advantages to Ravenna (I know that Sonos also currently does not run high sample rates)?

 

2. Similar to monteverdi's observation, I wish that sound quality would have been discussed more.  It seems like LAN-based devices should sound better, but do they in practice?  Would a (old school) CD transport sound better if it sent its data to the DAC over (Ravenna) LAN, rather than AES or S/PDIF or even I2S?

 

3. And finally,while the allure of multi-channel audio never caught on for me, multi-channel video remains interesting.  Would you be able to process, say, Dolby Atmos, DTS:X or Auro-3D on the Zman module (or on whatever server is sending data to the Zman) and remove the need to buy an expensive (and, soon-to-be obsolete) hardware surround processor?  That would be something cool....

 

Thanks again!

 

Dear "input username here",

 

1) On the pro side there are dozens of manufacturers who implemented Ravenna in both hardware and software. But I was talking about the consumer/audiophile side here. We are currently deploying our Linux Ravenna driver on various manufacturers products. We have shown working Aurender, Melco and Roon-ROCK systems in Munich in May, but there are a good number of others currently integrating our driver. Technically, we can add to the picture all applications running on Windows and MacOS as their manufacturers benefit from the Ravenna support through our Windows and MacOS drivers as well, like Roon, JRiver, Audirvana, etc, etc. This is for the "software" side. On the hardware side, the only consumer products supporting Ravenna today are NADAC and NADAC PLAYER, but the soon availability of the ZMan board will add many to the game. There are here also a good number of companies currently working on integrating ZMan in their DACs, speakers or other devices.

 

2) Stricto-senso, the "quality" of the bits transfered by Ethernet, USB, Firewire, Thunderbolt or even AES or SPDIF is the exact same. Every bits reach the destination intact and in the proper sequence. If it weren't the case we wouldn't discuss about sound quality but about glitches, pops, scratches or bangs. So, on the transport aspect, all those technologies are equal (if we put aside latency, number of channels, resolution, which are quantity considerations and not quality considerations). However, the sending and receiving devices can be more or less accomodating to the technology involved by these transmissions, some are synchronous, some are asynchronous, some have more or less jitter, some are more of less precise, etc... and it is then down to the sending and receiving devices to be properly designed to accomodate with these constraints. In theory all those transmission mediums can sound equivalent if the hardware implementing them is properly designed. So in that sense Ethernet can have some advantages, like a better galvanic isolation, specially if using optical cables. Ravenna is a very precise, low jitter and low latency protocol, so meaning that it can have, in theory, advantages in the design of hardware communicating with it. But it's main advantage is definitely its flexibility. Quality is really down to the engineers not to the technology.

 

3) Well, there are many aspects in  surround processors: the format decoding, but generally also processing (as its name suggest), like channel mapping, down-mixing, up-mixing, room correction and more creative effects (like stadium effect, action movie effect, etc...), and this can be quite costly in term of DSP. We plan offering some processing capabilities in ZMan, like room correction (to some level), EQ, up/down mixing, up/down sampling, etc... We are also investigating format decodings, however today any (decent) Bluray player actually properly decodes Atmos, DTS:X and Auro, so no real need to be done externally. So, on a purist approach of sound quality, if we were to use ZMan in this context, I would be more tempted to let the Bluray player decode the format itself, give up on special effects, and keep the DSP for either some room correction or some upsampling. Typically an HDMI audio to Ravenna converter would be the perfect link between Bluray players or video servers to Ravenna enabled DACs. As mentionned in the interview, Ravenna would allow as standard to start with a stereo DAC, then add more DACs later for the surround. Or even allow for hybrid solutions, with a very high-end "traditional" stereo system for the front, and active Ravenna enabled speakers for the rear and center. As Chris said: the limit is the imagination.

 

Best regards,

 

Dominique

 

[Kal Rubinson]

47 minutes ago, dbrulhart said:

Typically an HDMI audio to Ravenna converter would be the perfect link between Bluray players or video servers to Ravenna enabled DACs. As mentionned in the interview, Ravenna would allow as standard to start with a stereo DAC, then add more DACs later for the surround. Or even allow for hybrid solutions, with a very high-end "traditional" stereo system for the front, and active Ravenna enabled speakers for the rear and center. As Chris said: the limit is the imagination.

That's what I have been thinking.  

[bmoura]

7 hours ago, Kal Rubinson said:

That's what I have been thinking.  

 

7 hours ago, Kal Rubinson said:

Typically an HDMI audio to Ravenna converter would be the perfect link between Bluray players or video servers to Ravenna enabled DACs. As mentionned in the interview, Ravenna would allow as standard to start with a stereo DAC, then add more DACs later for the surround.

 

You could do that.  I prefer a purpose-built Multichannel DAC, like the NADAC, NADAC Player or exaSound e38, where you have all 6 to 8 channels in one box and don't have to deal with issues of drivers, clocks, sync, etc. 

 

[input username here]

8 hours ago, dbrulhart said:

We are also investigating format decodings, however today any (decent) Bluray player actually properly decodes Atmos, DTS:X and Auro, so no real need to be done externally. So, on a purist approach of sound quality, if we were to use ZMan in this context, I would be more tempted to let the Bluray player decode the format itself, give up on special effects, and keep the DSP for either some room correction or some upsampling.

 

Firstly, thank you for the excellent and informative responses to my questions--and thanks too for participating in the CA community.  That said, some (many/most?) computer audio/video-philes don't necessarily consume movies off spinning disks; some (many/most?) like to bypass the BR player completely and play from a file-server.  I, for one, have not so much as touched a non-CD 5" disk in several years--nor do I have any desire to ever do so again.  Perhaps I'm a member of too small a community for products to be designed for our particular needs, but being able to stream, say, Atmos movies straight from my server to a set of LAN-enabled MC speakers would be... the dream.

 

19 minutes ago, bmoura said:

You could do that.  I prefer a purpose-built Multichannel DAC, like the NADAC, NADAC Player or exaSound e38, where you have all 6 to 8 channels in one box and don't have to deal with issues of drivers, clocks, sync, etc. 

 

Just to save on clutter (and $$$ cables), I too would prefer a single box DAC (or, even better, no DAC at all and covert D2A right in the speakers).  But, if I am not misunderstanding things, the whole point to using a protocol like Ravenna would be so you don't have to ever mess around with (non-Ravenna) drivers, clocks, sync, etc.--that is, Ravenna obviates all of these problems by design.  (I hope I am correct in stating this and not over-stepping....)

[Kal Rubinson]

1 hour ago, bmoura said:

You could do that.  I prefer a purpose-built Multichannel DAC, like the NADAC, NADAC Player or exaSound e38, where you have all 6 to 8 channels in one box and don't have to deal with issues of drivers, clocks, sync, etc. 

I agree with you but this would open up new possible configurations for high-quality multichannel including the direct use of Ravenna-enabled speakers.

[bmoura]

1 hour ago, input username here said:

Just to save on clutter (and $$$ cables), I too would prefer a single box DAC (or, even better, no DAC at all and covert D2A right in the speakers).  

 

 

Hmm, then you'd need to replace your speakers and DAC when upgrading DACs?  Think I'll stay with keeping the Multichannel DAC and Loudspeakers separate.

 

[Kal Rubinson]

4 minutes ago, bmoura said:

Hmm, then you'd need to replace your speakers and DAC when upgrading DACs?  Think I'll stay with keeping the Multichannel DAC and Loudspeakers separate.

It also means that you might get a newer, better DAC when upgrading speakers.  

[bmoura]

Just now, Kal Rubinson said:

It also means that you might get a newer, better DAC when upgrading speakers.  

 

Assuming you want to pay for both speakers and a DAC. Rather than one or the other at a somewhat lower cost. 

 

[Kal Rubinson]

2 hours ago, bmoura said:

 

Assuming you want to pay for both speakers and a DAC. Rather than one or the other at a somewhat lower cost. 

 

Hey, Brian, I am not advocating this (note the smiley) but I do know that it will appeal to some and, frankly, promoting multichannel to them is a good thing.

[mourip]

12 hours ago, dbrulhart said:

 

Dear "input username here",

 

1) On the pro side there are dozens of manufacturers who implemented Ravenna in both hardware and software. But I was talking about the consumer/audiophile side here. We are currently deploying our Linux Ravenna driver on various manufacturers products. We have shown working Aurender, Melco and Roon-ROCK systems in Munich in May, but there are a good number of others currently integrating our driver. Technically, we can add to the picture all applications running on Windows and MacOS as their manufacturers benefit from the Ravenna support through our Windows and MacOS drivers as well, like Roon, JRiver, Audirvana, etc, etc. This is for the "software" side. On the hardware side, the only consumer products supporting Ravenna today are NADAC and NADAC PLAYER, but the soon availability of the ZMan board will add many to the game. There are here also a good number of companies currently working on integrating ZMan in their DACs, speakers or other devices.

 

2) Stricto-senso, the "quality" of the bits transfered by Ethernet, USB, Firewire, Thunderbolt or even AES or SPDIF is the exact same. Every bits reach the destination intact and in the proper sequence. If it weren't the case we wouldn't discuss about sound quality but about glitches, pops, scratches or bangs. So, on the transport aspect, all those technologies are equal (if we put aside latency, number of channels, resolution, which are quantity considerations and not quality considerations). However, the sending and receiving devices can be more or less accomodating to the technology involved by these transmissions, some are synchronous, some are asynchronous, some have more or less jitter, some are more of less precise, etc... and it is then down to the sending and receiving devices to be properly designed to accomodate with these constraints. In theory all those transmission mediums can sound equivalent if the hardware implementing them is properly designed. So in that sense Ethernet can have some advantages, like a better galvanic isolation, specially if using optical cables. Ravenna is a very precise, low jitter and low latency protocol, so meaning that it can have, in theory, advantages in the design of hardware communicating with it. But it's main advantage is definitely its flexibility. Quality is really down to the engineers not to the technology.

 

3) Well, there are many aspects in  surround processors: the format decoding, but generally also processing (as its name suggest), like channel mapping, down-mixing, up-mixing, room correction and more creative effects (like stadium effect, action movie effect, etc...), and this can be quite costly in term of DSP. We plan offering some processing capabilities in ZMan, like room correction (to some level), EQ, up/down mixing, up/down sampling, etc... We are also investigating format decodings, however today any (decent) Bluray player actually properly decodes Atmos, DTS:X and Auro, so no real need to be done externally. So, on a purist approach of sound quality, if we were to use ZMan in this context, I would be more tempted to let the Bluray player decode the format itself, give up on special effects, and keep the DSP for either some room correction or some upsampling. Typically an HDMI audio to Ravenna converter would be the perfect link between Bluray players or video servers to Ravenna enabled DACs. As mentionned in the interview, Ravenna would allow as standard to start with a stereo DAC, then add more DACs later for the surround. Or even allow for hybrid solutions, with a very high-end "traditional" stereo system for the front, and active Ravenna enabled speakers for the rear and center. As Chris said: the limit is the imagination.

 

Best regards,

 

Dominique

 

 

Thanks for joining us.

 

[bmoura]

38 minutes ago, Kal Rubinson said:

Hey, Brian, I am not advocating this (note the smiley) but I do know that it will appeal to some and, frankly, promoting multichannel to them is a good thing.

 

Promoting multichannel is a good thing.  

 

My experience is that speakers with built-in amps (active speakers) cost more than those without (passive speakers).  I'm guessing that if we see speakers with built-in amps or DACs - or both - they would be even more expensive.

 

But that may appeal to some, especially if price is not a key consideration.