AUDIO over IP - REDNET 3 & 16 Review. AES67 Sets A New Standard for Computer Audio

[occamsrazor]

The main disadvantages I see with the Burl for home use are:
 
- Lack of auto-sensing/auto-changing input rates, it's manual by dial.
 
- For those who like a DAC/Pre combined unit, the Burl doesn't have volume control.

 

[InsanityOne]

  Well you are right the BURL DAC plus a DANTE card is around $2700 - the RN3+Mutec is around $2200 - that is without the DAC.  Then factor in a decent SPDIF or AES cable so maybe you're between $2400-$2600 (unless you want to connect these multi $K boxes with a $50 cable - whatever works for you.  My experience is that the SPDIF/AES cable does make a significant difference).  Add in another decent BNC/AES cable between the RN3 and the Mutec...it all adds up.  Upgraded power cables - you need three for the RN/Mutec/DAC - versus one for the BURL,etc...
 
If the RN D16 is chosen then you are well past the BURL cost before adding in the DAC...

 
Indeed, you are right. The price of having a RedNet chain versus just the Burl is significantly higher. But just to be clear, the $999 Focusrite RedNet PCIe network card is recommend / required now? Or for users like me who have a high-end motherboard with multiple NICs, can we just stick to using the virtual soundcard software? I have not heard any definitive comments on if the Focusrite PCIe card adds a significant performance increase over an quality Intel NIC or not.
 
Also, another thing to take into account that I didn't think of earlier is whether or not the Burl benefits from an external WCLK (like the RN3 / D16 and the Mutec) which would also significantly increase the price, unless of course the internal clock on the Burl does not benefit from an external clock.
 
Edit: Another thing I noticed is that when looking at the Burl B2 DAC on Sweetwater, the photos must be of an older model because none of the inputs are the same and there is definitely not an Ethernet / Dante port on the back. If you bought yours from Sweetwater was it the newer model?
 
- InsanityOne 

 

 

[jazzfan]

 
Update on RedNet > Yggdrasil direct via spdif failure by way of information summary referring to my post here: http://www.head-fi.org/t/806827/audio-over-ip-rednet-3-16-review-aes67-sets-a-new-standard-for-computer-audio/555#post_12668123 and this reply: http://www.head-fi.org/t/806827/audio-over-ip-rednet-3-16-review-aes67-sets-a-new-standard-for-computer-audio/555#post_12668224 ...
 
Known user instances of RedNet " DAC via spdif(coax/Phono=RCA) *OK*

User	RedNet Source	DAC
Iving	D16 AES	Cambridge DacMagic Plus
Soundsgoodtome/gefski	REDNET 3	Matrix Mini-i
Soundsgoodtome/gefski	REDNET 3	R2R

 
Known user instances of RedNet " DAC via spdif(coax/Phono=RCA) *null*

User	RedNet Source	DAC
Iving	D16 AES	Yggdrasil
Iving	D16 AES	Linn AV 5103
Soundsgoodtome/gefski	REDNET 3	Yggdrasil

 
Further instance of *Yggdrasil null* (but not GuMBy) via spdif where Source is D16 AES:
Matrix of D16 AES(/Gustard U12) " GuMBy/Yggdrasil [credit: "atomicbob"]
Where the Reclocker and next column are marked "none" it means the Source Transport connected directly to the DAC input. The final column, simply titled "works" indicates whether you will hear music or not.

GuMBy reads spdif from D16 AES *but Yggdrasil does not* confirmed by gldgate here: http://www.head-fi.org/t/806827/audio-over-ip-rednet-3-16-review-aes67-sets-a-new-standard-for-computer-audio/645#post_12671773.
 
Re possible explantions:
 
Yggdrasil's successful reception of Pro and Consumer spdif [credit: "atomicbob"]
Here is another, highly regarded pro-audio interface *that works fine transporting SPDIF to the Yggdrasil* ... The SPDIF data stream has two sub-modes, one is for Professional audio use of the control word and the other is Consumer which has the copy restriction bit which prohibits copying of the SPDIF data stream.

Yggdrasil receives Pro *and* Consumer spdif OK albeit from a *non-RedNet source*. But in my less technical testing, a Cambridge DacMagic Plus works fine with the D16 AES whereas Yggdrasil does not - meaning that whatever the reading capabilities of the Yggdrasil, the Cambridge DacMagic Plus does it better.
 
Reportedly Yggdrasil has a history of difficulties around spdif BNC input: http://www.computeraudiophile.com/f6-dac-digital-analog-conversion/schiit-yggdrasil-best-digital-analogue-converter-available-24351/index18.html; see also http://www.head-fi.org/t/806827/audio-over-ip-rednet-3-16-review-aes67-sets-a-new-standard-for-computer-audio/630#post_12671711.
 
We await the results of Focusrite UK's further bench tests on my D16 AES / Yggdrasil.

 
I just realized my Yggy is a victim of this issue as well. I started a product page for the RedNet 3 here, and I added a Guide section to document these issues.
 
If anyone is aware of a fix from Focusrite, please fell free to update the Troubleshooting/Known Issues section in the new RedNet 3 Guide.
 
Thanks!

 

[rb2013]

The main disadvantages I see with the Burl for home use are:

- Lack of auto-sensing/auto-changing input rates, it's manual by dial.

- For those who like a DAC/Pre combined unit, the Burl doesn't have volume control.

True on both counts.

 

[rb2013]

Indeed, you are right. The price of having a RedNet chain versus just the Burl is significantly higher. But just to be clear, the $999 Focusrite RedNet PCIe network card is recommend / required now? Or for users like me who have a high-end motherboard with multiple NICs, can we just stick to using the virtual soundcard software? I have not heard any definitive comments on if the Focusrite PCIe card adds a significant performance increase over an quality Intel NIC or not.

Also, another thing to take into account that I didn't think of earlier is whether or not the Burl benefits from an external WCLK (like the RN3 / D16 and the Mutec) which would also significantly increase the price, unless of course the internal clock on the Burl does not benefit from an external clock.

Edit: Another thing I noticed is that when looking at the Burl B2 DAC on Sweetwater, the photos must be of an older model because none of the inputs are the same and there is definitely not an Ethernet / Dante port on the back. If you bought yours from Sweetwater was it the newer model?

- InsanityOne   

Good questions. I doubt a $1000 Rednet PCIe card is worth the money, certainly a slightly lower latency over DVS. I would recommend a newer and powerful PC for running DVS. I use a Haswell iCore 7/WIN10 with lots of memory.

On the external clock issue, the costs I mentioned above did not factor in an external OCXO clock for the Rednet chain. That would be an add-on for either path. The Mutec I referred to was for AES/Spdif reclocking, not needed in the BURL. The BURL bypasses the need for connecting by AES or Spdif coax, as the Dante Brooklyn II card is directly connected on the board.

I bought the Dante version of the BURL, it is sightly more expensive. The nonDante version is not convertible as far as I'm aware.

I hope one day every DAC has a mini PCIe Dante slot on it's board.

 

[mourip]

   
Edit: Another thing I noticed is that when looking at the Burl B2 DAC on Sweetwater, the photos must be of an older model because none of the inputs are the same and there is definitely not an Ethernet / Dante port on the back. If you bought yours from Sweetwater was it the newer model?

 
I contacted Sweetwater as I was considering trying the Burl in my headphone rig. They do not have the Dante version yet but Aric from Sweetwater emailed me to say that he has pricing now. I missed him today but hope to have more information tomorrow. The newer version has an obvious ethernet port in the rear panel.
 
I love the D16 in my speaker setup but I tallied up the dollar damage yesterday and realized that I had a bundle invested inbetween my PC and my DAC: D16 + Mutec USB + Antelope LiveClock + cables. This is the reason that I am so curious to find out where the stock Burl fits in "The Big List" and how much is needed to modify it to surpass what I have.
 
The other issue is that I really like my Yggy which I currently use in my HP rig with just a D16 and no other add-ons so the Burl would need to beat it. Too many unknowns for me to take the plunge yet.

 

[InsanityOne]

Good questions. I doubt a $1000 Rednet PCIe card is worth the money, certainly a slightly lower latency over DVS. I would recommend a newer and powerful PC for running DVS. I use a Haswell iCore 7/WIN10 with lots of memory.

On the external clock issue, the costs I mentioned above did not factor in an external OCXO clock for the Rednet chain. That would be an add-on for either path. The Mutec I referred to was for AES/Spdif reclocking, not needed in the BURL. The BURL bypasses the need for connecting by AES or Spdif coax, as the Dante Brooklyn II card is directly connected on the board.

I bought the Dante version of the BURL, it is sightly more expensive. The nonDante version is not convertible as far as I'm aware.

I hope one day every DAC has a mini PCIe Dante slot on it's board.

Some very good information here. I will be patiently awaiting your thread on the Dante version of the Burl B2 Bomber DAC and its modifications. I truly believe that it could be the "end-game" DAC until another technology comes along that is even better than Dante, (If that ever happens.) I just find the idea of a simplified audio chain so appealing. Just running three devices, a PC, a DAC, and an amp, but still getting the very best audio reproduction possible.
 
 

   
I contacted Sweetwater as I was considering trying the Burl in my headphone rig. They do not have the Dante version yet but Aric from Sweetwater emailed me to say that he has pricing now. I missed him today but hope to have more information tomorrow. The newer version has an obvious ethernet port in the rear panel.
 
I love the D16 in my speaker setup but I tallied up the dollar damage yesterday and realized that I had a bundle invested inbetween my PC and my DAC: D16 + Mutec USB + Antelope LiveClock + cables. This is the reason that I am so curious to find out where the stock Burl fits in "The Big List" and how much is needed to modify it to surpass what I have.
 
The other issue is that I really like my Yggy which I currently use in my HP rig with just a D16 and no other add-ons so the Burl would need to beat it. Too many unknowns for me to take the plunge yet.

 
Thanks for posting this, be sure to let every one know when you find out about the pricing / availability of the Dante equipped B2 Bomber DAC from Sweetwater. I am very curious to find out where the modified Burl B2 fits into "the big list" too. I am hoping it sits right at the top! As for the Burl B2 vs. the Yggy, the overall sound may come down to personal preference, but for me it is no contest because of Dante and the lack of a need for extra devices like Mutec, LiveClock, etc. with the Burl B2.
 
- InsanityOne 

 

[jabbr]

I just realized my Yggy is a victim of this issue as well. I started a product page for the RedNet 3 here, and I added a Guide section to document these issues.

If anyone is aware of a fix from Focusrite, please fell free to update the Troubleshooting/Known Issues section in the new RedNet 3 Guide.

Thanks!

It is not a Focusrite issue but an Yggy issue.
Schiit used off-spec SPDIF connectors (50 Ohm instead of 75 Ohm) for a batch of Yggy's. You should send your Yggy back to Schiit and have it fixed.

 

[jabbr]

Good questions. I doubt a $1000 Rednet PCIe card is worth the money, certainly a slightly lower latency over DVS.

...

Don't get hung up on this Dante latency as it is irrelevant imho, for how we use the Rednets.

This Dante latency is only important where multiple Dante chains are used in parallel and need to in-sync.
This Dante latency is the latency between the Dante source device and the Dante receiver device. When using multiple receivers you can imagine how different latencies in each of the chains can have an effect. Dante protocol makes sure all chains remain in sync by applying the longest latency to all the chains.

When using a single chain like we do, it is not too relevant how long a Dante latency is when kept in reasonable margins and it will have no effect on the sound result, except when it becomes too long for the buffer size/sample rate and dropouts will occur. This situation causing dropouts is then a network issue and not so much an issue of DVS vs PCIe-card. I can imagine that with bad network configuration and management you can create dropouts with a PCIe-card as well.

The main features of the PCIe-card is that it supports 128 channels and has 3usec latency, both of which are not particularly useful for us.

 

[johnjen]

What he said^^^^^
 
JJ

 

[johnjen]

Update on the Focusrite/Audinate DVS on Mac behavioral 'quirks'.
 
Since DVS is an Audinate program they are now 'looking into it' along with Focusrite and Jriver.
 
There are additional behavioral quirks I've been noticing in Media Center that may or may not be related to the Dante network, but I added this additional layer as well just in case they are related.
 
I have no idea how long it will take Audinate to come back with a response.
 
And I also raised the issue about the persistence of the Sample Rate Follow feature on Mac and how it differs from the pc.
They are sending this up the chain of command to see what can be done…
It would be nice if they were able to arrive at a 'set and forget' function instead of having to run RedNet Control all the time…
 
JJ

 

[Iving]

   
I just realized my Yggy is a victim of this issue as well.

 

It is not a Focusrite issue but an Yggy issue.
Schiit used off-spec SPDIF connectors (50 Ohm instead of 75 Ohm) for a batch of Yggy's. You should send your Yggy back to Schiit and have it fixed.

 
Just to point out that atomicbob's interesting findings (red cells in quoted post) re RedNet/Yggy failure (ostensibly because of obedience to a Copy Prohibited flag) via spdif *RCA/Coax* and BNC (not AES/XLR which is unaffected) got confounded with the eventual discovery that the spdif *RCA/Coax* input (not AES/XLR and don't know about BNC) of my new Yggy was faulty per se (and it got returned - I now use a Convert-2 like @mhamel). gldgate also experienced RedNet/Yggy failure via RCA presumably without the same fault; however, I'm not sure to what extent we established that Yggy always baulks at RedNet via RCA/BNC - since **Focusrite benchtested with a demo Yggy via RCA (not my faulty one) and asserted firmly that it worked**. (This being so it is unlikely that Focusrite will pursue the *putative* Yggy communication failure with any vigour at all.)

 

[Iving]

   
everything sounds wonderful with just setting it all to 192K even though RedBook 44.1K should probably be overclocked at a multiple such as 176K.  

 

  Ok. One more mystery that came with the upgrade. Previously I had 176K as a choice from the dropdown box in Rednet Control.
 
Now it is gone.
 
I know that the RN3 did not support 176K but my D16 did previously.
 
If you have a D16 and upgraded do you see 176K as a choice in Rednet Control?

 

if I set everything to 192K the sound is still wonderful.  

 

not seeing 176K in my RedNet Control app (and beginning to panic) but instead seeing it and being able to set it in the Dante Controller instead.  

 

   
You are right. I do see 176.4k listed in the Dante Controller app under Device view.

 
Yes - on the D16 AES (but not the RedNet 3 I understand) 176 is available in Dante Controller not RedNet Control. It is easy to (Upsample) x4 redbook in foobar (Sox) and make it work. My head told me it should be better than 192 (maths etc). But I changed back thinking 192 fuller and more polished. But I am not yet convinced and will experiment further. I can't at the moment as my system is in a second phase of suspension - last time speaker repair - this time PC replacement. (And I haven't updated firmware lately.) Will try to remember to post back whether 176 trumps 192 for redbook. Would like to hear other views. Interested in the mojo not the smoothness!

 

[Soundsgoodtome]

Interesting on the Burl B2 DAC, that they chose to use strictly an smps instead of linear. Is the idea of linear power strictly audiophile? While the guys recording our music are using stepped power units?

 

[occamsrazor]

I'd love to see an updated Burl, let's call it the B3 - half-width case, Brooklyn II Dante, auto-sensing inputs, and DAC/PRE volume control.
Yeah I know it'll never happen, but....