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

[Cornan]

I am sorry if I upset you or anyone else on this thread. Not my intension! I have followed this thread a long time but still have'nt read about anyone using what I call a great USB chain with 2-wire unshielded USB cables (ie. 3-wire cables with GND lift). It takes a long time to perfect USB audio and I guess that applies to AOyIP as well since so many "tweaks" are posted here. There have been a lot of break though In the USB audio. Latest one is the sc. leakage loop which quickly described require battery supplies or Ultracap supply for devices together with isolated ICs, isolated ethernet, limited AC connected devices, Isolation transformer with floating secondary and no filtered powerstrip to completely remove. Sounds amazing! I guess the Isolation transformer w/floating secondary would make great improvements upon AOIP as well.

Anyway, I am all for AOIP...but just as a replacement for Ethernet IMO.

Listen to it, then decide. Techno babble means not much, ears mean everything IMO. I had a decent chain with LPS on everything inc additional mains regenerator to supply my whole system. Rednet still cleaner, fluid, better.

IMO it isn't just the power / ground lift issue, USB data transfer has other (many) issues.

I agree that techno babble will not solve anything! IMO LPS is the wrong path since it will ensure leakage currents. You'll need to limit the leakage currents (which travels through neutral & live). Power the AOIP via battery supply or Ultracapacitors supply and you will enjoy further SQ improvements due to isolation to AC mains. Adding a IT w/ floating secondary downhill will ensure no backward noise into the AC mains. It will better both USB and AOIP!

 

[jabbr]

Is anyone using external clock in an AOIP chain? If so, what solution?

I use a Grimm CC1 clock which can follow sample rate of the input signal, though I use it at a fixed sample rate.

 

[rb2013]

I agree that techno babble will not solve anything! IMO LPS is the wrong path since it will ensure leakage currents. You'll need to limit the leakage currents (which travels through neutral & live). Power the AOIP via battery supply or Ultracapacitors supply and you will enjoy further SQ improvements due to isolation to AC mains. Adding a IT w/ floating secondary downhill will ensure no backward noise into the AC mains. It will better both USB and AOIP!

'Leakage Currents' Ha!  Isn't that 'techno' babble!
 
Just kidding - well the leak of a LPS is orders of magnitude less then these SMPS's like the one in the RN stuff.  And much depends on the design of the LPS - R-cores are the best and have the highest PSRR - see my three part PS threads.
 
Ultracapacitors are very good - but limited in current - and very expensive.  And the Uptone LPS-1 for example uses a SMPS to charge - so back to square one.  Of course  you can use a LPS to charge it as well - but that only adds to the $400 cost.
 
My long experience with battery power of all kind is they seem to lack dynamics - and many have high DC noise.  Battery power supplies suffer from slow rise times and the ability to deliver high current on demand.
 
Now another way to address the small leakage of a LPS is to use a AC line isolator and filter - which I do - a separate one for the DAC, DDC chain, PC server.  The Art Audio provide 40dB of attenuation and isolation - right up to 1Mhz levels.
 
The reason LPS's can create AC leakage is the charging of their PS filter caps - that is usually large on initial power up and diminishes with warmup stability.  The reason low ESR caps are important.  The degree of further leakage will be determined by the current draw - for min draw situations that would be quite small.
 
In any event with 40dB of isloation - then another 40dB of filtering on you have 80dB of AC line noise reduction.  To near immeasurable levels.  Now if your gear uses R-Core power supply transformers - they have very high PSRR in and of themselves.  Take the Rednet Ferex SMPS with 150,000uv of AC line noise injection - that is a problem.  The MEIYEN is approx 300uv after turn on.  On the DC side 13uv at 24VDC, 8uv at 5VDC.
 
Let's see you power a Rednet with a LPS-1 - good luck friend.
 
I have much more info here for anybody curious:
 
http://www.head-fi.org/t/821621/audio-power-supplies-part-1-smps-lps-supercap-battery-diy-route-new-devices-opens-up-new-options
 
http://www.head-fi.org/t/821731/audio-power-supplies-part-2-smps-lps-supercap-battery-diy-route-new-devices-opens-up-new-options
 
http://www.head-fi.org/t/822160/audio-power-supplies-part3-smps-lps-supercap-battery-diy-route-new-devices-opens-up-new-options
 
The MEIYEN is a discrete designed based on the single version of the AMB σ11 regulation scheme.
http://www.amb.org/audio/sigma11/

 

[Cornan]

A general rule is to limit the amounts of DC powered devices in the chain. Any "tweak" that is improving something needs to be passive...and if absolutely necessary it needs to be powered isolated from the AC mains. Keep that in mind for AOIP as well!

 

[rb2013]

  Yes debating USB v AOIP, and I said I would cast my opinion when I heard it. You haven't heard the Audio Note DAC 5. Rather comparing 'similar' design topologies and saying it will sound the same. I haven't heard the Zanden model you quote, but have friends who have Zanden and they say it does sound different to most Audio Note units. I don't cast my vote on that until I hear one.
 
Your DAC v AN DAC 5. Hmm, that is difficult to quantify. I can't find any reliable comparisons between the Zanden and my Audio Note DAC 5 Special. And your DAC obviously. The Audio Note has very expensive output transformers and a lot of work gone into the analogue filtering (no digital filter). That and Peter's I/V secrets, quality power supply with 2 x oversized mains transformers, big backgate caps, SHUNT power supply for the digital board. I am not saying your DAC project is no good. I am saying unless they are pitched together it is impossible to quantify.
 
I have owned 4 AN DACs of various levels and some had mods applied.
 
My DAC retails at over £25K, I bought used for a lot less. I also upgraded the output transformers to Signature level.
 
My own comparisons with the AN DAC 5 were C1 CH Precision, MSB Platinum Stack, Meridian 808, Naim CD555, Esoteric K-01, Lampizator Big 6, AMR DP-777. It beat all of those squarely, more real sounding with zero fatigue. The Lampi was the best IMO of those comparison DACs, the AMR a close third.

Well not to turn this into a DAC thread - the APL NWO uses the finest Lunduhl output transformers for coupling.  The history of Zanden and Audio Note are intertwined.  The Zanden using the TDA1541A and a proprietary analog filter.  http://www.stereophile.com/content/audio-note-uk-vs-kondo-audio-note-japan-1
 
I have compared the APL NWO GO 3.0 to the DCS stack and the Rossini - and much prefered the APL.  The DCS stuff is way to dry for my taste. 
Srajan of 6Moons has a bit more experience then either of us:

With today's digital source components, a clearly best CD player would have to beat AMR, Audio Aero, Burmester, dCS, EMM Labs, Esoteric, Orpheus Labs, Spectral and Zanden.
 
Now enter a generous benefactor. He gifted me with this NWO 3.0-GO to "have the best" - a true boon for any reviewer and most assuredly this one.* Like myself, my friend has owned Zanden's top quartet. Unlike myself, he's also owned the dCS triple stack, Meitner gear and continues to spin vinyl.
 
In seven years of reviewing, it's a first that I should even acknowledge such a possibility. Very little separates most expensive digital these days. Proclamations of best would have to be contingent on a substantial lead. If the NWO 3.0-GO was substantially better than my Zanden, AMR or Ancient Audio machines? At my friend's, I already had opportunities for telling extrapolations. Going into a formal review, I was quite confident that I had to merely ascertain the width of the performance delta over my personal stuff; assess whether the APL Hifi/Esoteric was just a bit or a lot better. If a lot -- the kind of lot I've not heard before between top-level digital -- I'd be a believer in best.
 
The wrap
Had I expected the big light show in the sky upon listening to the NWO after clocking 300 non-stop hours of endless repeat, I'da been disappointed. Both my Ancient Audio deck and Abbingdon Music Research's AM-77 -- two tubed machines as well -- perform at such a high plateau already as to simply not leave lots of room to climb much higher. The NWO does climb higher but what it gains is anything but showy, loud and fireworky. Quantitatively, with scale in hand to weigh treble, midrange, bass, dynamics et al, you'd have noticed subtle shifts at best. The NWO's mo betta operates qualitatively. It removes treble artifice that remains with even top-shelf machines. It has you respond differently - with less tension, long-term fatigue and subliminal annoyance. Most definitely, you'll practice no more avoidance of those beloved but overcooked hot-hot-hot albums which won'tsound any better now but quit annoying you. Their digital stinger has been removed. Poof. You can finally enjoy those discs for their musical merit and no longer get prevented by inferior mastering.

 
Now for me my reference is my old $30K analog rig - VPI Superscout Master Sig, Dynavecter XV1S, Benz Ebony LP, Bent Audio Silver step-up transformers, CJ tube 6922 tube phono pre, all Nordost Valhalla wiring - feeding a CJ ACT2 Pre.

 

This new source has left this analog in the dust.

 

[Cornan]

'Leakage Currents' Ha!  Isn't that 'techno' babble!

Just kidding - well the leak of a LPS is orders of magnitude less then these SMPS's like the one in the RN stuff.  And much depends on the design of the LPS - R-cores are the best and have the highest PSRR - see my three part PS threads.

Ultracapacitors are very good - but limited in current - and very expensive.  And the Uptone LPS-1 for example uses a SMPS to charge - so back to square one.  Of course  you can use a LPS to charge it as well - but that only adds to the $400 cost.

My long experience with battery power of all kind is they seem to lack dynamics - and many have high DC noise.  Battery power supplies suffer from slow rise times and the ability to deliver high current on demand.

Now another way to address the small leakage of a LPS is to use a AC line isolator and filter - which I do - a separate one for the DAC, DDC chain, PC server.  The Art Audio provide 40dB of attenuation and isolation - right up to 1Mhz levels.

The reason LPS's can create AC leakage is the charging of their PS filter caps - that is usually large on initial power up and diminishes with warmup stability.  The reason low ESR caps are important.  The degree of further leakage will be determined by the current draw - for min draw situations that would be quite small.

In any event with 40dB of isloation - then another 40dB of filtering on you have 80dB of AC line noise reduction.  To near immeasurable levels.  Now if your gear uses R-Core power supply transformers - they have very high PSRR in and of themselves.  Take the Rednet Ferex SMPS with 150,000uv of AC line noise injection - that is a problem.  The MEIYEN is approx 300uv after turn on.  On the DC side 13uv at 24VDC, 8uv at 5VDC.

Let's see you power a Rednet with a LPS-1 - good luck friend.

I have much more info here for anybody curious:

http://www.head-fi.org/t/821621/audio-power-supplies-part-1-smps-lps-supercap-battery-diy-route-new-devices-opens-up-new-options

http://www.head-fi.org/t/821731/audio-power-supplies-part-2-smps-lps-supercap-battery-diy-route-new-devices-opens-up-new-options

http://www.head-fi.org/t/822160/audio-power-supplies-part3-smps-lps-supercap-battery-diy-route-new-devices-opens-up-new-options

The MEIYEN is a discrete designed based on the single version of the AMB σ11 regulation scheme.
http://www.amb.org/audio/sigma11/

I do not wish to start a debate about battery supply vs LPS. All I know is that if you make sure that the Li-ion BPS have a charged value that is the same as the maximum voltage of your powered device you will be fine! Also the output A is important for the final result!

Leakage loops are very important for SQ. If you have LPS it is crucial to connect the all the LPS to the same unfiltered powerstrip connected to a IT w/ floating secondary for best results!

 

[rb2013]

A general rule is to limit the amounts of DC powered devices in the chain. Any "tweak" that is improving something needs to be passive...and if absolutely necessary it needs to be powered isolated from the AC mains. Keep that in mind for AOIP as well!

Battery power is passive?  Anyway not my kind of sound - not when I'm getting these explosive dynamics.  So lifelike!
 
For me all the theory is fine - but after doing this for 25+yrs I know what I like.
 
Weren't you using a UpNP/DNLA USb chain?
 
Cheers

 

[rb2013]

http://www.digikey.com/en/articles/techzone/2016/sep/a-designer-guide-fast-lithium-ion-battery-charging

 

[johnjen]

 
Is that on a Mac 10.9.5+ ? Does it sound any different? Does it follow the sample rate change of a preferred player? Sorry a few questions.

Yes 10.9.5
No, I can't hear any differences.
No, the SR Follow still doesn't work for me using Jriver
 
JJ

 

[astrostar59]

  Yes 10.9.5
No, I can't hear any differences.
No, the SR Follow still doesn't work for me using Jriver
 
JJ

Ok, thanks for that. I will wait for another update then.
 
I play 44.1 no up sampling as have an R-2R DAC NOS and use 95% 44.1 material.
 
What are your settings in Rednet Controller as mine?
Mode: AES
Word Clock Termination: Yes
Input Devices Off
Output Devices: Off
Input Protocol: Off
Output Protocol: Off
 
In Dante Controller
I set Preferred bit depth as PCM32
Unicast Delay Requests: Off
 
On Mac
I can set 44.1 as 24bit Integer (no option for 32bit)
 
In Audirvana+
I set Direct Mode and Asyncronious as active.
Max out hog mode.
 
Thanks for your help

 

[johnjen]

 
Ok, thanks for that. I will wait for another update then.
 
I play 44.1 no up sampling as have an R-2R DAC NOS and use 95% 44.1 material.
 
What are your settings in Rednet Controller as mine?
Mode: AES
Word Clock Termination: Yes
Input Devices Off
Output Devices: Off
Input Protocol: Off
Output Protocol: Off
 
In Dante Controller
I set Preferred bit depth as PCM32
Unicast Delay Requests: Off
 
On Mac
I can set 44.1 as 24bit Integer (no option for 32bit)
 
In Audirvana+
I set Direct Mode and Asyncronious as active.
Max out hog mode.
 
Thanks for your help

Yes to all of those settings.
 
I use Jriver which has a bit different setup to feed the data stream.
But for me the latency settings inside Dante Controller are all set using the smallest window possible.
I don't know if this is all that important but I would think the shorter the latency window that you can run, the better (based upon any occurrence of drop outs or 'hiccups').
 
JJ

 

[astrostar59]

 
Battery power is passive?  Anyway not my kind of sound - not when I'm getting these explosive dynamics.  So lifelike!
 
For me all the theory is fine - but after doing this for 25+yrs I know what I like.
 
Weren't you using a UpNP/DNLA USb chain?
 
Cheers

Not wanting to complicate things, but a power regenerator IMO gives huge gains in filtered and clean / stable power. Sonically made a big difference. Measured my gets 2.5%+ mains distortion in and 0.1% out. Put the LPS and the rest of the system on that and equates to cleaner power.

 

[astrostar59]

  Yes to all of those settings.
 
I use Jriver which has a bit different setup to feed the data stream.
But for me the latency settings inside Dante Controller are all set using the smallest window possible.
I don't know if this is all that important but I would think the shorter the latency window that you can run, the better (based upon any occurrence of drop outs or 'hiccups').
 
JJ

Thanks that is good to know. Yes latency is the enemy of Audio and jitter. One of the failures in USB IMO. I do graphic work for various clients and move big high resolution files around. Saving any back out to a USB external is a disaster. Too slow and stop starts.

 

[astrostar59]

  Well not to turn this into a DAC thread - the APL NWO uses the finest Lunduhl output transformers for coupling.  The history of Zanden and Audio Note are intertwined.  The Zanden using the TDA1541A and a proprietary analog filter.  http://www.stereophile.com/content/audio-note-uk-vs-kondo-audio-note-japan-1

Where has the history of Audio Note got anything in common with Zanden? Audio Note Japan and Audio UK split. Zanden has no link. Audio Note Japan doesn't have a DAC product right now.
 
Think this ego battle of your DAC v mine is going nowhere. Listening is the only way. Best get back on subject......

 

[Iving]

  Have you visited my listening room?  Who started this thread?  Who started the XU208 USB thread? 
 
I doubt there is anyone on this thread that has owned, lived with, tweeked both AOIP and USB chains to the degree I have.
 
What makes sense is immaterial - what sounds best is what counts!  I didn't think this new development would be popular here - but I'm and equal opportunity audiophile.  No hidden biases - just meticulous attention  to various configurations.  And I know what sounds best to me, and this ain't my first rodeo - of course YMMV.
 
I started this thread based on my perceived audio sound quality superiority of different forms of AOIP.  But as I stated many pages ago - the train does stop here - thank god!
 
What I heard yesterday has truly floored me - as did the Rednet RN3 did when I first heard it, and the subsequent improvements by adding the Mutec MC-3+ as SPDIF (later AES) reclocker, later the Antelope OXCO as Wclock, multiple Cerious Tech Graphene Extreme power cables (another of my finds that became very popular), Oyaide BNC Wclock cables, etc...
 
Yes I will start a new thread - especially since the XMOS XU208 F-1 was closed.
 

 
There are a lot of "I" assertions in that!
None of it speaks to my reluctance to re-contemplate USB.
This AOIP thread has been great for me building my present system - for which many thanks.
I am beginning to love what I have, and don't need to invite back in complex USB-hinged arrays in order to enjoy my music.
That is what I mean by "nope".
(Plus of course, "lol" means, "Look at how we so strenuously denied USB in the beginning!")
Absolutely nothing against you pioneering boys pushing the envelope.
But for me there is mojo where I'm at. I'll keep an eye on y'all tho' - even if I don't contribute