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

[johnjen]

   
My head tells me this should be so. But I keep listening to redbook upsampled to 192 instead of 176.4 - because of first hand experience. I enjoy 192 more - I think!

In my experiments I've tried the 192 and the 96 and the 88 Khz SR's using both step up and step down sample rate conversions.
The differences are slight, and are not, OMG! this is WAY better…
 
This to me is a high praise in that we can up sample and down sample with minimal differences in terms of SQ changes.
This isn't to say there are no differences, but it seems like sticking with even multiples (x2, /2) is slightly better than jumping across the 44.1 to 48KHz (and their multiples thereof) sample rate series.
Which is why I use the 88.2 as my SR of choice.
 
JJ

 

[goodvibes]

   
Science for the most part involves experimentation or an *empirical* approach (= "based on, concerned with, or verifiable by observation or experience rather than theory or pure logic"). Scientists "scrutinise" by *replicating* other people's results; i.e., subjecting  them to "falsification". You can't "falsify" someone else's observations from the armchair rather than the laboratory. Good reviews tend to focus more on "scrutinising" poor methodology - but that could hardly apply here where the weight of "evidence" (first hand aural experience) is so unmissable. Folks responding to you (as I) are trying to say that you have to hear a RedNet to understand what it can deliver in SQ terms. Those on this thread that have done so do tend to appreciate it. I can understand that if you've invested significantly in an alternative, cognitive dissonance may restrain your willingness to experiment, and perhaps it is difficult to road test these boxes before buying.

 I reserve the right to respond to quotes.
 
Just to be clear. The cost of a REDNET setup isn't an issue and I readily admit to not hearing it so have no 1st hand experience of performance nor have ever implied as much. That was never in question. Just pointing out what may or may not be limitations. I remember when USB via Itunes and before that, CDs were perfect sound forever and praised. I never accepted either of those 'truths' either, regardless of what others may have 'heard' or argued. Other's 1st hand experience is not first hand for me. I currently have no opinion of the actual sound of Rednet and never said I did. I do believe this could be much better than USB because, for me, USB isn't very good. Like I said, I was already using JET clocking via firewire a decade ago and know it's a better way. I appreciate that Rednet uses it as well. 

 
My point was that with UPNP, establishing a clean stream at the source is what's needed for a good result and clocking doesn't require a solution because all playback is localized within the renderer. With a good all in one renderer and no need for an additional SPdif or other digital interface, clocking is a non issue to the ability of the playback device which is a given in every setup.  The advantage of rednet isn't as much accuracy as it's realtime aspects which are great. That doesn't mean it isn't accurate (enough or very) but the emphasis is realtime with sufficient accuracy. It won't clock better than proper UPNP nor be more bit correct than proper UPNP. In fact, it could be less so but not necessarily so. It will have significantly less latency for real time type work but that isn't really important for home use. None of that means it isn't accurate or at least accurate enough to be invisible for most or even all but they are points worth making even if you don't want to hear them. Being this sensitive to these points does not make you appear more objective.
 
If there's someone with a setup in Chicago, I'd love to hear it and compare. PM me.

 

 

[Iving]

   I reserve the right to respond to quotes.
 
Just to be clear. The cost of a REDNET setup isn't an issue and I readily admit to not hearing it so have no 1st hand experience of performance nor have ever implied as much. That was never in question. Just pointing out what may or may not be limitations. I remember when USB via Itunes and before that, CDs were perfect sound forever and praised. I never accepted either of those 'truths' either, regardless of what others may have 'heard' or argued. Other's 1st hand experience is not first hand for me. I currently have no opinion of the actual sound of Rednet and never said I did. I do believe this could be much better than USB because, for me, USB isn't very good. Like I said, I was already using JET clocking via firewire a decade ago and know it's a better way. I appreciate that Rednet uses it as well. 

 
My point was that with UPNP, establishing a clean stream at the source is what's needed for a good result and clocking doesn't require a solution because all playback is localized within the renderer. With a good all in one renderer and no need for an additional SPdif or other digital interface, clocking is a non issue to the ability of the playback device which is a given in every setup.  The advantage of rednet isn't as much accuracy as it's realtime aspects which are great. That doesn't mean it isn't accurate (enough or very) but the emphasis is realtime with sufficient accuracy. It won't clock better than proper UPNP nor be more bit correct than proper UPNP. In fact, it could be less so but not necessarily so. It will have significantly less latency for real time type work but that isn't really important for home use. None of that means it isn't accurate or at least accurate enough to be invisible for most or even all but they are points worth making even if you don't want to hear them. Being this sensitive to these points does not make you appear more objective.
 
If there's someone with a setup in Chicago, I'd love to hear it and compare. PM me.

 

 
Arguing that one football team has better credentials than another when you've never attended a match between them is hardly legitimate "scrutiny". And I object to your (impolite) appropriation of "science" in a challenge to this thread when you said, "If you don't want any scrutiny, which is part of good science, make it an appreciation thread". That is my only sensitivity to what you are saying, and I have already explained why. Your remarks about UPnP don't mean anything to me - I am no longer interested in UPnP. This isn't a UPnP thread.

 

[astrostar59]

Oh my goodness 'goodvibes' lets drop the verbage and get to hear the thing, please. Hearing is believing. I for one am now 100% happy with digital after a 20 year love/hate affair with it.
 
Of course, AOIP is part of the story, you still need a good server, and then a good DAC and HPs or speakers. But getting USB out of the chain is a no brainer to me now. All those silly USB devices and money spent and it never got a clean sound. That saw tooth treble and on/off modulation in the high registers used to drive me nuts. The small soundstage and restricted dynamics, a haze about everything, the false detail. I could go on.
 
What we probably need now is an audiophile AOIP box that is a bit smaller and less channels, probably in black and more plug n play. Then it will be flying off the shelves! Hell, I may just make one myself.... what number is my bank manager? 

 

[Iving]

saw tooth treble and on/off modulation in the high registers used to drive me nuts. The small soundstage and restricted dynamics, a haze about everything, the false detail 

 
... matches my first hand experience (and nearly matches my own capacity for describing it )

Hell, I may just make one myself.... what number is my bank manager? 

 
lol - Go astro

 

[goodvibes]

I'm open. Chill guys. Can't wait to hear it. I'm hoping there's someone in this big city willing to help with a compare. Seems the criticism is of me and not my tech points so I'm good. The notion put forth that I'm protecting my investment with bias is what got this going and honestly, I'm all for cheaper and better so lets move on. I have never said it was worse than UPNP. Just don't understand why it would be better for home use and pointing at differences. I'm off till I get a listen and I hope someone PMs me with an opportunity. Cheers all.

 

[Iving]

  Just don't understand why it would be better for home use

 
My main criteria for choosing equipment are SQ and affordability.
It is Pro Audio gear. Some domestic audiophiles don't like that - but I don't mind it at all - in fact I like the robust feel.
Some folks don't like that these interfaces are red - but I don't mind the colour at all. (I like people and things just as they were created - you know - up to a point.)
So, for me, "better for home use" just translates as "sounds better and I can afford it" (even if I am deluded on that latter count).
Good luck with an audition.
I don't think we have encountered anybody yet who has dismissed the RedNets on SQ grounds after any kind of protracted hearing.

 

[rb2013]

  Not running DSD, but I am upsampling to 176.4 w/sinc filter.
 

 
Only reason for posting is to give you an idea of CPU load.

What CPU?  A iCore 7 is going to have a lower load then the iCore 5 they used.  One reason I'm a little hesitant to move away from the 7 to the lower processing N3150 Quad cores.
 
The Dante DVS does need to processing power, esp at the highest SR rates.

 

[rb2013]

  Because I like music, technology and may have a use for it at some point. Not everything needs to be the absolute best at everything to be the right thing for something. Why are you so defensive of the tech and offensive to posters at the same time? My posts are neither aggressive nor dismissive and I didn't see this as a REDNET appreciation thread but you're right, I have no need to be insulted here, multiple times. I left for a few weeks earlier after a couple posts for the same reason and here we are again. If you don't want any scrutiny, which is part of good science, make it an appreciation thread.
 
I honestly wish you well and hope you're more open to other views in the future. 

I'm all for more opinions - but I'm skeptical of those who declare their USB or UpNP/DNLA chains better then AOIP (whether DANTE or RAVENNA) - without ever hearing them.  Then go on and on theorizing as to why it sounds inferior.

  If you read the beginning of this thread - many, who have not heard, tried shooting down AOIP in the beginning.  Fortunately we persisted  - armed with our own listening experience.  In the many months since - quite a few others have tried the Rednet boxes and have come away very impressed.  You're a little late to the AOIP skeptics dance.  That's why I wonder why you spend so much energy fighting that losing battle?  Just to rain on other's parade?
 
I welcome any knowledgeable debate - and I certainly don't think we have reached computer Nirvana with DANTE - it does have it's warts.  But to just crash the party with silly critiques of this technology is going to get push back. Posting inaccurate, and somewhat silly stuff is going to get rebuked - sorry if you don't like that.  I cut you some slack from your posting - as you are new to this thread - but to set the record straight.  You wrote:

  10/23/16 at 7:56am

1. [img]http://cdn.head-fi.org/2/20/100x100px-LS-20d8384f_headphoneavatar.png[/img]

1. goodvibes
2. 
3.  
4. online

1. 7,955 Posts. Joined 12/2009
2. Location: USA

and favors low latency with error correction over buffering for accuracy. I'm not looking to digitally reformat and recover my music whether digital or analog. I can see this as great for real time pro use but I don't see myself ever going here for personal listening. Like most new formats, there is both good and bad. Early adopters concentrate on the good and the bad is usually discovered later, when practical applications are more scrutinized. Pro or studio kit isn't always better than Home kit. They have different needs and uses. I'm familiar with both and find adopting top home kit to studio use tends to work better than the other way around, if, durability and use are a fit and price isn't a major obstacle. I have yet to hear a Genelec speaker I could listen to at home yet they work well enough for their intended use. It's all rather relative and not all good or all bad and generally these sorts of things are designed to fill specific needs. The amount of latency for good home hidef UPNP listening is negligible and no issue in use. Bringing that up as a reason for REDNET is a complex solution looking for a problem.

Wrong on the 'error correction' and sorry to break the news to you - but almost all recorded music you hear has been 'reformatted and recovered' - both digital and analog.  How do you think those grooves get into the LP?  Pits in those CD's?  Bits get turned into analog signals your amps can use? Nobody ever said that 'Pro or studio kit' is better then 'Home kit' - where did you get that impression?  If you have read some of this thread - you will see my and others - criticism of Pro Audio gear.  I have pointed out the flawed SMPS power supplies as a big one.  You mention my recommending AOIP over UpNP/DNLA - saying Rednet is a 'complex solution' to a non-existent problem with UpNP/DNLA - and that's NOT complicated??
 
 
BTW this is audio and NOT science - but if you want to read about the hypocrisy in science - try reading Mathematician and Physicist Peter Woit's 'Not Even Wrong'.

 

[rb2013]

And lastly sox has been added to the latest version of Jriver Media Center and it does contribute to better SQ. Not by a huge amount (at least in my system) but it does help.
 
JJ

Good to know on the JMC add of SoX.  Thanks
 

   
You can't "falsify" someone else's observations from the armchair rather than the laboratory. Good reviews tend to focus more on "scrutinising" poor methodology - but that could hardly apply here where the weight of "evidence" (first hand aural experience) is so unmissable. Folks responding to you (as I) are trying to say that you have to hear a RedNet to understand what it can deliver in SQ terms. Those on this thread that have done so do tend to appreciate it.

Well said my friend!
 

   
My head tells me this should be so. But I keep listening to redbook upsampled to 192 instead of 176.4 - because of first hand experience. I enjoy 192 more - I think!

Same here - going to 192k with the SoX upsampler better then 176k for 44k Redbook files.  Same for 96k vs 88k with my R2R DAC.
 

  Thanks!
 
And yeah the lesson from using the LPS on the Fibre Media Converter, if it were applied to the RN3 or the RND16 etc., should yield some welcome SQ changes.
 
And to that end I may have found a triple output LPS that won't break the bank, even if it is ugly, and needs a box to put it into, etc.
 
At least it will allow for another experiment to test to see what would happen with swapping out the SMPS for a decent (2mv rms, 3mv P-P, rated noise) linear PSU.
 
The only unknown left to sleuth out is, do these RedNet boxes actually use the 51 volts being supplied by the stock SMPS, or can we ignore that voltage when we replace the SMPS.
I'm hoping it can be ignored…
 
And the last hitch in the git along is the lead time (4-6 weeks) for the triple PSU's I've found, but the price ≈ $100 each is attractive in and of itself.
 
JJ

Nice - looking froward to reading more about that.  That high voltage is an issue.  And why something like the external DC powered Arrakis Simple-IP AES3 XLR may be an interesting path.
http://arrakis-systems.com/pdfs/Simple-IP-8A&D%20Manual.pdf
I see it does not do 176k so it likely has the the Brooklyn I Dante card - like the RN3.  Wonder what the DC input voltage is?

 

[Fredoo]

After following this topic quite some time with interest, does anyone have an impression when the first simplified AOIP DDC or DAC -targeted at home users- will hit the market?? Either from the established pro-audio firms (i.e. Burl, Focusrite,..) of an innovative consumer HiFi brand (i.e. Schiit, ..).

 

[jabbr]

After following this topic quite some time with interest, does anyone have an impression when the first simplified AOIP DDC or DAC -targeted at home users- will hit the market?? Either from the established pro-audio firms (i.e. Burl, Focusrite,..) of an innovative consumer HiFi brand (i.e. Schiit, ..).

My idea is probably never or many years away

 

[3X0]

After following this topic quite some time with interest, does anyone have an impression when the first simplified AOIP DDC or DAC -targeted at home users- will hit the market?? Either from the established pro-audio firms (i.e. Burl, Focusrite,..) of an innovative consumer HiFi brand (i.e. Schiit, ..).

One or two years ago.
 
http://www.soulution-audio.com/en/serie5/560/index.php
http://www.soulution-audio.com/en/serie7/760/index.php
 
US$35k and US$50k respectively.

 

[InsanityOne]

  One or two years ago.
 
http://www.soulution-audio.com/en/serie5/560/index.php
http://www.soulution-audio.com/en/serie7/760/index.php
 
US$35k and US$50k respectively.

I don't think I would consider $35,000 - $50,000 anywhere near "targeted at home users", and probably not even "targeted at pro users" I am not sure what makes those two DACs special but that is "targeted at .0001% of users" pricing.
 
I don't think it is crazy to assume that within the next few years we will see fully AOIP capable DACs in the $500 - $1000 range, I couldn't see prices going much below $500 though as the BK2 card alone costs somewhere around $200 - $300, unless Audinate has some really good bulk pricing that we don't know about.
 
ASIDE: The device I am still very curious about is the Arrakis Simple IP. Is it a DAC, or just a converter like a RedNet device? At $999 it is an extremely tempting buy if it has a DAC. I cannot find any information about it anywhere, and the website is not very insightful. But, if it doesn't have a DAC then you might as well just stick to hunting for a used RN3 for ~$700.
 
- InsanityOne 

 

 

[3X0]

I don't think it is crazy to assume that within the next few years we will see fully AOIP capable DACs in the $500 - $1000 range, I couldn't see prices going much below $500 though as the BK2 card alone costs somewhere around $200 - $300, unless Audinate has some really good bulk pricing that we don't know about.

If that Audinate Brooklyn card pricing is accurate, it is unlikely we would see a AOIP DAC solution that is cheaper than even the RedNet 3. It would need to use something outside of the Brooklyn cards to actually be profitable.