[hopkins]

There is a Ultimo development board advertised on the Audinate web - I have inquired about it but have not received an answer yet. It has I2S outputs, so that could be an interesting solution as well, for those of us that have I2S inputs. I am currently using the F-1 XMOS card's I2S output, and would be curious to see how the two solutions compare.

 

[rb2013]

   
"Well USB will e around at the low end of audio and for mouses and keyboards - as it was designed."
>And I suppose Ethernet was designed for audio?  Great logic there.
 
"PC>PPA V2 USB card (TeraDak X1 - Nichicon HW caps)>Forza Twin Copper Split USB cable>iPurifier2>Startech LEX>BJC CAT6 UTP 550Mhz>Startech REX (with one Jitterbug - Breeze 24VDC LPS+DC iPurifer)>LH Labs 2G split USB cable (data only leg)>W4S Recovery(TeraDak LPS DC-30W+Cerious Graphene PC+ DC iPurifier)>Curious 200mm USB Link>F-1>AS Statement Silver SPDIF>APL DAC"
>???  Any one of those devices will kill the USB sound quality from a well set up computer source.  I am not surprised in the slightest that the Rednet 3 sounds better than the above.  More convoluted is not better.  I can recommend some parts that fix the problems without adding new issues: Wireworld Platinum Starlight USB (short as possible), JPlay JCAT USB card, HDPLEX 300 Watt ATX PSU, SSD storage, Daphile OS.  Each add-on fix-it box adds latency, potential grounding issues, and does not guarantee better operation of the USB receiver nor a "cleaner" USB signal.  
 
In my experience, the best way to do USB is to have a high quality USB receiver inside the DAC.  The signal has lowest jitter immediately after reclocking, so it makes sense to minimise distance and convolution between this and the DAC.  My DAC is not high end, but the humble USB card inside it has shamed even the highest performance external USB->SPDIF (In my experience, this is the Berkeley BADA USB).  Feed one of those with the best USB cable (mentioned above) and a well optimised computer source and that IME is the best way to do USB.
 
SPDIF or AES are not optimal input for most DAC as the jitter is much higher than USB.  SPDIF receiver PLL is around 50 ps, Amanero USB is 10 ps.  If you are concerned about USB noise - there are USB input cards that galvanically isolate after the USB receiver and before FPGA reclocking, and that clean low jitter low noise I2S can go straight into the DAC.  Without reclocking, galvanic isolation adds jitter, so a lot of DAC and D/D don't use it.  Maybe if DAC has good reclocking or clock recovery (possible word clock input) of SPDIF input the jitter might matter less.  LVDSI2S over HDMI is a fairly good way to use external converter without bottleneck of SPDIF.
 
Don't get me wrong I have no doubt REDNET 3 is very good, just given what it is being compared to, I don't really have much of a reference point.  IF we are comparing the REDNET to a more sensible chain with comparable price USB/SPDIF, then I would have more confidence. 
 
Right now I have moderate confidence that REDNET 3 would be very good alternative for external SPDIF interface.  Only thing I don't want external SPDIF interface because it is inherently compromised due to SPDIF.  If they have internal card that outputs I2S, and computer solution that works driverless with Linux, that would be great.  
I just don't like the prospect of going backwards to go forwards both for sound and ease of use (eg have to use Windows, SPDIF etc)

Yeah but USB is very limited in data capacity - that is why USB 2.0 Audio have to be created - but no error correction.  MS to this day does not support it.  Actually AES67/Ethernet is the ideal transport medium.  USB sucks because of the PC ground issues and data and power running side by side, just to name a few.  But hey if you think USB is perfect and you have it nailed down - all the power to you.  Enjoy!
 
>???  Any one of those devices will kill the USB sound quality from a well set up computer source.  I am not surprised in the slightest that the Rednet 3 sounds better than the above.  More convoluted is not better.  I can recommend some parts that fix the problems without adding new issues: Wireworld Platinum Starlight USB (short as possible), JPlay JCAT USB card, HDPLEX 300 Watt ATX PSU, SSD storage, Daphile OS.  Each add-on fix-it box adds latency, potential grounding issues, and does not guarantee better operation of the USB receiver nor a "cleaner" USB signal. <
 
Well I can see we are on separate pages I had a Plat Starlight  USB (one of about a dozen I tried) mediocre for sure.  Lightspeed crushes it - and not even the 10G just the 2G.
HDPlex - not the last word in PC PS.  SSD - big deal.  JCAT card - highly doubt this is that much better then a PPA V2 fed by a LPS.  But Marcin is going to send one to me to try - so will give that a go.  Anyway still no galvanic isolation - like the ICRON/Startech GB LAN Iso Audio USB...so highly doubtful of your comments.
 
>In my experience, the best way to do USB is to have a high quality USB receiver inside the DAC.  The signal has lowest jitter immediately after reclocking, so it makes sense to minimize distance and convolution between this and the DAC.  My DAC is not high end, but the humble USB card inside it has shamed even the highest performance external USB->SPDIF (In my experience, this is the Berkeley BADA USB).  Feed one of those with the best USB cable (mentioned above) and a well optimised computer source and that IME is the best way to do U<
 
OK now you have lost ALL credibility with me - just jumped the shark.  Inside the DAC causes PHY AGC induced modulation  - feeding high noise levels into the DAC clocks.  You need to do a little more research my friend...that and the countless folks who have reported that a good external DDC feeding by SPDIF/AES just crushes their internal i2s card (Amerno in particular).  The rest of you comment is just rehashing old ground that has be put to bed long ago...next up why we absolutely have to have 784kPC and 512 DSD to get any decent SQ...and on and on it goes...
 

 
OK here comes the 'USB is the greatest crowd'  - wonder what took them so long?
 
Oh and BTW I have 100% confidence as I have REDNET playing right now as I type this - the SQ crushes the Mutec MC-3+ USB - ever hear that one chappie?

 

[rb2013]

  I got back this reply from Audinate, makers of the Dante technology and the DVS software, when I was trying to config my D16 to 192K. Actually it was in reply to my letting them know I had already figured it out but wanted to prompt them about the blooming audiophile market.
 
Sounds promising...
 
"Funny you mention SPDIF to Dante.
I have been asked for a simple conversion box for the last couple of months by a lot of end users. The least expensive solution right now is the Focusrite RedNet3 box, but that’s over 3000 dollars.
So I mentioned it at the InfoComm trade show last week to: Focusrite, RDL, and AtteroTech.
So maybe we will see something come out that can do that in the next few months.
As for AES67, we are compatible in the newer cards with our latest firmware update, but the Dante standard is not based on it.
Dante Virtual Soundcard automatically adjusts its sample rate to match the software on the computer, but if we allowed auto adjustment across the network that would cause a major problem in pro audio settings.
Can you imagine what would happen if a twenty thousand seat arena suddenly switched sample rates just because someone turned on DVS?
That would be bad."

Nice!  So it can be done with DANTE - not what BURL told me.  He said they specifically asked Audinate to update DANTE to give it that capability.  One question for them  - why the 'SR Follow' button?  It's not like the DANTE card is separate from the RN3 or 16 - they intergral.  Maybe having that ready for a later firmware update?

 

[Danutz]

Hello,
 
Which one to buy Rednet 3 or D16 (almost same price) or even A8R-8...?
 
Any major difference? Will be used PC-Rednet-NadM51Dac
 
Thank you very much,
 
Dan

 

[mhamel]

  Hello,
 
Which one to buy Rednet 3 or D16 (almost same price) or even A8R-8...?
 
Any major difference? Will be used PC-Rednet-NadM51Dac
 
Thank you very much,
 
Dan

No need to spend extra on the "r" models - those are built for redundancy and failover which really aren't needed in consumer settings.  
 
Also, the A8 is analog A/D and D/A, so if you're looking for digital out, it's not the one you want.

 

[Danutz]

Hi mhamel,
 
so no difference between Rednet 3 an d16.... I'm really not understanding their configurations, but heard that d16 would use the newer (better???) Brooklyn II Dante.... is that older version than Dante Ultimo?
 
It's used by this product: http://www.fouraudio.com/en/products/dbs1.html
 
but I can't find a price of this item.... Is it comparable to the Rednets?
 
Thanx,
 
Dan

 

[rb2013]

  Hi mhamel,
 
so no difference between Rednet 3 an d16.... I'm really not understanding their configurations, but heard that d16 would use the newer (better???) Brooklyn II Dante.... is that older version than Dante Ultimo?
 
It's used by this product: http://www.fouraudio.com/en/products/dbs1.html
 
but I can't find a price of this item.... Is it comparable to the Rednets?
 
Thanx,
 
Dan

No the REDNET 3 uses the first Brooklyn card - it did not do 178k SR.
 
The noise specs are exactly the same:
 
RN3:

Input sample rate range	32 to 216 kHz
Gain error	-0.3 dB
Dynamic Range	> 138 dB (-60 dBFS method)
THD+N	< -130 dB (0.00003%); 0 dBFS input

RN16AES:

 

Input sample rate range	32 to 216 kHz
Gain error	'-0.3 dB
Dynamic Range	> 138 dB (-60 dBFS method)
THD+N	< -130 dB (0.00003%); 0 dBFS input

 
 
Here is the difference:
RN3:

44.1 / 48 / 88.2 / 96 / 192 kHz

 
RN16AES:

 

Supported sample rates

44.1 / 48 / 88.2 / 96 / 176.4 / 192 kHz (-4% / -0.1% / +0.1% / +4.167%)

 

[Danutz]

ok, thanks.... what about the fouraudio productline? Dante Ultimo

 

[mhamel]

  ok, thanks.... what about the fouraudio productline? Dante Ultimo

 
Ultimo is their lower cost chipset, it maxes out at 96KHz.

 

[drez]

  Yeah but USB is very limited in data capacity - that is why USB 2.0 Audio have to be created - but no error correction.  MS to this day does not support it.  Actually AES67/Ethernet is the ideal transport medium.  USB sucks because of the PC ground issues and data and power running side by side, just to name a few.  But hey if you think USB is perfect and you have it nailed down - all the power to you.  Enjoy!
 
>???  Any one of those devices will kill the USB sound quality from a well set up computer source.  I am not surprised in the slightest that the Rednet 3 sounds better than the above.  More convoluted is not better.  I can recommend some parts that fix the problems without adding new issues: Wireworld Platinum Starlight USB (short as possible), JPlay JCAT USB card, HDPLEX 300 Watt ATX PSU, SSD storage, Daphile OS.  Each add-on fix-it box adds latency, potential grounding issues, and does not guarantee better operation of the USB receiver nor a "cleaner" USB signal. <
 
Well I can see we are on separate pages I had a Plat Starlight  USB (one of about a dozen I tried) mediocre for sure.  Lightspeed crushes it - and not even the 10G just the 2G.
HDPlex - not the last word in PC PS.  SSD - big deal.  JCAT card - highly doubt this is that much better then a PPA V2 fed by a LPS.  But Marcin is going to send one to me to try - so will give that a go.  Anyway still no galvanic isolation - like the ICRON/Startech GB LAN Iso Audio USB...so highly doubtful of your comments.
 
>In my experience, the best way to do USB is to have a high quality USB receiver inside the DAC.  The signal has lowest jitter immediately after reclocking, so it makes sense to minimize distance and convolution between this and the DAC.  My DAC is not high end, but the humble USB card inside it has shamed even the highest performance external USB->SPDIF (In my experience, this is the Berkeley BADA USB).  Feed one of those with the best USB cable (mentioned above) and a well optimised computer source and that IME is the best way to do U<
 
OK now you have lost ALL credibility with me - just jumped the shark.  Inside the DAC causes PHY AGC induced modulation  - feeding high noise levels into the DAC clocks.  You need to do a little more research my friend...that and the countless folks who have reported that a good external DDC feeding by SPDIF/AES just crushes their internal i2s card (Amerno in particular).  The rest of you comment is just rehashing old ground that has be put to bed long ago...next up why we absolutely have to have 784kPC and 512 DSD to get any decent SQ...and on and on it goes...
 

 
OK here comes the 'USB is the greatest crowd'  - wonder what took them so long?
 
Oh and BTW I have 100% confidence as I have REDNET playing right now as I type this - the SQ crushes the Mutec MC-3+ USB - ever hear that one chappie?

 
Power and data running side by side is only a big deal if the power line has noise.  USB card with low noise 5V this wont be an issue.  
 
I have tried Lightspeed 10G.  Very sterile and lacking in nuance level detail.  Beaten by even the series 6 Platinum.  The full linear HDPlex PSU is the best fully ATX compatible solution I have tried.  JCAT is better than PPA V2.
 
Galvanic isolation is necessary for Ethernet because it needs to connect between rooms which might have different ground reference and therefore might create ground loop.  You can galvanically isolate USB if you like - it doesn't really help a whole lot and in my experience is a step backwards.  Even isolating the ground pin is a step backwards.  The galvanic isolation does provide some noise rejection and DC isolation but it also adds jitter.  The net gain is just not there.  Most of the good USB solution use galvanic isolation after the USB receiver, that way the PHY noise is isolated from the output.  Again the net gain is not a given.
 
What makes you think Ethernet is free from PHY noise?  In my experience NIC's put out a lot of noise.  I need to enable my NIC because I am running a headless machine.  If were not running a headless machine, I would disable the NIC as it produces much noise.
 
Secondly, have you measured the PHY noise and compared it to Ethernet, or are we just comparing one D/D implementation to another.  There are so many more factors that might be affecting the sound of the Rednet vs your USB based D/D.  Not least the quality of the reclocking, SPDIF output quality, power supplies etc.  To simply decide that the difference is 100% down to using AOIP is an assumption.  It could be valid assumption, but we would need to isolate so many other factors to make that determination.
 
I cannot account for people preferring AES/SPDIF over USB.  As mentioned the interface is a bottleneck.  If they were using LVDSI2S I would have slightly less reservations.  Impressions come from so many different setups - perhaps they have poor computer sources.  Also different systems tend to show or hide things in different ways.  Even in my own system as it develops, positions are overturned.  As resolution improved the true nature of things come out more.  What was previously observed to by detailed turns out just to be thin or harsh.  Could be that my system is still at an interstitial stage where USB appears to sound better.  However my own preference for USB came only AFTER upgrading to the HDPlex PSU and case.  We have to rationalise the world with the information we have available.  All I know is that on a technical level SPDIF is a sure bottleneck, USB evils are contingent and for the most part unverified.  LVDSI2S is much less bottleneck.
 
I used to use an external D/D until I improved my computer, but of course [computer upgrading] was costly and not possible for everyone.

 

[drez]

I could add another anecdote on how noisy and un-isolated Ethernet is in practice.  See photo below:
 

 
As you can see the ethernet switch is powered from linear PSU and isolated from vibrations.  I thought this would be dumb.  Ethernet is not even in the signal path.  Even so adding linear power and vibration isolation to the switch improved the sound from my system 

 

[jabbr]

Nice!  So it can be done with DANTE - not what BURL told me.  He said they specifically asked Audinate to update DANTE to give it that capability.  One question for them  - why the 'SR Follow' button?  It's not like the DANTE card is separate from the RN3 or 16 - they intergral.  Maybe having that ready for a later firmware update?

I had contact with Focusrite support about RD3 / RD16 automatically following sample rates on the PC.
To be short: it cannot be done.

This is the answer from Focusrite support:

The RedNet 3 and RedNet D16 AES will not automatically follow the sample rate/bit rate change when you make changes to your DAW or Dante Virtual Soundcard.

Whenever you may a change to your DAW you will need to make the change to Dante Virtual Soundcard as well as whatever RedNet or Dante Device that is in the Dante Network. Having to manually make the changes to all devices in the Dante Network allow for a more flexible and robust audio eco system. This will also allow for you to have a customized environment in which you can have different devices at different sample rates. You can make custom presets in Dante Controller in which you can create that will change the sample rate of both your RedNet interfaces and Dante Virtual Soundcard.

 

[rw35]

  I could add another anecdote on how noisy and un-isolated Ethernet is in practice.  See photo below:
 

 
As you can see the ethernet switch is powered from linear PSU and isolated from vibrations.  I thought this would be dumb.  Ethernet is not even in the signal path.  Even so adding linear power and vibration isolation to the switch improved the sound from my system 

 
Both USB and Ethernet are noisy but Ethernet seems to handle the data transfer better. USB seems more vulnerable to noise and thus it works best when the sender is of very high quality (Aurender, etc). LAN seems to negate much of the electrical issues that computers generate in terms of SQ degradation.
 
My gut feeling is that the same levels of SQ are achievable with both, but LAN transfer bypasses most of the electrical problems, whilst even the best USB chains need helper devices. That said my LAN has fibre media converters to remove noise.
 
Buy any PC with a normal USB port and an AOIP device - AOIP will probably trounce USB, that's what everyone who has tried RedNet has said to date.
 
I'm lucky that my USB play is fabulous (bet it's very close or as good), but I'm not typical of what people are typing.
 
AOIP is early but it's got serious legs in my book, it makes sense. My only issue is I don't generally like pro audio (owner of recoding studio) people as they are not usually as attentive to detail as high-end guys.
 
A 2 channel high-end AOIP device is sure to be on its way soon, then I'll have to decide whether to drop Aurender (if they don't adopt it).
 

 
---
 
Source:
*Aurender N100 (no internal disk : LAN optically isolated via FMC with *LPS) > DIY 5cm USB link (split for *LPS) > Intona Industrial (injected *LPS) > Curious short USB link > W4S Recovery (*LPS) > DIY 5cm USB link (5v rail removed) > *Auralic VEGA (balanced)

Control:
*Jeff Rowland CAPRI S2 (balanced)

Playback:
2 x Revel B15a subs (balanced) > ATC SCM 50 ASL (balanced - 80Hz high pass filtered from subs)

Misc:
*Via Power Inspired AG1500 AC Regenerator
LPS: 2 x Swagman Lab Audiophile SE (W4S & FMC) & 1 x SMSL P1 (Intona power injection)
Storage: QNAP TS-253Pro 2x 3Tb, 8Gb RAM
Cables: DIY heavy gauge solid silver (balanced)
Mains: dedicated distribution board with 5 x 2 socket ring mains, all mains cables: Mark Grant Black Series DSP 2.5 Dual Screen

 

[mtoc]

I had contact with Focusrite support about RD3 / RD16 automatically following sample rates on the PC.
To be short: it cannot be done.

 
I've said before, you have to add a MCLK to it if you wanna auto-sample rate switching, and if you use some software to unsample all your stuff to 192kHz, that'z no good. A few more clicks everything's fine.

 

[jabbr]

I've said before, you have to add a MCLK to it if you wanna auto-sample rate switching, and if you use some software to unsample all your stuff to 192kHz, that'z no good. A few more clicks everything's fine.

Not true i'm afraid, adding an external master clock will not create automatic sample rate switching.

Just got some additional information from Focusrite about developments that will be released in a few months that will allow it:

I just wanted to follow up on your conversation with Louie yesterday.

At present, DVS cannot be used to change the sample rate of RedNet devices on the network, however this is something we're looking to alter with the next release of RedNet Control, which we are aiming to have available towards the end of August (version 1.10).

Please note that, for this to work in the way you desire, the playback program you are using would need to be able to change the sample rate of the playback driver (in this case, DVS). Typically, this is only possible from playback software that utilises ASIO drivers (pretty much all DAW software + some media players such as Foobar).

With the next release of RedNet Control, once the playback software changes the sample rate DVS is working at this can be configured to also change the sample rate of other RedNet devices on your network.

So with a little patience it will become available.

Cheers