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

Aug 15, 2016 at 3:18 PM Post #1,486 of 3,694
I'm pretty sure the new Mutec Ref10 will not be a Rubidium clock. In a post several months ago I inferred that it was Rubidium and was corrected pretty quickly. My guess is also high quality OCXO.
 
Aug 15, 2016 at 4:38 PM Post #1,487 of 3,694
 
Thanks Iving!
 
I have sent a mail to Focusrite. As far as I can se the specs are the same.
 
Tommy

 
Good idea!
See also particularly here (right hand panel for different models): https://uk.focusrite.com/ethernet-audio-interfaces/rednet
and other resources:
https://support.focusrite.com/hc/en-gb
https://global.focusrite.com/downloads
https://www.audinate.com/products/software/dante-virtual-soundcard?
https://www.audinate.com/training-and-tutorials
 
Aug 15, 2016 at 8:17 PM Post #1,488 of 3,694
No, they don't disclose anything on their components, just measurements of their circuits and functioning of the complete device.

Their marketing speak only goes thisfar:
It is also their PLL design with a 0.1 Hz bandwidth which makes me think a reference clock will not improve this master clock. This bandwidth of 0.1 Hz indicates that any difference between internal oscillator and external reference clock that is greater than 0.1 Hz will be discarded and not applied to correct the internal clock. This means that virtually everything will be discarded, and they can only do that and still have a good master clock when their internal crystal oscillator is extremely good and stable otherwise their master clock would not give good results.


Since the purpose of these devices is clocking - unlike the REDNETs - the clock stability and accuracy is a measure of the functioning of the complete device.  It's just another measure like phase noise.
 
Does it even have a 10M clock input?
 
Aug 15, 2016 at 8:24 PM Post #1,490 of 3,694
Here is an interesting alternative.  Only $149
 
It's OXCO but one that takes a clock signal from orbiting GPS satellites to act as a disciplining ref rate.  Bet those satellites have some pretty fine atomic clocks.
 
Note the 10MHz output and GPS antenna hookup.  Ext DC power as well.
 
http://www.ebay.com/itm/GPS-Disciplined-Oscillator-Clock-GPSDO-10M-Output-Sinusoidal-Wave/251802969291?_trksid=p2047675.c100005.m1851&_trkparms=aid%3D222007%26algo%3DSIC.MBE%26ao%3D1%26asc%3D20131003132420%26meid%3D2f69a02dab9545d8a0afc8e5ce2da852%26pid%3D100005%26rk%3D1%26rkt%3D6%26sd%3D281649576001
 
Aug 15, 2016 at 11:50 PM Post #1,491 of 3,694
Quick update
 
I took out the Liveclock and tried a direct connection from Cybershaft into Mutec again. Same outcome -  still prefer sq with Cybershaft going into liveclock first. Could be as simple as by going into Liveclock first I can synch both RedNet and Mutec to same clock. Music just seems to have more presence and energy. Direct connection to Mutec does not sound bad by any means but it feels slightly more subdued.
 
Also, if Antelope owners out there do get a 10M clock, don't jump the gun on the calibration process. When I connect the Cybershaft and LIveclock and hit "calibration" the initial error is around.22Hz or so. Don't do anything for about about 2-3 minutes. The error rate slowly decreases and finally seems to stabilize for me around .005Hz. I then hit the "calibrate" button again to compensate for the difference to get to 0Hz. The first go around I did not provide enough time for the clocks to calibrate on their own. Live and learn.
 
Aug 16, 2016 at 12:02 AM Post #1,492 of 3,694

 
 
Yes, I noticed the one above on the Cybershaft webpage as well when I was first browsing around. Lots of interesting options indeed.
 
 
Quote:
  Here is an interesting alternative.  Only $149
 
It's OXCO but one that takes a clock signal from orbiting GPS satellites to act as a disciplining ref rate.  Bet those satellites have some pretty fine atomic clocks.
 
Note the 10MHz output and GPS antenna hookup.  Ext DC power as well.
 
http://www.ebay.com/itm/GPS-Disciplined-Oscillator-Clock-GPSDO-10M-Output-Sinusoidal-Wave/251802969291?_trksid=p2047675.c100005.m1851&_trkparms=aid%3D222007%26algo%3DSIC.MBE%26ao%3D1%26asc%3D20131003132420%26meid%3D2f69a02dab9545d8a0afc8e5ce2da852%26pid%3D100005%26rk%3D1%26rkt%3D6%26sd%3D281649576001

 
Aug 16, 2016 at 12:33 AM Post #1,493 of 3,694
Since the purpose of these devices is clocking - unlike the REDNETs - the clock stability and accuracy is a measure of the functioning of the complete device.  It's just another measure like phase noise.

Does it even have a 10M clock input?


Now you come to mention it, it does have a clock input but it appears to be only a word clock?!
The manual doesn't mention 10MHz signals. I had assumed (never assume anything :D ) that it would accept 10MHz because in their white papers they answered a 'question' whether the CC1 could improve the clock in a studio where the house clock was a 10MHz. I had assumed that meant the CC1 could be connected to the 10MHz, but that doesn't have to be the case.
So probaby it doesn't have a 10MHz signal input, and therefore this case has closed (quite firmly :D )
 
Aug 16, 2016 at 2:21 AM Post #1,494 of 3,694
  Here is an interesting alternative.  Only $149
 
It's OXCO but one that takes a clock signal from orbiting GPS satellites to act as a disciplining ref rate.  Bet those satellites have some pretty fine atomic clocks.
 
Note the 10MHz output and GPS antenna hookup.  Ext DC power as well.
 
http://www.ebay.com/itm/GPS-Disciplined-Oscillator-Clock-GPSDO-10M-Output-Sinusoidal-Wave/251802969291?_trksid=p2047675.c100005.m1851&_trkparms=aid%3D222007%26algo%3DSIC.MBE%26ao%3D1%26asc%3D20131003132420%26meid%3D2f69a02dab9545d8a0afc8e5ce2da852%26pid%3D100005%26rk%3D1%26rkt%3D6%26sd%3D281649576001


Meh. Phase noise is probably bogus compared to the Cybershaft units.

I was checking out the user reviews page on google translate. Check out this review - they said the Premium is better than the GPSDO.
 
Edit: Looks like they are comparing to this http://www.trimble.com/timing/thunderbolt-e.aspx and not the GPSDO offered by Cybershaft. Their datasheet shows -115dB/Hz @ 10Hz.
 
Now what is still useful is that they seemed to prefer the holdover (non-GPS) mode on the Thunderbolt GPSDO, so it might be the case that the Cybershaft Premium sounds better than the GPSDO? I don't see any phase noise specs listed for the GPSDO model.
 
 https://translate.google.com/translate?hl=en&sl=ja&u=http://www.cybershaft.jp/a-products/a-products-gps.html&prev=search - 3rd review from the top
 
This is the experience you have changed to cyber shaft, Inc. Premium a 10MHz clock from T Inc. Thunderbolt GPSDO. About three years ago, I was reminded of the importance of the clock in the introduction of the Thunderbolt DigitalAudio playback. To pursue the possibility of Digital playback starting with its experience, it has continued its efforts even now. Clock to receive correction from GPS was thought to be the most accurate to date. It is not a mistake at the correct point. But this time, try listening to Premium, necessarily it was now that the thought-provoking about what is appropriate (preferred or not) in the Audio. I have to teach various things over the phone conversation with Mr. Hasegawa of cyber shaft company. The necessary clock in the performance of the in Audio, it can be understood as a theory, which is a short-term stability, also was able to realize it by listening to the sound.
On the recommendation of Mr. Hasegawa was the next attempt. Is a comparison audition at the Thunderbolt GPS Lock state and HOLDOVER mode. Because to say that compared to the GPS Lock state are listening also another three years on a daily basis, you do not have it, such as the signal-type switching. Listening in HOLDOVER mode only if you feel it. The result, although significant changes, such as it can be said in the words did not feel, taste the atmosphere, such as set foot in one step a quiet forest, I remember feeling as calm. HOLDOVER mode Good !! As a result, the correction of the GPS of every moment is not necessarily required for the Audio, is better not to apply a correction to the contrary has led to think that it might than settle. OCXO short-term stability is excellent from this it is thought than preferable returned to the Audio, we decided to try the Premium.
userp03.jpg

A result, this attempt was not a mistake. Now Premium has gained the main seat, GPSDO has become a spare parts. I am listening to the 95% classic. Equipment configuration, NAS (QNAP TS-121, SSD) → WSmartHub → SFORZATO DST-01 (Digital Strem Transport) → Accuphase → Mytek Digital Stereo 192-DSD (DAC) → Accuphase A-47 (AMP) → B & W802D, clock system (10MHz) → Antelope OCX → DST-01 and Mytec

 
Aug 16, 2016 at 10:17 AM Post #1,496 of 3,694
  Quick update
 
I took out the Liveclock and tried a direct connection from Cybershaft into Mutec again. Same outcome -  still prefer sq with Cybershaft going into liveclock first. Could be as simple as by going into Liveclock first I can synch both RedNet and Mutec to same clock. Music just seems to have more presence and energy. Direct connection to Mutec does not sound bad by any means but it feels slightly more subdued.
 
Also, if Antelope owners out there do get a 10M clock, don't jump the gun on the calibration process. When I connect the Cybershaft and LIveclock and hit "calibration" the initial error is around.22Hz or so. Don't do anything for about about 2-3 minutes. The error rate slowly decreases and finally seems to stabilize for me around .005Hz. I then hit the "calibrate" button again to compensate for the difference to get to 0Hz. The first go around I did not provide enough time for the clocks to calibrate on their own. Live and learn.


Great feedback thanks.  Yes in the Live Clock manual they say to allow for time for the clock sync - what a great feature.
 
So you are using the Live Clock to fed Wclock into the REDNET and the Mutec as SPDIF/AES reclocker right.  Then the Cybershaft (that name gets me chuckling every time) just to the 10M input on the Live Clock.
 
Aug 16, 2016 at 10:19 AM Post #1,497 of 3,694
Now you come to mention it, it does have a clock input but it appears to be only a word clock?!
The manual doesn't mention 10MHz signals. I had assumed (never assume anything
biggrin.gif
) that it would accept 10MHz because in their white papers they answered a 'question' whether the CC1 could improve the clock in a studio where the house clock was a 10MHz. I had assumed that meant the CC1 could be connected to the 10MHz, but that doesn't have to be the case.
So probaby it doesn't have a 10MHz signal input, and therefore this case has closed (quite firmly
biggrin.gif
)


Your wallet is quite happy about that!  Great clock in any regards - their jitter management is sota.
 
Aug 16, 2016 at 10:23 AM Post #1,498 of 3,694
Any solutions for Linux platforms, NAS and micro PC included?


I have not yet heared any plans for Linux support on the Dante protocol. You could ask Audinate directly, they reply pretty fast.

On the Ravenna protocol there seems to be an imminent release of also a Linux based virtual sound card {source: forum from Mivera Audio}. I don't know if this will be sold only with Mivera or other Ravenna products or wether it might become available as stand alone app. This new app will be AES67 comoatible, so you could use it to communicate with RedNet devices once they have been upgraded to AES67 too. I don't know how far Focusrite is with releasing AES67 compatibility upgrades, but Audinate has already released the required updates for the Brooklyn firmware.

Availability of a virtual sound card for a NAS platform seems weird to me, as the NAS-es don't have playback software but only DLNA-server software and that is useless without an additional DLNA renderer, in the context of AOIP.

Cheers
 
Aug 16, 2016 at 10:56 AM Post #1,500 of 3,694
Well, after gotten assurance from Focusrite Tech Support that the 16 R and the 16 AES are more or less the same, I jumped the Rednet wagon (and the Mutec MC-3 + USB) :)
 
"Apart from the front panel design, the redundant power supply and ethernet there are no notable differences between the D16 and D16R"
 
"Yes, both the D16 and D16R use the Brooklyn II module and the RedNet 3 does use a Brooklyn I module. The only difference for the end user between Brooklyn I and II is the ability to assign a static IP address to the device and in the future, AES67 compatibility. Both devices are purely digital so there is no sonic difference between them. 

I would not anticipate any issues using either the D16 or D16R in a home audio setup, using a mac or pc and Dante Virtual Soundcard. If you did choose the D16R you would only need to have one ethernet cable and one power cable connected to use the device."
 
Got it working after downloading DVS, Dante Controller, Rednet Controller (do I need this?) and Java (pre 2015) in about 30 minuttes. But shouldnt there be a valid license for DVS included? Nothing in the box so I use the Trial download for now.
 
Initially the sound is very, very good! I feel I am missing a little inner detail but the "flow" of music is fantastic!  The sound is very close  to my USB chain (Berkeley usb/spdif converter, Regen, Curious Cables) after 10 min. Looking forward getting som hours and playtime on it. And more to come later on..... (the Mutec) :)
 
 
Tommy
 

Users who are viewing this thread

Back
Top