dCS Network Bridge

[AmusedToD]

This is what the NB unit status reports:

 

 

Not sure whether the lock status refers to the clock inputs (in case of using an external master clock) or the clock output. But in any case, it shows “not locked”. 

[AmusedToD]

6 minutes ago, skatbelt said:

 

And continue with speaker placement... 

Unfortunately not possible at this point in time, but once we move to a new place - absolutely! 

[AmusedToD]

1 hour ago, mecani said:

I do not understand

 

Deleted

[AmusedToD]

I have some new findings I would like to share with you.

 

Since getting my new speaker and interconnect cables last week (Transparent Audio Super Gen5 bi wire and Transparent Audio Super Gen5 balanced XLR), I went through a rough week of burning them in, and I almost lost my nerves in the process. The sound I was getting from the system was playing tricks with my mind, one day sounding too dark, the next day sounding flat and lifeless, then too bright, then compressed etc. Absolutely mind boggling. Overall, I could say it was decent, but nowhere near the level I intended to have and wished for, especially having in mind the value of the components.

 

However, a “miracle” happened today. A digital AES/EBU cable I ordered almost a month ago was delivered today (Transparent Reference) to complement my Transparent Premium AES/EBU (lower in line than the Reference), so I could finally test the dual AES mode between the Network Bridge and the Paganini.

 

So the setup looks like this:

 

iMac (Roon core) -> Supra Cat8 Ethernet cable -> SOtM ISO-CAT6 passive isolator -> SOtM CAT6 Ethernet cable -> dCS Network Bridge -> dual AES (Transparent Reference & Transparent Premium AES/EBU cables) and silver BNC clock cable by Oyaide -> dCS Paganini DAC with its clock slaved to the Network Bridge  -> Transparent Super Gen5 balanced XLR (brand new, still burning in) -> Simaudio Moon I-7 RS integrated amp -> Transparent Super Gen5 bi wire speaker cable (brand new, still burning in) -> TAD ME1 speakers.

 

Up to this point in time I used a single Transparent Premium AES/EBU cable for connecting the NB to the DAC. After receiving the Reference cable, I put it to the test by playing a few tracks (Marvin Gaye - “What’s going on” in 24/192 from his 1971 album “What’s going on”; Oscar Peterson Trio - “You look good to me” in 16/44.1 from the 1964 album “We get requests”; Dave Grusin - “Keep your eyes on the sparrow” from his 1976 album “Discovered again”, etc.). The Transparent Reference AES/EBU proved to be a much more serious cable compared to the Premium (as it should be being more than double the price), giving a much more relaxed sound with a bigger soundstage, more precise imaging, deeper and better defined bass and a buttery smooth midrange. Wow! Very happy at this point.

 

Having established it’s a worthy upgrade, I went on to finally test the dual AES mode. I played the above mentioned 16/44.1 track by Oscar Peterson Trio - You look good to me. What surprised me was the sample rate on the Paganini - it displayed 16/88.2 instead of 16/44.1. I thought ok - it must be the settings, but I let it play anyway. The first thing I noticed was the completely incorrect positioning of the instruments in space! The acoustic bass was not playing on the far right but in the center, and there was a sense of extreme reverberation. What the hell!? Is this how the dual AES should sound? This isn’t right at all!

 

After confirming all the settings in the NB and the Paganini were right, I fired up google and started searching, and then stomped onto this on the dCS website:

 

Dual AES is a digital audio interface invented by dCS in 1995 to allow high sample rate data to be split between two base-rate data streams for easy storage on the digital recorders of the time. For example, a stereo pair of 96kHz data could be split into two 48kHz streams and recorded as 4 tracks on a 48kS/s recorder. To replay, the 4 tracks are transmitted over two AES3 cables to the DAC, which reconstructs the original 96kS/s data.

 

Dual AES was offered as an open standard and it was added to the AES3 specification. Almost 20 years later, we use Dual AES at sample rates as high as 384kS/s. Dual AES features lower jitter than Single AES or S/PDIF, so there is a sonic benefit.

 

The raw data must be specially formatted as a Dual AES pair at the transmitter and then decoded at the receiver. Two identical Single AES streams used together will be decoded by the DAC as a strange ‘phasey’ mono – this is not the same as Dual AES at twice the sample rate.

 

What the hell? Phasey mono? That’s exactly what I heard. No trace of the “lower jitter playback with a sonic benefit”.

 

But then I thought - perhaps the NB can’t split a 16/44.1 file into “two base rate data streams”. So I fired up Roon again and set it to upsampling mode to the maximum 24/384 rate that both the NB and the Paganini could read.

 

Guess what? Bam! Holy moly, I was absolutely stunned by the sound quality I received from my gear within the first second of playback! The soundstage exploded in all 3 dimensions, with glass like clarity and superb imaging/layering and texture. It was so good that I was left speechless. What the hell happened??? 

 

So now, after a whole hour of experimenting, I am still amazed by the improvement I got from the dual AES mode. My system finally sounds “expensive”! The benefit from going dual AES far exceeded the benefit I got from upgrading the SOtM sMS-200 to SOtM sMS-200 Ultra (my previous streamers).

 

P.S. Unfortunately upsampling in Roon is necessary for it to work properly, so no bit perfect streaming in this scenario. I don’t know why this is the case, I thought it would work as in a dCS CD Transport - dCS DAC setup, but obviously it doesn’t. The files need to be upsampled in the audio player for the Network Bridge to properly split the rates into 2 streams and extract the maximum sound quality with the lowest possible jitter. And it’s a huge sonic upgrade indeed!

 

So this matter requires further investigation as well.

[AmusedToD]

So just to confirm my findings: 44.1, 48, 96, 176.4,  192 and DSD64 sample rates do now work properly through the dual AES outputs on the NB. However, when upsampled to DXD (352.4 and 384) - dual AES works magically. I still can’t say whether this is due to the settings (I tried all possible combinations) or the Paganini DAC.

[AmusedToD]

Update to the dual AES saga. I finally managed to get it working without upsampling hi res files (only redbook and 24/48 need upsampling in Roon).

 

This is what the manual says:

 

 

So first I reduced the dual AES sampling rate to the minimum possible in the Network Bridge settings, which is 88.2.

 

 

Anything lower than 24/88.2 will not work right over dual AES, so I had to set the upsampling options in Roon to  “custom”. As you can see from the picture below, only 44.1 and 48 are upsampled (because they lower than the minimum dual AES sample rate in the NB settings), all other sample rates starting from 88.2 and higher up to 384 (including DSD64) are played bit perfect!

 

 

I hope this will be helpful for others who are using the NB with a dCS DAC capable of dual AES connection. 

 

 

[AmusedToD]

23 minutes ago, KunterK said:

After testing in my system I "upgraded" to dCS Network Bridge from Aurender N10. Since dCS is playing from the Network or from an attached USB drive,  not from an internal disk,  one should be careful with the surrounding items (network, data source etc)  when doing the comparison. Both were connected to a TotalDAC D1-six (which is now up for sale, to be replaced by a Vivaldi DAC) via transparent XL AES. Proper network feed performs better than attached USB) 

When driven from roon (linear powered NUC i5) the sound is sublime with the NB on all dimensions. 

To my ears it even challenges W20 quite nicely.

 

Cheers

 

Congrats on getting the Vivaldi! I heard the full stack recently and was blown away by it.

 

You should definitely try the dual AES, the improvement over the single AES is simply stunning! Especially since you already have one of the best digital cables in the world (Transparent Reference XL). 

[AmusedToD]

24 minutes ago, KunterK said:

I did try it hence the change. Dual Aes and clock from NB to Vivaldi with roon upsampling was very bad for the wallet. 

 

I will use Shunyata sigma for Aes and clock though. Will report performance once I have them. 

 

Xl trio is too expensive. 

 

Please report how the Shunyata’s fare against the Reference XL. They really are darn expensive. I am using a regular Reference plus Premium for dual aes until a good deal for a used Reference aes emerges.

[AmusedToD]

My mainboard version is 1.0, and I bought a brand new unit last month. The network board version is 324.

[AmusedToD]

This is what Mr. Martin Reynolds of dCS replied to my question regarding the NB firmware update:

 

”We hope to release the MQA update for Rossini by the end of the year, then we will roll MQA out to NBR and Vivaldi over the following months. I cannot be more precise, as some of the delays are beyond our control. The Network Bridge will then perform the first unfold on bit-perfect MQA data, which will result in good-quality audio at 24/88.2 or 24/96. Unfortunately, the Paganini DAC is no longer in production, so we cannot update it to perform the final MQA rendering, but it will give good results.“

[AmusedToD]

3 minutes ago, rickca said:

Could you please ask him about availability of USB audio output on the NB?  That's more important to me than MQA.

 

Well, it took them a week to reply to my question which was sent through their website. You could drop them an email the same way: https://www.dcsltd.co.uk/product-support/contact/

 

[AmusedToD]

10 minutes ago, rickca said:

Done.

 

So according to the update you got from Martin Reynolds, the MQA schedule has already slipped from the timeframe Darko mentioned only 3 weeks ago.  Clearly Darko's input was stale by the time he published his article.

 

Yes, apparently so. According to Martin’s email, we shouldn’t expect any updates this year.

[AmusedToD]

25 minutes ago, firedog said:

I wrote and asked them what the issue was with installing RoonBridge (Roon streaming app) on the Onset Streamer and they answered me right away and said this:

 

 

 

So according to them, the Stream board (made by Stream Unlimited in Vienna) is not good sounding, thus dCS made the wrong decision of using this module in the Rossini DAC and the Network Bridge. Everybody hurry and dump your Rossini's and NB's in favor of the Onset Streamer! And forget Roon along the way too!

[AmusedToD]

1 minute ago, rickca said:

I guess I don't blame them on MQA support because MQA has to sign off on the implementation and the MQA process doesn't sound exactly streamlined.

 

I'm hoping they can do a better job telling us when the USB output will be available.  And I really hope the timeline isn't linked to the MQA stuff.

 

Perhaps the USB out update will not be linked to the MQA update, in which case we might get it sooner.

 

Personally, I don’t care about the USB output as my DAC does not have an USB input. On the other hand, I doubt it will be sonically superior to the dual AES connection, but this is strictly when used with dCS DACs which are designed to sound best over dual AES.

 

I think enabling the USB output on the NB is for the purpose of making the product more accessible to the general market, including people that use good quality non-dCS DACs that have superior USB inputs (Chord Dave comes to mind).

[AmusedToD]

46 minutes ago, Chewy560 said:

Are there any users out there who are using NB with Chord Dave. If so have you tried to interface the dual AES output from NB to DAVE? DAVE manual states; "Dual Data Mode - Dave supports dual data mode on its BNC inputs. This is where a 176kHz data rate is supported by sending a left 88kHz signal and a right 88kHz signal via separate BNCs." Therefore I think it might be possible with appropriate connectors.

 

 

Interesting. You could try these adapters from Neutrik: https://www.thomann.de/gb/neutrik_naditbncfx_aesebu_bncadapter.htm

 

Then set the dual AES mode in the NB settings to 176.4 khz and see if it works.

[AmusedToD]

5 minutes ago, Chewy560 said:

 

Excellent - those connectors would appear to be suitable.

Unfortunately I am only contemplating the DAVE at the moment, however if the above works it would be a big +. (the alternative dCS options are too expensive for me at the moment).

 

 

In my experience the dual AES mode is a significant step up from any other possible digital connection between the NB and DAC (single AES or regular rca spdif), so you should definitely have it in mind when choosing your DAC. 

[AmusedToD]

31 minutes ago, Cyrus said:

@AmusedToD  Unfortunately, I arrived at the same conclusion (based on reading/researching) on the best use case; i.e. using the dual AES out was really the best choice.  That, and the uncertainty about the USB output, it's compatibility, and the quality vs AES.

 

Unfortunate, since I was (still am) very interested in the dCS NB but I wasn't able to play at the price point I needed to, for a dual AES DAC. 

 

Well, you can always get a used dCS DAC, they are built like tanks. I had an older Elgar Plus (15 years old) for audition in my system before purchasing the Paganini, and it blew away everything I ever used before. I am not even sure if the Paganini is a real upgrade to the Elgar Plus (apart from being prettier in its shiny silver chassis and for having the ability to play 24/384 and DSD64). 

[AmusedToD]

Did anyone notice the delay when playing DSD files over Roon, it started yesterday after the new Roon Core update. It takes approx 6-7 seconds for a DSD song to start when played through Roon into the NB. I restarted everything several times after the update, but nothing helped. The same happens if I try to skip a DSD file, the long delays simply happen every time with DSD.

[AmusedToD]

36 minutes ago, The Computer Audiophile said:

I had breakfast with dCS a few hours ago. I’ll ask them this afternoon about this. 

 

Good man! Please ask them if they have a specific date for the MQA update too

 

Thanks!

[AmusedToD]

41 minutes ago, beessy said:

Yes same issue for me

 

Mainboard v1.0

 

Mainboard v1.0 on mine too. 

 

Did it start after the Roon update?

[AmusedToD]

13 minutes ago, beessy said:

Yes it seems. It is like a 6s-7s resync delay is applied on all dsd track (even set at 0ms)

 

Exactly!

 

12 minutes ago, The Computer Audiophile said:

dCS is talking to Roon now. 

 

Hope they sort it out soon enough, it's pretty annoying for people who have large DSD collections.

[AmusedToD]

Here is a great suggestion for those who like to watch YouTube music videos, concerts, etc.

 

You get an Apple TV 4K and set the audio output to dCS Network Bridge via AirPlay and voila! The best sound you will ever get from YouTube  Especially if the NB is connected through dual AES and a clock cable to a dCS DAC!

 

This goes for all streaming services (Netflix and the likes) and everything else you can play through the Apple TV.

 

[AmusedToD]

10 hours ago, beessy said:

 

Issue seems to be fixed with roon build 276.

 

Yes, they did fix it, DSD is back to normal with 276. Great job Roon!

[AmusedToD]

23 minutes ago, beessy said:

i still have a strange thing since 274, when i use 'rewind button' on a playing track, the sound makes a little pop/scratch and then the track is played at beginning.

 

Did you restart both the server and the NB after the last update?

[AmusedToD]

6 minutes ago, nkbg said:

 

Many audio video sync issues with this?

 

None at all.