dCS Network Bridge

[mcgillroy]

On 5/3/2018 at 3:55 PM, AMP said:

 

Yes, quite a bit actually. We chose to implement based on a significant amount of customer demand.

 

 

Correct. Based on feedback from dealers and distributors as well as direct customer interactions the demand for USB audio output is comparatively low.

 

 

 

Interesting - could you give an indicator of the geographies of MQA demand? Europe, USA, Japan, Asia in general?

[stevebythebay]

3 hours ago, The Computer Audiophile said:

Disable volume?

I only see one place to increase/lower volume in the dCS app and that appears to be in the lower right corner of the player.  It's at the extreme of 0.0 db.  I'm assuming that's the only way to "disengage" the volume.  Is that the correct place?  That's the only volume reference in the dCS NB manual that I can find.

 

Did one last thing: disabled Volume Leveling.  That was the missing piece.  I can now have the dCS NM perform the first unfold and the Berkeley DAC render.  Both MQA indicators, dCS app and Berkeley LED are showing the corresponding info.

[Pedro Romão]

I will be VERY vocal on this. I was told 6 months ago by an official dCS agent that USB functionality was just around the corner. I was told the same by dCS staff on various emails. I am facing again another delay (definitive?) on the usb out enabling.

Definitely NOT what I was expecting from a company like dCS. Starting to regret this buy, unfortunately.

[MetalNuts]

On 5/3/2018 at 9:55 PM, AMP said:

 

Yes, quite a bit actually. We chose to implement based on a significant amount of customer demand.

 

 

Correct. Based on feedback from dealers and distributors as well as direct customer interactions the demand for USB audio output is comparatively low.

 

 

I don't know what other consumers demand but MQA is  surely one of the feature I ignore in deciding to buy dCS Rossini Dac and Clock.  So may be I am the minority. 

 

The reason I switched from Esoteric N-01 to dCS Rossini is that I can choose the filter I want (surely I will not activate the MQA filter) in dCS Rossini whereas the Esoteric N-01 has no such choice and I assume it will automatically activate the MQA filter when playing MQA files.  However, having regard to the test by some learned member here, not necessary MQA files can tricker the MQA filter, so I have doubt that the MQA filter is the only filter in Esoteric N-01 and is always active.

 

It is also a pity that dCS does not have a forum on its own like Linn where its users can discuss and voice out their concern.  It seems like a official dCS forum is where one can see the demand of the dCS actual user, and not someone who expect many from dCS without actually buying any.

 

 

[AMP]

6 hours ago, mcgillroy said:

Interesting - could you give an indicator of the geographies of MQA demand? Europe, USA, Japan, Asia in general?

 

Sorry, we don't share this type of information publicly. I can say that demand was fairly uniform across the board but some geographies yelled a lot louder than others.

[AMP]

4 hours ago, stevebythebay said:

Did one last thing: disabled Volume Leveling.  That was the missing piece.  I can now have the dCS NM perform the first unfold and the Berkeley DAC render.  Both MQA indicators, dCS app and Berkeley LED are showing the corresponding info.

 

We discovered an issue this morning having to do with Roon's DSP and MQA. They identified and fixed a bug after the last alpha build prior to 1.5 going public and it turns out that their bug was masking a yet unknown one of ours. We had already completed our final testing of this functional area and had no information about any subsequent changes to the code.

 

As of right now Roon's DSP functions used in conjunction with the Network Bridge will not properly trigger the MQA rendering process on a capable DAC.

 

We've identified a workaround and hope to have that implemented soon. In the mean time functions like volume leveling, parametric EQ, headroom adjustment, etc will collide with the MQA signaling information. Best to leave them off if you're looking for the full MQA experience.

 

If you use Roon's upsampling with the bridge then that function will work as intended. It will be enabled for all MQA content and disabled when MQA streams are played.

[AMP]

2 hours ago, Pedro Romão said:

I will be VERY vocal on this. I was told 6 months ago by an official dCS agent that USB functionality was just around the corner. I was told the same by dCS staff on various emails. I am facing again another delay (definitive?) on the usb out enabling.

Definitely NOT what I was expecting from a company like dCS. Starting to regret this buy, unfortunately.

 

I'm very sorry that you are frustrated, but we've encountered numerous hurdles with the USB output and until we're certain that the performance is up to par we aren't going to release the feature. We're targeting June for this, but until we've completed a full round of testing on the latest build we can't be certain of that date.

[AMP]

19 minutes ago, MetalNuts said:

It is also a pity that dCS does not have a forum on its own like Linn where its users can discuss and voice out their concern.

 

This is coming. Having learned from the MQA releases we aren't going to provide a timeline, but this is high up on my to-do list.

 

20 minutes ago, MetalNuts said:

It seems like a official dCS forum is where one can see the demand of the dCS actual user, and not someone who expect many from dCS without actually buying any.

 

What's interesting about this comment is that many of you have expressed displeasure about our choices regarding MQA. I know that some of you are dCS customers, but many who have made comments in this thread are not. So as to set the record straight we based our decision on feedback from known customers as well as prospects who were known to the dealer or distributor to be truly serious about a dCS purchase.

 

We're a small company and over the course of a year spent thousands of man-hours on MQA. We're not done yet so that number continues to go up. We would never have considered this level of investment (at no cost to the customer) if real demand wasn't present.

 

I know that a lot of you don't like MQA and I can appreciate that stance. I can also appreciate that some of you would have preferred it if another feature had been given priority over MQA. Again, I get it. The fact of the matter is that our actual customers requested this either directly or through their dealers. The pressure from customers and dealers was overwhelming at times and we couldn't ignore that.

[One and a half]

USB functionality and glitching for DSD transitions stopped purchase of the Network bridge for me.

 

Based on a sodden experience with a competitor bridge, the existing DAC wouldn’t have worked any way, so which of the three reasons put a stop to buying, doesn’t really matter, just proves the technology is in its infancy.

 

The inclusion of MQA instead of USB functionality means a decision that dcs will have to live with in the interim or even long term since buyers are looking elsewhere and won’t come back. For me MQA is displaying foxtails on car aerials, fashionable value only.

 

Needless to say, but will anyway, dcs network bridge implementation  is far away on the horizon.

[Patatorz]

7 hours ago, stevebythebay said:

I only see one place to increase/lower volume in the dCS app and that appears to be in the lower right corner of the player.  It's at the extreme of 0.0 db.  I'm assuming that's the only way to "disengage" the volume.  Is that the correct place?  That's the only volume reference in the dCS NB manual that I can find.

 

Did one last thing: disabled Volume Leveling.  That was the missing piece.  I can now have the dCS NM perform the first unfold and the Berkeley DAC render.  Both MQA indicators, dCS app and Berkeley LED are showing the corresponding info.

I think they are mentioning the volume fixed @ 100% (0 dB) in Roon configuration. By the way the dCS app should reflect this volume setup

[Patatorz]

5 hours ago, Pedro Romão said:

I will be VERY vocal on this. I was told 6 months ago by an official dCS agent that USB functionality was just around the corner. I was told the same by dCS staff on various emails. I am facing again another delay (definitive?) on the usb out enabling.

Definitely NOT what I was expecting from a company like dCS. Starting to regret this buy, unfortunately.

Same from my side. I thought my dealer would have been more vocal but it seems he was less than others on MQA. I don’t understand how dealers could be vocal on mqa when you clearly see the « dark side » of MQA (read below).

 

To be honest I have mixed feelings. On one side I’m fed up with usb announcement from officials and not seeing it coming as it was one of my request to invest in a network bridge (yes Amp I even if we are not owners you can understand that we can become owners). On the other side I see dCS moving and providing updates, investing on people and it is not what happen with other companies that I would not mention here where they release their product and after there is no follow up at all leaving the product as it is and making is lifetime becoming smaller and smaller.

i hope dCS, as other companies (Lumin, auralic, etc...) designed their product in order to be able to implement new functionalities in the coming months/years (should be the case with a fpga). My biggest fear is to invest in a network bridge and understanding within 6 months to 1 year that because of design constraints they are releasing a new one...

 

[rickca]

5 hours ago, AMP said:

The fact of the matter is that our actual customers requested this either directly or through their dealers. The pressure from customers and dealers was overwhelming at times and we couldn't ignore that.

So when the novelty wears off and people discover that MQA doesn't live up to its hype we can get back to developing things that are actually useful.

[MetalNuts]

5 hours ago, AMP said:

What's interesting about this comment is that many of you have expressed displeasure about our choices regarding MQA. I know that some of you are dCS customers, but many who have made comments in this thread are not. So as to set the record straight we based our decision on feedback from known customers as well as prospects who were known to the dealer or distributor to be truly serious about a dCS purchase.

 

I have reservation with the feedback from the dealers or distributors.  My experience tells me that most, if not all, the sales of the dealer and distributor (at least in my place) do not have much knowledge with the equipment they are selling.  The sell what they are told to sell with the most profit, most do not own reasonable HiFi equipment.  They just bluffed.  What they are selling are always the best in the market and the brands they do not have are always inferior to what they have.  Cannot blame them, it is business anyway.

[Patatorz]

I must admit that you are right. 99% of the dealers today understand nothing to computer music, streamers, streaming and are able to support their customers on this topic. I have the chance that one of mine, managing dCS, is part of the 1% 

As you i would have a lot of reluctance to take for granted all what they would bring from their customers. That's why i'm very suspicious about MQA. We don't know what MQA could become : like HD-DVD or like Blu-ray...So,as customer, i would prefer a product that brings me more options( like USB output, Qobuz streaming, Tunein integration, capacity to switch on and off USB 5V if playing through usb, world clock with same form factor as the NB,..) than in MQA. 

[mcgillroy]

6 hours ago, AMP said:

 

Sorry, we don't share this type of information publicly. I can say that demand was fairly uniform across the board but some geographies yelled a lot louder than others.

 

Interesting thank you!

[MetalNuts]

1 hour ago, Patatorz said:

I must admit that you are right. 99% of the dealers today understand nothing to computer music, streamers, streaming and are able to support their customers on this topic. I have the chance that one of mine, managing dCS, is part of the 1% 

 

Agreed, dCS is one of the few that replied to my email of enquiry.  Other brands just ignored it.

[Patatorz]

8 hours ago, AMP said:

 

We discovered an issue this morning having to do with Roon's DSP and MQA. They identified and fixed a bug after the last alpha build prior to 1.5 going public and it turns out that their bug was masking a yet unknown one of ours. We had already completed our final testing of this functional area and had no information about any subsequent changes to the code.

 

As of right now Roon's DSP functions used in conjunction with the Network Bridge will not properly trigger the MQA rendering process on a capable DAC.

 

We've identified a workaround and hope to have that implemented soon. In the mean time functions like volume leveling, parametric EQ, headroom adjustment, etc will collide with the MQA signaling information. Best to leave them off if you're looking for the full MQA experience.

 

If you use Roon's upsampling with the bridge then that function will work as intended. It will be enabled for all MQA content and disabled when MQA streams are played.

 

Roon released this morning fixing some MQA issues (https://community.roonlabs.com/t/roon-build-323-is-live/42850). Any relation with what you mention above ?

[Beolab]

With the new SW / FW + App , the app crasches occationally as usual. 

 

The sound from Tidal feels slight more native to the Master recording than before, with higher resolution ?

 

Tidal MQA 24/96 vs same album but recorded in 16/44.1 slight better timing and you get one step closer in to the layers in the music, but it is not a huge, on some tracks, but on some there is a great diffrence actually.  

 

I cant see that i can make any settings for MQA unfolding in the NB app settings, so the maximum unfolding is 24/96, i thaught that it could unfould up to 24/192 and send it out to a non MQA dac ? 

 

Can i find the all MQA Tidal Masters album in som dCS playlist like BlueSound got ? 

 

 

[stevebythebay]

On 5/8/2018 at 11:22 AM, AMP said:

 

The list is too long to post and a number of the fixes have nothing to do with the NBR.

 

The network firmware is built from a shared source tree that covers all of our products so one of the main benefits of this release is that it brings the underlying infrastructure to parity with the Rossini and Vivaldi One. This includes an update to the Roon Ready SDK (Roon endpoint code) as well as a significant number of fixes surrounding the Roon implementation. The end result is much more stable operation with Roon. In addition there have been a number of fixes to the UPnP implementation as well as native Tidal streaming.

 

At this point in time we do not publish specific bug IDs nor do we publish a detailed changelog. We track this information internally, but we do not publish it.

Well my testing confirms that some HDCD material coming from Roon still is not correctly passed through to the Berkeley DAC. Still no HDCD light on the unit. Guess passing this from Roon support to dCS support has yet to yield a fix. Maybe next time...

[The Computer Audiophile]

34 minutes ago, stevebythebay said:

Well my testing confirms that some HDCD material coming from Roon still is not correctly passed through to the Berkeley DAC. Still no HDCD light on the unit. Guess passing this from Roon support to dCS support has yet to yield a fix. Maybe next time...

That’s a strange one I’ll have to test when I get back from Munich. If the HDCD indicator isn’t illuminated then the LSB isn’t getting through properly and the music isn’t bit perfect. 

 

The usual suspect it volume attenuation somewhere in the chain. 

[AMP]

On 5/9/2018 at 6:36 PM, AMP said:

We've identified a workaround and hope to have that implemented soon. In the mean time functions like volume leveling, parametric EQ, headroom adjustment, etc will collide with the MQA signaling information. Best to leave them off if you're looking for the full MQA experience.

 

I have the updated firmware image and will be testing through the weekend. It appears to be working fine and assuming no issues we'll be releasing the update early next week.

[AMP]

13 hours ago, stevebythebay said:

Well my testing confirms that some HDCD material coming from Roon still is not correctly passed through to the Berkeley DAC

 

Is this still the one Reference Recordings album that's causing you problems or is it more this time?

[stevebythebay]

16 hours ago, The Computer Audiophile said:

That’s a strange one I’ll have to test when I get back from Munich. If the HDCD indicator isn’t illuminated then the LSB isn’t getting through properly and the music isn’t bit perfect. 

 

The usual suspect it volume attenuation somewhere in the chain. 

Just did some additional testing.  What I failed to do was revert to non-DSP and volume leveling status (as I did with my MacMini - now using a Nucleus).  Previously only one ripped CD had the issue, namely Copland 100 from Ref. Recordings. 

 

Once I disable both DSP and volume leveling, ALL of my confirmed HDCD discs now are correctly detected.  If either or both settings are enabled NONE of my CDs with HDCD are detected. 

 

So, it begs the question as to how Roon is altering the bitstream.

 

p.s. orig. Roon thread https://community.roonlabs.com/t/bit-perfect-rip-with-hdcd-not-engaging-dac-processing/30435/34

[stevebythebay]

2 hours ago, AMP said:

 

Is this still the one Reference Recordings album that's causing you problems or is it more this time?

See my response to Chris above.  At least all CD rips are consistent, though the result is not what I would have expected, unless, by design, DSP and volume leveling do something to the bitstream that, by nature, will defeat HDCD.

 

FYI: Roon site thread - https://community.roonlabs.com/t/bit-perfect-rip-with-hdcd-not-engaging-dac-processing/30435/34

[AMP]

48 minutes ago, stevebythebay said:

So, it begs the question as to how Roon is altering the bitstream.

 

45 minutes ago, stevebythebay said:

unless, by design, DSP and volume leveling do something to the bitstream that, by nature, will defeat HDCD.

 

ALL DSP functions alter the bitstream and this is the very nature of what DSP is. Any correction is going to fundamentally alter the data and from a pure numerical standpoint even a minor change (like volume leveling) is going to result in a massively different set of bits. It's going to sound similar, but the data is very different.

 

This is why you can't apply any DSP to an MQA stream before it's unfolded. If you do that you lose everything that allows the MQA stream to be reconstructed. HDCD operates in a very similar way to MQA's first unfold in terms of the information embedded in the bitstream. Any change, no matter how minor, will break HDCD.