Bricasti M1 > M5 > M12: Network Player - Member REVIEW - w/OEM Response from Brian Zolner President Bricasti

[matthias]

2 hours ago, One and a half said:

I would think the former would be worse since USB is untreated, and no mention of galvanic isolation on the output for USB for the M5.

The AES3 output would be better, but no DSD256 direct.

 

Certainly the Ethernet input of M1 is best, but according to this review

http://www.hifi-advice.com/blog/review/digital-reviews/network-player-reviews/bricasti-m5-network-player/

the USB connection between M5 and M1 is better than AES.

 

Matt

[One and a half]

5 hours ago, matthias said:

 

Certainly the Ethernet input of M1 is best, but according to this review

http://www.hifi-advice.com/blog/review/digital-reviews/network-player-reviews/bricasti-m5-network-player/

the USB connection between M5 and M1 is better than AES.

 

Matt

I have the deepest respect for the writings of Christiaan Punter and I enjoy his reviews very much. As opposed to the founder of CA, Hi fi Advice does comparisons there and then, and that's valuable, since people have an idea where the reference line is.

 

On the M5 review specifically, he says USB shines, AES3 and coax still have excellent attributes (see quote below). I wonder if overall liveliness means an increase in jitter, dunno, it's symptomatic. This is the problem of reviews and the manufacturer being mum on what's inside to make a valid comparison with another component. A clean 5V rail is not enough, there's a computer in there, on Linux, USB cabling also has an effect on sources and components.

 

The main take away from the review was the redundancy of the likes of Aurender with Roon and that a dedicated computer can still hold its own. 

 

"Connected via USB the Bricasti definitely sounds best to my ears. This way it retains all of the organic and full bodied sound, large soundstage and superbly sonorous bass that the coax output provided, but gains more speed and attack as well as more overall liveliness. So far the latter was one of the main dividers between the (Antipodes) DS and M5, but with USB the two come closer together."

 

 

[Ran]

I prefer AES to USB on the M5 by a tiny margin.

[matthias]

1 hour ago, One and a half said:

I have the deepest respect for the writings of Christiaan Punter and I enjoy his reviews very much. As opposed to the founder of CA, Hi fi Advice does comparisons there and then, and that's valuable, since people have an idea where the reference line is.

 

On the M5 review specifically, he says USB shines, AES3 and coax still have excellent attributes (see quote below). I wonder if overall liveliness means an increase in jitter, dunno, it's symptomatic. This is the problem of reviews and the manufacturer being mum on what's inside to make a valid comparison with another component. A clean 5V rail is not enough, there's a computer in there, on Linux, USB cabling also has an effect on sources and components.

 

The main take away from the review was the redundancy of the likes of Aurender with Roon and that a dedicated computer can still hold its own. 

 

"Connected via USB the Bricasti definitely sounds best to my ears. This way it retains all of the organic and full bodied sound, large soundstage and superbly sonorous bass that the coax output provided, but gains more speed and attack as well as more overall liveliness. So far the latter was one of the main dividers between the (Antipodes) DS and M5, but with USB the two come closer together."

 

 

 

I do neither like the detour via USB nor via AES or SPDIF. 

In my view the best Ethernet input has Linn Klimax Katalyst, maybe the second best is the Bricasti. Ethernet will be the future, not USB. But to design a great Ethernet interface is much more difficult than to design a great USB one. So for USB-DACs the M5 is a good solution. I hope that good manufacturers like Sonore will offer Ethernet input boards that will communicate directly to DSP/DAC without the detour via USB.

 

Matt

[One and a half]

2 minutes ago, matthias said:

 

I do neither like the detour via USB nor via AES or SPDIF. 

In my view the best Ethernet input has Linn Klimax Katalyst, maybe the second best is the Bricasti. Ethernet will be the future, not USB. But to design a great Ethernet interface is much more difficult than to design an great USB one. So for USB-DACs the M5 is a good solution. I hope that good manufacturers like Sonore will offer Ethernet input boards that will communicate directly to DSP/DAC without the detour via USB.

 

Matt

Ethernet is the way of the future, you sound like a politician, some vague promise of glory, that's usually unattainable!  What about now, wait it out? 

 

Due to the cost of research, USB input to DACs as well as AES3 and Coax will remain for many years. A renderer (is) needs a computer, which needs a program to run and process Ethernet Packets and output to something in the same box that doesn't cause interference. So far, an XMOS type chipset for Ethernet is not on this planet. Processing by way of a computer needs to stay out of a DAC, just look at the lengths Lumin and Linn take to keep that processing away from the DAC, single billeted aluminium, heaven knows how that cool all that gear....and all that machining costs a bomb.

 

Designing a rendering interface is completely different to a DAC, these days, the DAC market is certainly crowded, a lot of research is necessary to produce a product to sell now to make it stand out from the rest of the pack.

 

Two boxes, the diversion you put it are necessary. OK, you could add a fibre optic transmission layer, I'm sure the audio would agree on one protocol, not. It will be a while to wait, or for now just cough up the cash for the Linn or gamble on an Bricasti M1, Lynx Hilo on Dante (no Roon!).

[barrows]

The facts are that both USB and Ethernet inputs require a "computer".  The Async USB protocol indeed requires a bit less processing power than Ethernet, but nowhere near the processing power of a full blown laptop, or even tablet or phone (these days).

Either input can be implemented to be transparent; essentially, either input can be "perfect", but many USB DACs still do not have the best USB implementations, and most Ethernet input DACs do not have the best Ethernet implementations.

A really good Ethernet Renderer (I am familiar with the tech of Sonore here, so that is my benchmark) with USB output, into a DAC with a well implemented USB inout will be at NO disadvantage to a direct Ethernet DAC performance wise-but this depends on the implementation of each device being well done.  As implementation matters so much, everyone with different set ups is going to report different results, and one would be mistaken to draw any conclusions from reports using components different from theirs.

 

The only inherent advantages of Ethernet over USB, is that Ethernet allows for much longer cabling with no problems, and Ethernet should have no limit on sample rate (even DSD 1024 or 2048 should be possible) capabilities (some USB inputs are limited).

 

 

[matthias]

12 minutes ago, barrows said:

A really good Ethernet Renderer (I am familiar with the tech of Sonore here, so that is my benchmark) with USB output, into a DAC with a well implemented USB inout will be at NO disadvantage to a direct Ethernet DAC performance wise-but this depends on the implementation of each device being well done. 

 

I can not imagine that a Linn Klimax Katalyst would perform excellently with the detour via USB. And Brian Zolner admits that the direct Ethernet input into the M1 or M12 is superior to the detour via the M5 and USB or AES. So the Sonore approach via USB is a solution for USB-DACs to get network capability. But the best way is to have a direct Ethernet input into a well designed Ethernet board inside a DAC. Period.

 

Matt

 

[barrows]

5 minutes ago, matthias said:

I can not imagine that a Linn Klimax Katalyst would perform excellently with the detour via USB. And Brian Zolner admits that the direct Ethernet input into the M1 or M12 is superior to the detour via the M5 and USB or AES. So the Sonore approach via USB is a solution for USB-DACs to get network capability. But the best way is to have a direct Ethernet input into a well designed Ethernet board inside a DAC. Period.

Matt, I am not here to discuss Sonore gear.  

 

But you are incorrect, while I have already stated that I think the future will be direct Ethernet connected DACs, and as far as I am concerned he Klimax DS is the best of those, there is no performance disadvantage in using a properly implemented Ethernet renderer with USB output onto a properly implemented USB DAC.  If using USB results in any disadvantage, then there is an implementation problem somewhere causing that disadvantage.  USB inputs, when properly implemented, are transparent, and not subject to analog like "problems", and so are Ethernet inputs.

This is why it is so important to not suggest that anyone's results will be repeatable in different systems, as every implementation is different.

[matthias]

10 minutes ago, barrows said:

Matt, I am not here to discuss Sonore gear.  

 

But you are incorrect, while I have already stated that I think the future will be direct Ethernet connected DACs, and as far as I am concerned he Klimax DS is the best of those, there is no performance disadvantage in using a properly implemented Ethernet renderer with USB output onto a properly implemented USB DAC.  If using USB results in any disadvantage, then there is an implementation problem somewhere causing that disadvantage.  USB inputs, when properly implemented, are transparent, and not subject to analog like "problems", and so are Ethernet inputs.

This is why it is so important to not suggest that anyone's results will be repeatable in different systems, as every implementation is different.

 

OK, maybe,

BUT Bricasti use the same board in the M1, M5 and M12. So it is all about the most direct pathway. And I am sure if Sonore design an suitable Ethernet board like Linn or Bricasti it will be better than the existing Sonore solutions. USB is a compromise for the existing USB-DACs.

 

Matt

[fsmithjack]

1 hour ago, One and a half said:

Ethernet is the way of the future, you sound like a politician, some vague promise of glory, that's usually unattainable!  What about now, wait it out? 

 

Due to the cost of research, USB input to DACs as well as AES3 and Coax will remain for many years. A renderer (is) needs a computer, which needs a program to run and process Ethernet Packets and output to something in the same box that doesn't cause interference. So far, an XMOS type chipset for Ethernet is not on this planet. Processing by way of a computer needs to stay out of a DAC, just look at the lengths Lumin and Linn take to keep that processing away from the DAC, single billeted aluminium, heaven knows how that cool all that gear....and all that machining costs a bomb.

 

Designing a rendering interface is completely different to a DAC, these days, the DAC market is certainly crowded, a lot of research is necessary to produce a product to sell now to make it stand out from the rest of the pack.

 

Two boxes, the diversion you put it are necessary. OK, you could add a fibre optic transmission layer, I'm sure the audio would agree on one protocol, not. It will be a while to wait, or for now just cough up the cash for the Linn or gamble on an Bricasti M1, Lynx Hilo on Dante (no Roon!).

 

Gamble on the M1? How do you come up with that? 

 

The M1 is no gamble. It’s one of safest bets in all of high end Audio. The M1 is not just a Network Player it happens to be a pretty decent DAC as well.

 

Their Network Player is absolutely fantastic. If you follow digital audio at all then you will know that Ethernet is absolutely the future. 

 

He was just stating a point that is pretty simple and blatantly obvious to everyone. Ethernet is what every single OEM is either developing or improving right now. All of them.

 

Yes Linn was first in and they make some stunningly good sounding products no doubt. 

 

USB is fine. I loved it for years but who cares if someone makes something better then great I’m in. Out with the old and in with the new. This is Computer Audio. It changes daily. Bring on the new products and approaches.

 

Some people feel the need to defend their USB products. That’s fine, there are some seriously good sounding USB gear out there but I just want the best sound however I get it. USB or no USB no problem.

 

USB had its time in the sun but those days are fading briskly! 

 

Uptone Audio is working on a great Ethernet switch because Alex knows where everything is headed! I bet it’s going to be awesome! 

 

Everything will be Ethernet based. It just fits. No more physical media. Cloud based and storage based streaming via Ethernet and at some point WIFI will be good enough as well but hard wire still sounds best today but it will come. 

 

These are not promises of granger by a forum member but rather simple industry observations by a knowledgeable CA member. 

 

 

[matthias]

@barrows

When USB is transparent, why do we have ongoing discussions about USB cables, USB regenerators, USB reclockers, LAN Rovers etc. between USB output and USB input?

The debate about a single USB cable like the Lush from Phasure fills a thread of 97 pages!

With an Ethernet input into a great Ethernet board inside a DAC you get rid of this with one stroke.

You can stream from a Mac Mini or MBP with Audirvana to the Ethernet DAC and it is done.

 

Matt

[barrows]

52 minutes ago, matthias said:

USB is a compromise for the existing USB-DACs

The above is an incorrect assumption.  SOME USB DACs have a compromised USB interface, not all.  And inherently, there is nothing "wrong" with USB as an audio transmission protocol.  Take a look at the "massively improve" thread if you want to see what some people are doing to "fix" all of Ethernet's so called "problem"!  Audiophiles will always find a way to "improve" upon any interface, whether it has  any problems or not...

A properly implemented USB interface is transparent, and at matters not.  If a dAC sounds better via another interface, then its USB implementation is not as good as it should be.

 

I have tested this extensively, using the same DAC, and switching in and out a direct Ethernet interface, and a direct USB interface.  In my case the USB interface outperformed Ethernet.  Why?  Because the USB interface was essentially perfect, and the Ethernet interface was not.  With this DAC, using an external Ethernet Renderer-USB-DAC was the best solution for performance.

 

The clear issue here is one of implementation, and not which interface os better, both can be "perfect" when done properly.

[barrows]

18 minutes ago, matthias said:

@barrows

When USB is transparent, why do we have ongoing discussions about USB cables, USB regenerators, USB reclockers, LAN Rovers etc. between USB output and USB input?

The debate about a single USB cable like the Lush from Phasure fills a thread of 97 pages!

With an Ethernet input into a great Ethernet board inside a DAC you get rid of this with one stroke.

You can stream from a Mac Mini or MBP with Audirvana to the Ethernet DAC and it is done.

 

Matt

All the same discussions are happening about Ethernet as well.  And SPDIF, etc.  Audiophiles will always find a way to "improve" things, whether real or imagined.

Cables always matter, regardless of data format, or digital vs. analog.  Sometimes they may matter more than others, but they will always matter.

[matthias]

21 minutes ago, barrows said:

A properly implemented USB interface is transparent, and at matters not.  If a dAC sounds better via another interface, then its USB implementation is not as good as it should be.

 

OK, then M5 > USB > M1 or M12 is on par to Ethernet > M1 or M12.

Ask Brian Zolner what he prefers. 

Maybe the Bricasti USB implementation is not as good as it should be.

The other question is which USB cable to chose between M5 and M1 or M12.

 

Matt

[matthias]

1 minute ago, matthias said:

 

Double post, sorry.

[barrows]

1 minute ago, matthias said:

 

OK, then M5 > USB > M1 or M12 is on par to Ethernet > M1 or M12.

Ask Brian Zolner what he prefers. 

Maybe the Bricasti USB implementation is not as good as it should be.

The other question is what USB cable to chose between M5 and M1 or M12.

 

Matt

Sure, what this tells me is that the Bricasti Ethernet interface is better implemented than their USB interface, not that Ethernet is better than USB in general.

[audio.bill]

43 minutes ago, matthias said:

@barrows

When USB is transparent, why do we have ongoing discussions about USB cables, USB regenerators, USB reclockers, LAN Rovers etc. between USB output and USB input?

The debate about a single USB cable like the Lush from Phasure fills a thread of 97 pages!

With an Ethernet input into a great Ethernet board inside a DAC you get rid of this with one stroke.

You can stream from a Mac Mini or MBP with Audirvana to the Ethernet DAC and it is done.

 

Matt

The debate over the impact of various Ethernet cables continues among users, along with the claimed benefits of various Ethernet isolators, optical conversion and its related cables, enhanced low noise switches, etc. Hardly eliminating the add-ons being used with USB, rather replacing them with others unique to Ethernet. The quest for perfection will move ahead in whatever medium or format you choose.

[matthias]

24 minutes ago, barrows said:

Sure, what this tells me is that the Bricasti Ethernet interface is better implemented than their USB interface, not that Ethernet is better than USB in general.

 

I prefer a direct Ethernet input like Linn and Bricasti to a detour via USB.

Further I prefer Wifi to wired Ethernet.

 

Matt

 

[fsmithjack]

On 3/15/2018 at 5:07 AM, One and a half said:

@Miska, @fsmithjack says the improvement in sound with the Linear Switch and Audioquest Vodka cables improved, by a notable difference. Take that as is & no need to be dismissive of the Linear switch right away, let's try and figure out what's going on.

 

Audioquest's technobabble makes it difficult to provide any meaningful and tangible evidence as to what the construction of the cable is. It maybe conforming to Cat7 standards, since it boasts speeds for Ethernet audio that it can handle, but high speed Ethernet audio is not really necessary. In any case, Cat7 requires individual pairs of the cable are screened, plus the cable has an overall shield. To mitigate common mode noise, filters can be included in either the Linear Switch or the Vodka Cable, to also reduce the effects of differential noise as well.

 

So, the suggestion of using a shielded Ethernet cable, was to avoid EMI escaping the plain vanilla UTP cable, and to provide a bond between the Ethernet switch and the M1 DAC. Galvanic Isolation wrecking? When the transformers inside RJ45 jacks still have inter-winding capacitance to allow differential noise to jump easily at undesired frequencies. Unless of course the RJ45s have filters built in, but no one says anything about them, or deflect the truth with some secret sauce like 'Noise-Dissipation System'.

 

Here's an extract out of an App note from Texas Instruments:

 

 

From the Appnote:

Recommendations for limiting chassis coupled noise include:
• Use a shielded connector on the network interface PCB. The shielded connector should be connected
to a PCB chassis ground plane that is decoupled from the PCB system ground.
• Connect the chassis ground plane to the system ground plane using size 1206 zero ohm resistors
symmetrically placed on either side of the RJ45 connector. These resistors can be removed or
replaced with alternative components (i.e. capacitors or EMI beads) if necessary during certification
testing. See Figure 3.
• Use common mode choke transmission components in the network interface PCB design. These
devices are commonly available in discrete form, integrated into network transformers, and integrated
with transformers in network connectors.
• If possible, the use of shielded cable can reduce emission levels by 6 to 10 dBµV or more.

 

@fsmithjack, how far away is the Ethernet switch from the M1?

Common Mode Choke Filtering Laird Technologies -App-Note-1.pdf

Texas Instruments AN-1862 Reducing Radiated Emissions in Ethernet 10-100 snla107a.pdf

 

Hi sorry - just noticed this question. It’s about 2 feet away? Is that ok? Thanks 

[barrows]

1 hour ago, matthias said:

I prefer a direct Ethernet input like Linn and Bricasti to a detour via USB.

Further I prefer Wifi to wired Ethernet.

 

You are of course welcome to prefer whatever you like, but that does not change the reality of the technical concerns.  In an ideal world, with lots of choices of great DACs with exceptional Ethernet interfaces, i would prefer a direct Ethernet input on my DAC, unfortunately we are not in that world (yet).

[ChrisG]

5 hours ago, One and a half said:

Processing by way of a computer needs to stay out of a DAC, just look at the lengths Lumin and Linn take to keep that processing away from the DAC, single billeted aluminium, heaven knows how that cool all that gear....and all that machining costs a bomb.

I don't think the casework on Lumin or Linn has anything to do with keeping the PC away from the DAC - vibrations maybe, PC noise interference, I don't think so.

One of the big advantages of Ethernet vs USB is you can run very long runs of cable, so you don't have to have the PC anywhere near the DAC. Can't say the same with USB since, without powered repeaters, the cable length can't be much longer than 15 meters.

Back to the casework - I've owned both Akurate and Klimax streamers and currently own a Klimax DSM and, while it is true that Klimax (which is housed in billeted aluminum) is much better than the one tier down Akrurate (which isn't housed in billeted aluminum), I don't think the SQ difference has much to do with the casework. 

[ChrisG]

3 hours ago, matthias said:

 

OK, maybe,

BUT Bricasti use the same board in the M1, M5 and M12. So it is all about the most direct pathway. And I am sure if Sonore design an suitable Ethernet board like Linn or Bricasti it will be better than the existing Sonore solutions. USB is a compromise for the existing USB-DACs.

 

Matt

Not sure what you're asking for here...even if Sonore did design what you've suggested, somehow it would still have to fit inside your USB DAC. Are you suggesting that they become a supplier for other manufacturers?

[matthias]

5 hours ago, barrows said:

In an ideal world, with lots of choices of great DACs with exceptional Ethernet interfaces, i would prefer a direct Ethernet input on my DAC, unfortunately we are not in that world (yet).

 

Agree.

 

Matt

[matthias]

4 hours ago, ChrisG said:

Are you suggesting that they become a supplier for other manufacturers?

 

Yes.

 

Matt

[One and a half]

7 hours ago, ChrisG said:

I don't think the casework on Lumin or Linn has anything to do with keeping the PC away from the DAC - vibrations maybe, PC noise interference, I don't think so.

One of the big advantages of Ethernet vs USB is you can run very long runs of cable, so you don't have to have the PC anywhere near the DAC. Can't say the same with USB since, without powered repeaters, the cable length can't be much longer than 15 meters.

Back to the casework - I've owned both Akurate and Klimax streamers and currently own a Klimax DSM and, while it is true that Klimax (which is housed in billeted aluminum) is much better than the one tier down Akrurate (which isn't housed in billeted aluminum), I don't think the SQ difference has much to do with the casework. 

Processors give off EMI, even the heatsink acts as an antenna. This noise couples to the PCB and ground planes, therefore layout and materials are important to avoid coupling. Please refer to the pdf in the post from Sun.

 

Why did you sell the Linn?

An Overview of Chip Level EMC Problems - IEEE Santa Clara Valley ...032010Radu.pdf