Introducing The Sonore Signature Rendu SE

@barrows, leaving apart the cost and complexities, would a fiber port on the SRse isolate the ethernet side better ? I use a FMC with microRendu and find the benefits are substantial.

Dev, I am of the "simple is better" approach in most things.  Of course, simple is not always better, but at least I start there.  Typically, in networking, fiber is only called for in very long runs.  While it offers the benefit of galvanic isolation, Ethernet is already galvanically isolated by transformers at each end (as long as one is not using metal shell connectors with grounded shields on their Ethernet cabling which can introduce a non-isolated ground connection).  Additionally, the process of going from an electrical signal to an optical signal and back is not lossless, doing this will increase the amount of errors and the the number of re-sends of data (re-sends of course increase processing and therefore noise generation), so there are trade-offs.

Because of these technical reasons I find it surprising when audiophiles find using fiber for relatively short runs results in performance increase, but I would never dispute your own findings.  For the technical reasons above we would do not use optical in our products, but of course you are free to experiment with it and use whatever you find to work best in your system.

I'll be keeping a close eye on this product. Bravo...for deciding to implement a proper PS within the same box. I cant tell you how tired I am of seeing/needing/hearing about all these external LPS boxes being hung off the back of "insert component here" in order to get the most out of the gear its powering.

Bob buys a Fiber Media Converter to isolate his system from the grunge on the ETH line. Bob decides he now needs to power the FMC with an LPS. Bob then buys a few more LPS's to power his music server his NAS his NAA and his USB Decrapifier too! Then and only then can Bob sit back and enjoys the show.

Blaaah

10 hours ago, barrows said:

Dev, I am of the "simple is better" approach in most things.  Of course, simple is not always better, but at least I start there.  Typically, in networking, fiber is only called for in very long runs.  While it offers the benefit of galvanic isolation, Ethernet is already galvanically isolated by transformers at each end (as long as one is not using metal shell connectors with grounded shields on their Ethernet cabling which can introduce a non-isolated ground connection).  Additionally, the process of going from an electrical signal to an optical signal and back is not lossless, doing this will increase the amount of errors and the the number of re-sends of data (re-sends of course increase processing and therefore noise generation), so there are trade-offs.

Because of these technical reasons I find it surprising when audiophiles find using fiber for relatively short runs results in performance increase, but I would never dispute your own findings.  For the technical reasons above we would do not use optical in our products, but of course you are free to experiment with it and use whatever you find to work best in your system.

Barrows, I am also of the opinion of keeping things as simple as possible and only put things wherever it absolutely matters for audio. Somehow the fmc seems to belong to that category. I am very well aware of the galvanic isolation that the transformers provide in the ethernet PHY but it seems like somehow the noise makes through - could be the type of Ethernet cable used ? I use a BJC Cat6a but if you have some other suggestion, I would like to hear. In other forums, I have read AQ Diamond (an uber expensive for an ethernet cable) makes a big difference but I haven't tried.

On the additionally processing note you mention - if you talking about protocol re-transmission, like TCP, then given the network bandwidth required for audio, I hardly suspects there is re-transmission involved from media conversion. They are supposed to do the conversion at line rate, which is very large compared to what is needed for audio.

DEV, I am loathe to try Ethernet cables costing thousands of dollars, so I have not.  If I was going to try some audiophile Ethernet cables, I would start with Nordost and Wireworld.  I need about 25' from upstairs room where Router and NAS are, and then another 1 M cable between router and NAS.  From a technical standpoint, the only way I can see anything other than a reasonable quality Ethernet actually mattering (and BJC CAT 6A is quite nice, with Belden bonded pairs) is if it produces less errors/re-sends-a cable cannot stop noise on the data lines... and differential signaling should cancel most noise at the transformers anyway.

The only "real" solution to reduce noise on Ethernet I can think of would be a purpose built "audiophile router", but this is likely a long way off.  I did find that a better power supply for my router made a difference, go figure...

A lot of this stuff is annoying!  Noise form Ethernet transmission has a long way to go, through many stages of isolation, for it to get to where it can cause harm (at the DAC or potentially the output side of the USB interface): transformers at both ends of Ethernet transmissions and then through the isolation at the USB interface.

I have been meaning to get some audiophile Ethernet cable samples for evaluation, so we might make better informed recommendations to customers, but with all the hardware development I do (and that is where much bigger gains are made) I have not had time-maybe things will calm down a bit now with the Signature Rendu SE out and I can try some Ethernet cabling...

57 minutes ago, barrows said:

 

I have been meaning to get some audiophile Ethernet cable samples for evaluation, so we might make better informed recommendations to customers, but with all the hardware development I do (and that is where much bigger gains are made) I have not had time-maybe things will calm down a bit now with the Signature Rendu SE out and I can try some Ethernet cabling...

 

Barrows, please share your thoughts when you get a chance

14 hours ago, barrows said:

DEV, I am loathe to try Ethernet cables costing thousands of dollars, so I have not.  If I was going to try some audiophile Ethernet cables, I would start with Nordost and Wireworld.  I need about 25' from upstairs room where Router and NAS are, and then another 1 M cable between router and NAS.  From a technical standpoint, the only way I can see anything other than a reasonable quality Ethernet actually mattering (and BJC CAT 6A is quite nice, with Belden bonded pairs) is if it produces less errors/re-sends-a cable cannot stop noise on the data lines... and differential signaling should cancel most noise at the transformers anyway.

The only "real" solution to reduce noise on Ethernet I can think of would be a purpose built "audiophile router", but this is likely a long way off.  I did find that a better power supply for my router made a difference, go figure...

 

A lot of this stuff is annoying!  Noise form Ethernet transmission has a long way to go, through many stages of isolation, for it to get to where it can cause harm (at the DAC or potentially the output side of the USB interface): transformers at both ends of Ethernet transmissions and then through the isolation at the USB interface.

 

I have been meaning to get some audiophile Ethernet cable samples for evaluation, so we might make better informed recommendations to customers, but with all the hardware development I do (and that is where much bigger gains are made) I have not had time-maybe things will calm down a bit now with the Signature Rendu SE out and I can try some Ethernet cabling...

Went through the Ethernet testing bit of late.  What I did find is more to do with mixing cable designs is a no-no.  Tested Audioquest and WireWorld.  Each is using a very different approach to how the wires are laid out and the philosophy around the design.  Previously I'd been using Audioquest Vodka for all Ethernet connections.  Net: I have been happy with WireWorld Cat8 for NAS to switch and MacMini to switch with WireWorld Platinum from switch to dCS Network Bridge.  The overall improvement was in lowering the noise floor and providing a wider, taller and deeper soundstage along with more focus/pinpoint sonic images in space.  I'll say that might personal preference in sound is ultra revealing, so I hear warts and all.  For those who prefer a more tubelike, vinyl analogue experience go with Audioquest.  

Being an IT guy by trade I cant get past the absurdity of an Audiophile Ethernet Cable. I guess this is similar to the many EE's on the web feeling the same way about the slu of other Audiophile cables on offer.

Anyway, I wont disagree that there are differences to be heard by swapping ETH cables as long as the swapper does realize that what they are hearing is NOT what the original Artist intended and may not even pass a CRC Checksum test between point A & B compared to an ETH cable designed to spec found in any store selling such a product.

But with that said, as long as you like what you here, that's all that matters

4 hours ago, cjf said:

Being an IT guy by trade I cant get past the absurdity of an Audiophile Ethernet Cable. I guess this is similar to the many EE's on the web feeling the same way about the slu of other Audiophile cables on offer.

 

Anyway, I wont disagree that there are differences to be heard by swapping ETH cables as long as the swapper does realize that what they are hearing is NOT what the original Artist intended and may not even pass a CRC Checksum test between point A & B compared to an ETH cable designed to spec found in any store selling such a product.

 

But with that said, as long as you like what you here, that's all that matters

From what I've seen, most ethernet cables sold in retail stores (hanging from a rack in hardshell plastic packaging) for a few bucks, don't actually pass spec when tested. They obviously "work", but don't actually meet spec. 

So that would be one reason to buy even ethernet cables from a company like bluejeanscable, who custom make each order and test each cable, and send you the results. More expensive than your commonly found cable, but still modestly priced and clearly better made than most "off the rack" cables. 

On 2017-07-21 at 9:35 AM, barrows said:

Dev, I am of the "simple is better" approach in most things.  Of course, simple is not always better, but at least I start there.  Typically, in networking, fiber is only called for in very long runs.  While it offers the benefit of galvanic isolation, Ethernet is already galvanically isolated by transformers at each end (as long as one is not using metal shell connectors with grounded shields on their Ethernet cabling which can introduce a non-isolated ground connection).  Additionally, the process of going from an electrical signal to an optical signal and back is not lossless, doing this will increase the amount of errors and the the number of re-sends of data (re-sends of course increase processing and therefore noise generation), so there are trade-offs.

Because of these technical reasons I find it surprising when audiophiles find using fiber for relatively short runs results in performance increase, but I would never dispute your own findings.  For the technical reasons above we would do not use optical in our products, but of course you are free to experiment with it and use whatever you find to work best in your system.

I am finding FMCs, with the downstream FMC powered by an LPS-1, lower the noise into the microRendu and thus into the DAC, where the noise alters the D/A results. The downstream FMC is not ideal in that it has switching regulators and introduces noise, although with the right power it drops more noise from the ethernet line than it introduces.  I think before you declare the use of FMCs an idea for end users to play with you should build an experiment with optical ethernet input and make a decision based on results obtained.

When I installed the FMCs (tried both 1Gbps and 100mbps) the first thing I did was connect a computer and do FTP transfers.  I got the same results with wired or fibre and was satisfied with both.

20 minutes ago, d_elm said:

I am finding FMCs, with the downstream FMC powered by an LPS-1, lower the noise into the microRendu and thus into the DAC, where the noise alters the D/A results. The downstream FMC is not ideal in that it has switching regulators and introduces noise, although with the right power it drops more noise from the ethernet line than it introduces.  I think before you declare the use of FMCs an idea for end users to play with you should build an experiment with optical ethernet input and make a decision based on results ontained.

 

When I installed the FMCs (tried both 1Gbps and 100mbps) the first thing I did was connect a computer and do FTP transfers.  I got the same results with wired or fibre and was satisfied with both.

You say that the use of the FMCs "lower(ed) the noise into the microRendu", I am curious how you actually determined that there was lowered noise?  Was there audible noise somewhere in your system that went away?  Or did you do spectral analysis somewhere?

What exactly was the improvement which you experienced?  I am trying to be open minded here, of course.

3 minutes ago, barrows said:

You say that the use of the FMCs "lower(ed) the noise into the microRendu", I am curious how you actually determined that there was lowered noise?  Was there audible noise somewhere in your system that went away?  Or did you do spectral analysis somewhere?

What exactly was the improvement which you experienced?  I am trying to be open minded here, of course.

I do not have test equipment for measuring noise.  I can only use my ears and listen to music.  A microRendu 1.4 board or an ultraRendu are both about lowering the noise sent to the DAC.  So are FMCs.  You do not have to agree with me and others but before you start selling the next latest and greatest you should at least experiment with an optical ethernet input.

12 minutes ago, d_elm said:

I do not have test equipment for measuring noise.  I can only use my ears and listen to music.  A microRendu 1.4 board or an ultraRendu are both about lowering the noise sent to the DAC.  So are FMCs.  You do not have to agree with me and others but before you start selling the next latest and greatest you should at least experiment with an optical ethernet input.

I asked how you determined that the FMCs "lower(ed) the noise into the microRendu".  It appears that you assumed this, based on hearing some change in the sound produced through the loudspeakers.  There is nothing wrong with that, I am just asking to know what that change was which led to believe it was due to "lower" noise into the microRendu: what change did you hear in your system?

I am skeptical, as technically speaking there is very little to suggest that such an approach would make any difference, and there are also technical reasons not to use two conversions from electrical signal to optical, but as mentioned before I am open minded.

17 minutes ago, barrows said:

I asked how you determined that the FMCs "lower(ed) the noise into the microRendu".  It appears that you assumed this, based on hearing some change in the sound produced through the loudspeakers.  There is nothing wrong with that, I am just asking to know what that change was which led to believe it was due to "lower" noise into the microRendu: what change did you hear in your system?

I am skeptical, as technically speaking there is very little to suggest that such an approach would make any difference, and there are also technical reasons not to use two conversions from electrical signal to optical, but as mentioned before I am open minded.

I have not heard the 1.4 board or ultraRendu in my system but others are pleased with the SQ improvement.  Same for me, in the same way, with FMC at 100mbps downstream powered by LPS-1.  It is my conclusion that all three products, or product improvments in Sonore's case, are about reducing noise that goes over the USB interface to the DAC.

6 minutes ago, d_elm said:

I have not heard the 1.4 board or ultraRendu in my system but others are pleased with the SQ improvement.  Same for me, in the same way, with FMC at 100mbps downstream powered by LPS-1.  It is my conclusion that all three products, or product improvments in Sonore's case, are about reducing noise that goes over the USB interface to the DAC.

There is actually a lot more to it than that  There is a lot of strange things going on, and we have found many things which improve the sonic performance.

So, it could be that with an Ultra FMCs could offer no improvement.  I am not saying this so, but it could be.  at some point you get to where there is no "problem" to solve. 

28 minutes ago, barrows said:

There is actually a lot more to it than that  There is a lot of strange things going on, and we have found many things which improve the sonic performance.

So, it could be that with an Ultra FMCs could offer no improvement.  I am not saying this so, but it could be.  at some point you get to where there is no "problem" to solve. 

Could that be 'no USB' and optical ethernet input ?  But then, that would have to done by a DAC.

There is nothing wrong with USB when does right, in both sender and receiver.

But, I do believe that the future will be direct ethernet input DACs.  Right now it appears that only Linn and Merging are doing that very well, other DACs with Ethernet inputs appear to be compromised on their Ethernet inputs (probably using poor Ethernet input boards from relatively affordable sources, 192 only, etc...)

It is a little tricky to put Ethernet inside the DAC, one has to be very careful about noise isolation as Ethernet processing can produce a lot of noises nd this needs to be well isolated from the DAC/analog sections.  Linn in the Klimax handles this very well, at a cost of cost.  For now, a separate Ethernet streamer is a good way to go to avoid that noise source inside the DAC.

37 minutes ago, barrows said:

There is nothing wrong with USB when does right, in both sender and receiver.

But, I do believe that the future will be direct ethernet input DACs.  Right now it appears that only Linn and Merging are doing that very well, other DACs with Ethernet inputs appear to be compromised on their Ethernet inputs (probably using poor Ethernet input boards from relatively affordable sources, 192 only, etc...)

 

It is a little tricky to put Ethernet inside the DAC, one has to be very careful about noise isolation as Ethernet processing can produce a lot of noises nd this needs to be well isolated from the DAC/analog sections.  Linn in the Klimax handles this very well, at a cost of cost.  For now, a separate Ethernet streamer is a good way to go to avoid that noise source inside the DAC.

I saw that Ayre partnered with Sonore at Munich 2017.  I hope that leads to something good for Ayre's QX-5 ethernet support.

12 hours ago, barrows said:

There is nothing wrong with USB when does right, in both sender and receiver.

But, I do believe that the future will be direct ethernet input DACs.  Right now it appears that only Linn and Merging are doing that very well, other DACs with Ethernet inputs appear to be compromised on their Ethernet inputs (probably using poor Ethernet input boards from relatively affordable sources, 192 only, etc...)

 

PS Audio DS Sr w/ Bridge II and DS Jr are both Ethernet capable - with DLNA, Tidal, Roon and now MQA support. Ethernet in and I2S out in the inside.

I really appreciate these posts that are "to the point" because it seems to me there is a fine balance between mentally accepting what "should" be happening and what actually is happening (subjective as it is) when the signal finally comes out of the speaker.

It's clear Ethernet has isolation designed into it, but why do cables make a sonic difference? Maybe the noise that undermines the final audio signal operates at different frequencies? Maybe it's just cable design? Or something else? Yes, I'd like to know why as much as the next person but at the same time, I'm not an electrical engineer and the complex interdependencies cannot really be appreciated by me despite a conceptual knowledge of such.  

So, thanks @barrows (and @Superdad and @PeterSt and @JohnSwenson and others here) for making the effort to be open-minded while holding the conventional wisdom lightly. It is truly rewarding to hear what's cooking in audio today, and reproduced music has never sounded so great, so conveniently.

One really cool thing about much of this stuff, is that you can try it for free. No manufacturer likes to take things back, but many of them offer free in-home trials. What more could we want? 

With respect to Ethernet cables, I'm sure there are dealers or manufacturers who will offer free in-home demos. Plus, this stuff is easy to ship. We aren't talking about 200 lbs. amplifiers. 

On 7/23/2017 at 11:27 PM, Dev said:

 

PS Audio DS Sr w/ Bridge II and DS Jr are both Ethernet capable - with DLNA, Tidal, Roon and now MQA support. Ethernet in and I2S out in the inside.

I am aware there are many DACs with Ethernet inputs, my point was that there are very DACs with really well implemented Ethernet inputs.  It is not a simple matter to incorporate an Ethernet receiver inside the same chassis as a DAC, as the Ethernet receiver creates a lot of noise, and must be very well isolated from the rest of the circuitry.  Additionally, a lot of DACs incorporate ethernet receivers of limited capability (like only to 192 kHz, or DSD1, etc).

6 hours ago, The Computer Audiophile said:

One really cool thing about much of this stuff, is that you can try it for free. No manufacturer likes to take things back, but many of them offer free in-home trials. What more could we want? 

 

With respect to Ethernet cables, I'm sure there are dealers or manufacturers who will offer free in-home demos. Plus, this stuff is easy to ship. We aren't talking about 200 lbs. amplifiers. 

 

 

Thanks Chris! This is nice to know as I've tried a few cables in my system (Supra, Blue Jeans, and "generic") and I can clearly hear a difference, but beyond my current set up (Blue Jeans) I will be well into diminishing returns territory. I'd rather spend that on a new DAC...

I posed the Ethernet cable question as an example of the fine balance discussed here at CA, partly because it's supposedly isolated by definition, and because critical listeners are posting in the "massively improve" thread about their shift toward USB from Ethernet (not necessarily "away from" Ethernet). As @barrows points out, fewer boxes/power supplies is probably desirable, and just having Ethernet on chassis in a DAC is one thing; doing it right is altogether another.

Y'all are really helping me minimize the dent in my bank account! My kid's college fund thanks you, too!

I must express my disappointment after investing about 2200. only 10 months ago fore the loaded Signature LPS and mR only to see one product discontinued and the other replaced. Good to hear that Sonore may be offering those like myself an upgrade path from the Signature LPS to the new all in one solution. Good marketing and will aid in customer retention.

8 hours ago, rjc1845 said:

I must express my disappointment after investing about 2200. only 10 months ago fore the loaded Signature LPS and mR only to see one product discontinued and the other replaced. Good to hear that Sonore may be offering those like myself an upgrade path from the Signature LPS to the new all in one solution. Good marketing and will aid in customer retention.

The mR wasn't replaced; still being sold.

Sonore has a series of units in price from $299 to $2999 that use the SO  operating system and functionally do the same thing. Each buyer can choose how far up the audiophile ladder to go. 

I don't see how this is any different than what speaker manufacturers do as a matter of course: come out with a standmount speaker, and then a succession of larger floorstanders in the same "line". 

No one objects to that or thinks their speaker has been "replaced"  when a larger, more expensive, more capable model is released. Why do people get bent out of shape when Sonore does the same thing with the Rendu series?