EtherREGEN: The long development thread. [Some Gen2 dev. pics and update starting on page 92.]

[kennyb123]

Can the ability to eliminate phase-noise fingerprints also provide benefit at the point where USB enters a DAC?  I’m thinking about the clocks inside some of these music servers and renderers.  Would their phase-noise get fingerprinted onto the USB signal?  Music servers sound different, even with an ISO REGEN in use.  Might that be part of the reason why?

 

I have a feeling the EtherRegen might actually make server differences easier to hear, as it will reduce the harm all our network gadgets cause and thus give us clearer insight into the server.  I could be wrong though.  Can’t wait to find out!

[kennyb123]

Are we there yet, (super)dad?  😐

[kennyb123]

4 hours ago, Superdad said:

 

Setting aside that we are avoiding the term "clean side" (as the design is rather symmetrical), potential cable differences will be lessened on the upstream side of the isolation moat.  That is opposite of the downstream side you are calling the "clean side."  

The point being that what comes before the isolation--and final clocking "out" to the DAC-connected device--is what will matter much less.  But the cable from that--typically the lone 'B"-side port--will be the one that still matters.  Hence we suggest keeping the EtherREGEN close to the DAC-connected Ethernet renderer (or Ethernet DAC) and using a short, high-quality cable for that connection.

 

It makes perfect sense to avoid the term “clean side” when viewed from the perspective of the EtherREGEN.  It’s agnostic as you’ve described. I think it’s going to be tougher for the rest of us to abandon the terms “clean side” and “dirty side” when we view things from the perspective of our networks. On one side of the EtherREGEN will be simply my streamer.  On the other side will be the rest of my network, which has all kinds of noise being dumped onto it.  The goal of adding to EtherREGEN is to create a “clean side” (of my network) that’s isolated from the noisy “dirty side” (of my network).   

 

I think as long as folks are clear that it is their network that has the dirty/clean sides - and not specifically the EtherREGEN - that those terms can be helpful.  

[kennyb123]

3 hours ago, Superdad said:

Lest anyone think the above is a "simple" switch:

You are looking at a boatload of innovative engineering and techniques never previously applied to an Ethernet switch.

And the board you see, in 250 unit runs, will cost us about $250 each--a full $200 of which are parts on the top of the board. Those 10GHz-capable ultra-low-jitter flip-flops, 600MHz LVDS isolators, clock synth, and pile of best-of-breed regulators don't come cheap!

 

Impressive Alex - thanks for sharing.

 

But are we there yet, dad?

 

(ducking)

[kennyb123]

2 hours ago, stevebythebay said:

 

If anyone doesn’t think their choice of Ethernet cables is inconsequential,  just try some different designs in your system setup.

 

I finally got my hands on a couple of Shunyata’s, a new cable design. It ran rings around my existing WireWorld Platinum and Starlight 8 cables, in/out of the Cisco 2960 switch.  Needless to say, I’ve put a couple of these on order. 

 

Really looking forward to the EtherREGEN to see how much farther my system can go.

 

Are you referring to the Sigma or Alpha?  I will be purchasing one of these, but am waiting first for the EtherRegen, as I will use it between this and my Zenith.  Like you, I’m looking forward to seeing how much further these two moves will take my system.  

[kennyb123]

27 minutes ago, stevebythebay said:

What I've found is that you really don't want to mix differing technologies -- the WireWorld is Cat8 (shielded) untwisted flat design with the Shunyata Cat6a unshielded type with a traditional type configuration.  They seem to work against each other, or at least diminish the intended results.  That's what I discovered when initially swapping just the Sigma for the Platinum.  Better but nothing like adding the Alpha into the mix.

 

Interesting finding.  I have a long run of Wireworld Chroma from wall jack to my Zenith presently.  The plan was to run that into the EtherRegen and then have an Alpha or Sigma feeding the Zenith.  Hopefully the EtherRegen does succeed in allowing the cable on one side of the moat to have no audible impact on the other side of the moat.

 

Your comments about the Sigma have me rethinking the order of my moves.  The high demand for the EtherRegen may make it tough to obtain one once it’s finally available.  

 

[kennyb123]

On 8/8/2019 at 8:07 PM, stevebythebay said:

Used both the Sigma and Alpha in my config.  

 

Looking at the specs Shunyata has published for their Ethernet cables, it appears the only difference is the number of CMode modules.  Venom has 0, Alpha has 1 and Sigma has 2.  The EtherRegen should lessen (or eliminate?) the need for any common mode filtering.  It’ll be interesting to hear if the differences between their three Ethernet cables vanish when used with the EtherRegen.

[kennyb123]

2 hours ago, stevebythebay said:

But the proof will be in the “pudding”.

 

True that.

 

I was just struck by the fact that this thread was started in March of 2018.  Hopefully not much longer.

[kennyb123]

5 hours ago, marce said:

The Cmode filters are ferrite rings.

 

Where did you get that information?

[kennyb123]

6 hours ago, vortecjr said:

However, the silly thread will say you need two ERs in series each with external reference clocks in order to get good sound. Also, for good measure the silly thread will say you need two linear power supplies that you have to wait over a year to get or it just won't sound quite right. I'm just saying... 

 

Why misrepresent what they are pursing in that thread?  It should be obvious to anyone that they are hoping to go way beyond getting "good sound" and are looking for the best possible sound they can achieve.  

 

Aren't you also trying to do the same with the products you sell?  Why even advance beyond the microRendu if all that matters is getting "good sound"?   

[kennyb123]

4 hours ago, Superdad said:

 

As promised. Arguments about ABX and cables gone, along with a few other random OT posts. It was a fun diversion while it lasted. 

 

I very much appreciate the great job you do at keeping us on point here.  

[kennyb123]

4 minutes ago, Puma Cat said:

Even within the strict confines of their "schtick", they don't know how to conduct a scientifically controlled experiment in the context of "real-world" use, and furthermore, they've not done a formal MSA (measurement systems analysis) prior to the measurements they take. Without an MSA in place before taking measurements, there is no way to know that their measurement system has sufficient precision to observe real differences from noise (all data sets contain noise, some contain signals) and make statistically valid inferences, if there are differences. For example, they don't know what their smallest effective measurement increment is, they don't know what the "% Contribution" of their MS is, and they don't know anything about the "part" (i.e., DAC) x Operator interaction. 

 

So very well said.  Over the years I had tried to make the same point about their measurement having sufficient precision.  I couldn't explain it as precisely as you did.

 

Sorry for going off topic here Alex.  I just thought the above was an excellent point.

 

[kennyb123]

@Superdad thank you so much for getting the first batch shipped out so far ahead of the promised date.  I’m loving what I’m hearing so far!  @JohnSwensonand you have another winner here.

 

My initial impressions:

 

 

 

[kennyb123]

On 9/9/2021 at 11:18 AM, PYP said:

Personally, I don't think that optical is a magic solution.  Just like everything else, it depends upon implementation.  

True that.

[kennyb123]

13 hours ago, Tony P said:

I look forward to seeing what the Gen2 will do for my streamed content.

I hope you report your findings once you are able to obtain a Gen2.  I also plan to purchase a Gen2 but it will be used in my audio system where it will replace a Gen1.  The Gen1 will then go to my video system.  Like you, I look forward to seeing what an EtherRegen will do for my streamed video content.  It’s indispensable in my audio system.

[kennyb123]

19 hours ago, agladstone said:

I’m curious if the new EtherRegen 2 would actually be best served for streaming 4K video because of the 1GB throughput? 
Maybe I’ll need to get two of the newer ones :) 

Keeping my fingers crossed that 100 mbps is sufficient.  What’s interesting is that my LG OLED only supports up to that rate over wired - though I’ll likely run the B side into my Nvidia Shield Pro and that supports gigabit.  I suspect though that the compressed streams will be fine over 100 mbps.

[kennyb123]

On 8/28/2023 at 3:24 PM, Superdad said:

My father, at 93.5 years, passed away late Friday night.

 Very sorry for your loss.

[kennyb123]

On 8/29/2023 at 7:16 AM, The Computer Audiophile said:

I dreamed of owning Hovland gear back in the day, but could never afford it

Me too.  Beautiful stuff.