UpTone Audio EtherREGEN Listening Impressions

[Mike Rubin]

19 minutes ago, Superdad said:

You should experiment.  I think Jud's experience has nothing to do with the EtherREGEN's SFP port.  It most likely has to do with the SMPS of his Cisco switch and its effect on his system--but not via the optical cable.

But yes, it has been confirmed (first by John Swenson in his own system, and since then by others) that the EtherREGEN goes beyond the opticalModule.  This is not a slight on the very fine opticalModule--easily the world's best FMC for audio. But the core architecture and objectives of the EtherREGEN is the reason that people are hearing and reporting these results.

Thanks, Alex.  If I can round up the scratch, I will place an order and see how this works out.

[Superdad]

44 minutes ago, stevebythebay said:

Won’t dwell on medical stuff, though I had a gout attack last Wednesday on my left big toe, I’m still recovering from…

Ouch! Hope you are feeling better.

 

Quote

Sorry for the rather lengthy post...

Are you kidding? Based on what you wrote I almost feel like just locking the thread now. What more can be said? You said it it all and said it so well.  Really happy to know that your already advanced system responded so well. Maybe dCS will want to offer EtherREGEN to go with their components. Of course they would need to put it in a fancy box and multiply the price a few times...

 

40 minutes ago, Mike Rubin said:

If I understand your configuration properly, you are connecting to your endpoint using copper on the B side RJ45. In contrast, I want to use the B side copper port upstream of the SFP.  In other words, I want the signal to go from the B side to the A, connecting my rendu end point to the SFP.  I think Jud tried that, if I understood his first post correctly, but preferred going from A to B.

No, I do not think that is what Jud did. He had two Cisco switches--and before the EtherREGEN he had them connected to each other via SFP/optical fiber, and from the second one ran just one copper cable to his microRendu.  When he got the EtherREGEN, he compared:

1) Eliminating the second Cisco Catalyst 2960, running optical to EtherREGEN's 'A'-side and copper to his microRendu; 

then

2) Eliminating the first Cisco Catalyst 2960 and just running copper to an 'A'-side EtherREGEN port and still the 'B' port to his microRendu.

He preferred #2, but other than the Cisco having an internal SMPS--maybe affecting his wall AC--we do not have an explanation.

---------

 

As for your request for reports running B>A (copper to optical, about the only reason someone would do so), I think some folks are already doing this or will shortly.

 

Remember, we spent a bunch of extra time and money to make EtherREGEN symmetrical--just so that B>A would perform equally as well as A>B--primarily for optical endpoint users.  Otherwise we would not have put the differential flip-flops on the 'A' side as well, and the second high-speed differential isolator across the moat could have be down-graded to run just a single-ended 25MHz clock line back to the 'A'-side (as opposed to both 25MHz and 250MHz differential clocks).

[Mike Rubin]

Thanks for clarifying this for me, Alex.  I will place an order.  Early post-holiday splurge for me.

[Puma Cat]

47 minutes ago, stevebythebay said:

Won’t dwell on medical stuff, though I had a gout attack last Wednesday on my left big toe, I’m still recovering from…though it did keep me on the couch listening to the EtherREGEN break-in.  And break-in it really did.  I’m glad I raised this break-in question on the forum with Alex, as it proved to be very, very true.

I was initially uncertain about how much better replacing my existing Cisco 2960 would turn out to be.  I’d done my best to mitigate all digital and analogue related noise coming from devices, and passing along to others, along with using whatever cabling and interconnects that would prevent pickup or passing of electrical or other mysterious crap.  My config. is as simple as I can make it with Roon player and Eero WiF wired to the ER while the “B” port going to dCS gear.  That’s it.

However, after plugging in the ER there was no immediate and obvious change in SQ for my Roon local music playing.  However I did notice that OTA streaming of even low quality radio, let alone FLAC music from Radio Paradise, improved nicely.  However, that’s upstairs cable modem to Eero WiFi access point,  then to another Eero co-located with the downstairs audio system and wired into the ER.  Caveat: the Shunyata Sigma Ethernet cables incorporate technology to mitigate both signal and high frequency noise distortion.  That's what's wired between the Roon, ER, and DAC components So, no immediate revelations from what the ER would come to reveal.

After 24 hours of playing music w/o the amps turned on, I found a very dramatic change when I finally got around to hitting the switches on the amps and doing any critical listening.

Replacing the Cisco 2960 with the EtherREGEN results in separating the wheat from the chaff - the overall sound level seems louder.  This must be some psycho acoustic effect.  Background is blacker, and sonic decay lingers on, along with much clearer sense of the sonic space surrounding instruments and singers.  

I agree with Jud’s perception of how the ER really started “opening up”.  Clearly lines between all elements of the sonic picture came into a much clearer separation in all areas - location in space, size, realism, etc.  Even listening to live performance the audience seemed more “real” and clearly hearing members shout out things I’d failed to pick up on in the past.

And everything Puma Cat has stated I hear as well.  The palpable nature of live playing, especially percussion in attack and decay are very evident.  But it goes way beyond that to what others have heard: fretwork on guitars, reed instruments that just seem far more realistic, and the piano is just so right and clearly displayed.  I’ve just begun listening to the same piece played by different pianists in solo just to capture their style and how the recordings were miked and engineered.  Already very revealing.

By keeping things simple in my chain the ER can reveal just how great it is as more than an ideal switch.

Will be fascinating to read John’s technical brief when it arrives.  Hoping he can help provide the laymen among us with what the actual effect that results from the many wiz bang elements he’s put into the ER.  

As for all of us who are sharing, I agree with Puma Cat re: the diverse nature of our systems.  Best we can hope for is just a subjective reaction to before/after the addition of the ER, making one simple change.  Too many changes and you’re uncertain what really affected what.

Always feel when I’ve made a positive change in my whole listening environment the result is greater “ease” in listening.  Simply great pleasure and pulls me into the performance rather than showing off.  On the other hand the ER can’t make a silk purse out of a sow’s ear: bad recordings will stick out like a sore thumb that much more.  I’ve already experienced that with numerous albums in my collection.  But then I focus on the performance…

Examples of listening sessions:

Sgt. Peppers - uncluttered and quiet with Ringo’s drumming unpacked; all of the micro details exposed.

Jimi Hendrix - Electric Ladyland - phase experiments revealed; Voodoo Child - Jimi’s guitar fully realized; Gypsy Eyes shows off black background and allows exposure of all micro details; all around spatial expansion width/depth/height and precise size and placement of instruments & vocalists; Voodoo Child (slight return) even better showing off many of these attributes).

Sorry for the rather lengthy post...

 

Good post! My personal findings, listening to a lot of full-scale classical symphonic content last night are very concordant with yours, Steve.  👍

 

Agree with Steve that ER starts to show its stuff with ~24 hours burn-in. Sunday night was a notably different and improved experience over Saturday night. If my hypothesis holds, by ~72 hours of burn-in, should be "up on the cam, and hitting on all cylinders", to use an aphorism from motor racing. 

[Puma Cat]

48 minutes ago, octaviars said:

 

Well it sounds really good already so if the etherREGEN can improve it by just a little bit I am happy with that.

 

I hope Alex will ship it in the coming days so perhaps early next week I will have it in my system.

I'll bet it already sounds really good;  you've got a superb system there. Love the TAD gear and dCS digital components. 

[cat6man]

2 hours ago, Superdad said:

I know that there are several opticalModule/opticalRendu owners who have received--or are about too--an EtherREGEN. We believe that best performance will be obtained by directly attaching the EtherREGEN's 'A'-side SFP port to the optialRendu. But I am sure people with all these devices will report in due time. 

 

can't wait (but will have to.............)

 

congrats Superdad on a great introduction!

[Puma Cat]

4 minutes ago, charlesphoto said:

 

I would be curious what the opticalModule into etherRegen sounds like. 

Go back and read my comments/impressions on page 2 (IIRC)....

[k-man]

2 hours ago, Superdad said:

Maybe dCS will want to offer EtherREGEN to go with their components. Of course they would need to put it in a fancy box and multiply the price a few times...

Here’s a thought! How about designing a DAC streamer utilising the best parts of the etherREGEN and ISO REGEN? No fancy box, <$3000-$4000. dCS killer 

[incus]

Just picked up my etherRegen and eagerly attached it to my opticalRendu.  Powered by an LPS1.2.  No grounding yet.  A-side Ethernet via SOtM cable from a SOtM switch, B-side optical direct to oRendu.  Nothing else attached to the eRegen.  No external reclocking yet.  Signal then goes into a tX-USB and on into a Hugo TT2.  Already great sound.  At very very early first blush I noticed a slight lowering of the noise floor, inkier background, a bit more depth, as if I can better perceive the directionality/three dimensionality of the soundwaves being picked up by the mics. These are changes I like!  Details, transients, dynamics, tonality -- these were already excellent in my system so no difference I can hear yet.  I have three hours on the unit so these are super crazy early impressions.  But I especially like the increased depth.  A hard thing to achieve.  A metric I use when measuring the accuracy of clocks.  To me personally I have always felt that phase noise manifests itself as a loss in this 3D/directional information.  The removal of electrical noise for me manifests as a more relaxed presentation - less extraneous information vying for limited brain bandwidth.  Will report back later with updated impressions after more burn-in time and with the unit in different configurations - used on the server side instead, re-clocked with my SOtM clock, grounded in different ways, different Ethernet cable, etc.  Great start, tho.  Cheers

[Superdad]

4 minutes ago, incus said:

A-side Ethernet via SOtM cable from a SOtM switch, B-side optical direct to oRendu.

 

Hi Incus:

Thanks for posting.  The above is confusing and I just want to be sure that you are crossing the moat.

 

The EtherREGEN's SFP cage is on the 'A' side--with the four Gigabit RJ45 ports.

The 'B' side is the single 100Mbps RJ45 copper port, right next to the BNC jack.

 

So for use directly to an opticalRendu, you want to be running B>A.  So the copper cable from your other switch should be plugged into the port labeled "'B' PORT OUTPUT," and the fiber optic cable will be run from the SFP port on the 'A' side to your opticalRendu.

[incus]

Sorry got A and B side mixed up above.  Configuration is copper ethernet into the B side with optical out of the A-side.

[incus]

And I should add a few details - SOtM switch powered by Farad Super3 12V, oRendu powered by Sonore 7V ps designed for it, tX-USB powered by Farad Super3 12V and TT2 powered by Farad Super3 15V.  All Audio Sensibility silver DC cables except Ghent Audio JSSG360 Neotech coming from LPS-1.2 to eRegen.

 

Going to try removing the SOtM switch later too.

[cat6man]

1 hour ago, incus said:

And I should add a few details - SOtM switch powered by Farad Super3 12V, oRendu powered by Sonore 7V ps designed for it, tX-USB powered by Farad Super3 12V and TT2 powered by Farad Super3 15V.  All Audio Sensibility silver DC cables except Ghent Audio JSSG360 Neotech coming from LPS-1.2 to eRegen.

 

Going to try removing the SOtM switch later too.

 

@incus

 

how were you feeding the opticalRendu before adding the EtherRegen?  via SOtM switch directly or via opticalModule?

 

also, what is your source/server and what mode is the opticalRendu running?

[Jud]

6 hours ago, Mike Rubin said:

It sounds as though Jud tried but abandoned using the optical port as the downstream one for sound quality reasons.  Is anyone else connecting a device through an SFP module and, if so, how are you feeling about the ER in that application?  I ask because I have an optical rendu now connected to the network with a Sonore optical module that I would switch to another system if the ER is an upgrade of the magnitude this thread suggests it is.

 

Upstream - Copper Ethernet from router feeding Cisco, then Cisco via optical to EtherREGEN. I liked copper Ethernet from router directly to EtherREGEN better. My speculation is that this was due to the Cisco's SMPS putting noise back into the wall and the system. Someone who owned a device with an optical Ethernet output that you could power from a good quality low electrical noise low leakage current power supply (LPS-1.2 for example) might have a very different experience.

[incus]

1 hour ago, cat6man said:

 

@incus

 

how were you feeding the opticalRendu before adding the EtherRegen?  via SOtM switch directly or via opticalModule?

 

also, what is your source/server and what mode is the opticalRendu running?

I was feeding the opticalRendu straight from one of the optical ports on the SOtM switch.  Server is SOtM sMS-1000SQ Ultra SE (with sCLK-Ex clock board, external master clock connection, upgraded Evox caps, UPOCC silver wiring, eABS absorption paper), used as Roon core, using opticalRendu as Roon endpoint. Occasionally I upsample via HQp on another server and use the sMS server or the opticalRendu as an NAA. 

[Jud]

6 hours ago, stevebythebay said:

I’ve just begun listening to the same piece played by different pianists in solo just to capture their style and how the recordings were miked and engineered.  Already very revealing.

 

Yes, this is the sort of thing I feel the ER makes more possible, or easier, depending on where your system was before.

[LowMidHigh]

Has anyone tried this?

ER -> Mac Mini (or PC) -> DAC.

 

I wonder if feeding the Mac Mini the ER would make a SQ difference.

[McNulty]

Anyone who has experimented with different power supplies on the ER yet? Maybe one of the beta testers?

[Superdad]

8 minutes ago, McNulty said:

Anyone who has experimented with different power supplies on the ER yet?

 

Wow, I was so enjoying that we made it 5 pages/120 posts before someone brought up power supplies. 

 

Of course you are all free to chat about that.  But for those wondering--and I am sure users will confirm--EtherREGEN sounds fantastic with the stock PS we ship with it.  Whatever folks report about PS upgrading can only be a fraction of its right-out-of-the-box goodness. 

[McNulty]

16 minutes ago, Superdad said:

 

Wow, I was so enjoying that we made it 5 pages/120 posts before someone brought up power supplies. 

 

Of course you are all free to chat about that.  But for those wondering--and I am sure users will confirm--EtherREGEN sounds fantastic with the stock PS we ship with it.  Whatever folks report about PS upgrading can only be a fraction of its right-out-of-the-box goodness. 

Well, I saw the influence of Ethernet cables was already shortly discussed, so I thought why not drop the other bomb... 😉 

 

For now I’m anxiously waiting for delivery of my ER with stock supply from Joel with Metawave, but of course curious if I would be missing out on a further SQ upgrade when using the stock supply.

[HumanMedia]

15 hours ago, LowMidHigh said:

Has anyone tried this?

ER -> Mac Mini (or PC) -> DAC.

 

I wonder if feeding the Mac Mini the ER would make a SQ difference.

I think that might defeat the purpose of the ER.  RFI-noisy general purpose computers are some of the things the ER is meant to separate your DAC from. You could attach it to the B side of the ER with your DAC on the A side which would work well. But since computers transmit airborne RFI as well, I would also try and keep the computer as far away physically as possible.

[Superdad]

14 minutes ago, HumanMedia said:

I think that might defeat the purpose of the ER.  RFI-noisy general purpose computers are some of the things the ER is meant to separate your DAC from. You could attach it to the B side of the ER with your DAC on the A side which would work well. But since computers transmit airborne RFI as well, I would also try and keep the computer as far away physically as possible.

 

I do not agree. The EtherREGEN is designed to isolate (typically on its 'B' side) the DAC-attached computer/streamer/renderer endpoint. Its benefits will be realized even if that DAC-attached endpoint is general purpose computer, Mac mini, laptop, NUC, whatever.  

And for @LowMidHigh's example of ER -> Mac Mini (or PC) -> DAC, the DAC likely has only USB input so the DAC can not be directly attached to the EtherREGEN.

[wwc]

From Superdad:    

"I do not agree. The EtherREGEN is designed to isolate (typically on its 'B' side) the DAC-attached computer/streamer/renderer endpoint. Its benefits will be realized even if that DAC-attached endpoint is general purpose computer, Mac mini, laptop, NUC, whatever."

 

Thank you, @Superdad, for clarifying that point.   There is so much information bouncing around here that it can be difficult to know which end is up, down, or sideways.    

 

I am using a Server (Taiko Audio)  which has significantly addressed RFI. but I am planning to insert a Mac Mini in between the Taiko Server and the DAC.  (The Mini is necessary to run some Bacch software).   I'm hoping the Mini will not add too much of a noise signature.

 

My plan is: 

Copper ethernet > Cisco 2960 > Copper ethernet >  ERegen "B"side > ERegen Optical "A" side > Optical Taiko Server > usb  Mini >  usb DAC

[k-man]

@wwc Have you considered the ISO Regen+ LPS1.2 after the Mac Mini? I think this is a critical component to add before the USB DAC. Quite a few posters on this listening impressions thread will have already benefited from such an optimised approach. 

[Superdad]

6 minutes ago, wwc said:

... I am planning to insert a Mac Mini in between the Taiko Server and the DAC.  (The Mini is necessary to run some Bacch software).   I'm hoping the Mini will not add too much of a noise signature.

 

Well in the context of a top system (seeing you mention Taiko Server indicates you have some other investments there), I really need to point out that using a Mac mini as the DAC-attached endpoint will be a real compromise, especially if you stay with macOS.  If you must stay with a Mac--I assume to run macOS--then you can help it considerably with our Mac mini DC-conversion/Linear Fan Controller Kit (MMK) and our JS-2 choke-filtered, dual-output, 5-7.4A linear power supply.

 

6 minutes ago, wwc said:

My plan is:  

 

Copper ethernet > Cisco 2960 > Copper ethernet >  ERegen "B"side > ERegen Optical "A" side > Optical Taiko Server > usb  Mini >  usb DAC

 

Mac mini aside, what you propose is not the optimal set up.  You will get much better performance by feeding copper from the EtherREGEN's 'B' port to the Mac mini.  You can still use optical from the Taiko, running that to the EtherREGEN's 'A'-side SFP cage.  And run a copper Ethernet cable from your Cisco (or direct from your router) into one of the other 'A'-side Gigabit port.