So I've had an EtherRegen in rack for 2 weeks...

[plissken]

Here's my setup:

 

Playback Machine:

Dell Optiplex with Intel Dual Port Server NIC (Pro 1000), Win7 Pro X64, Kingston SSD, 4GB RAM. Using Intel's ANS to run both ports in a static LAG (so not using LACP negotiation). The team is running VLAN 2 tagged.

 

JRiver 26 set @ defaults

 

Port 1 goes to the B side EtherRegen and the other side goes to port 23 on my Aruba 2530

 

Port 2 goes direct to port 24 on the Aruba

 

On the Aruba: trunk 23-24 trk1 trunk (at the end it's either trunk (on) or LACP (negotiated) ) so two physical cables but seen at layer 2 MAC as a single connection.

 

File Server:

HP Laptop connected to port 1 on the 2540. Port one is in VLAN 2 also. Win10 Pro X64. So on the Aruba it's 'VLAN 2 untagged 1'. This places the port in VLAN 2. The Cisco equivalent is 'Switchport mode Access' then 'Switchport access vlan 2'.

 

DAC:

Emotiva DC-1

 

Transducers:

Audio Technica ATH-M50X and AKG 701's

 

Music:

http://www.2l.no/hires/

 

I simply used the EtherRegen for the first week. No veils lifted, no more slam, no more air around the highs.  During this 2nd week while queuing up and letting tracks play I started to use the static LAG that I built.

 

I'm able to pull either cable and have zero drops in ping. So the LAG is working as intended. There are also zero drops in music and if I pull both connections for a few seconds. For even modern streamers this could mean a minute or more of buffered playback. For JRiver I can option the entire track be buffered. Even at 11MB/s that the EtherRegen handicaps me at 24/192 tracks will load in 10-15 seconds on average.

 

It all sounded great regardless.

 

My 1st issue is that streamers and computers buffer data, we aren't playing in real-time off the Ethernet port. When I pull both cables I'm able to still experience playback. Using the 'Leakage Currents' argument that is when the audio should have taken an appreciable, and readily apparent, uptick in quality.

 

My 2nd issue is that GBe Ethernet gives us a 1000% improvement in transfer speeds therefore reducing any theoretical exposure to gremlins and we get 802.3az.

 

My 3rd issue, and this is mostly for people with DIY computer head end and nodes, is that for the money I can go and get an MikroTik switch for $240 with eight 10GBe SFP+ ports , 10GBe SR LC/LC SFP+'s for $19 a piece, MM OM4 cabling, and 10GBe PCI-E SFP+ NICs for about the same $640. The fiber piece also takes care of any current leakage and I get a modern, feature rich switch in the process.

 

 

 

 

 

[plissken]

Here is a quick video going through the config and showing play back while I'm removing one cable or the other and then finally both.

 

What is the EtherRegen doing when the music is playing but no cables are plugged in?

[plissken]

4 hours ago, mansr said:

How does it perform as a switch? Throughput, latency, packet loss, etc.

 

I'll go into port counters on the switch and post a screen shot.

 

Throughput was typical of a 100Mbit connection. Pings where good and sub 1ms. For the money I think it's a terrible value proposition.

 

 

[plissken]

1 minute ago, mansr said:

What about between the gigabit ports?

 

I didn't even bother to test since that isn't the problem it's positioned to solve. I'll take a look.

 

[plissken]

2 minutes ago, mansr said:

We already knew that.

 

With so many subjectivists asking "Did you try it" and pointing out there is a 30 day MBG I wanted to give it a shot even though it flies in the face of logic and buffered systems.

 

 

[plissken]

Just now, Blackmorec said:

Use your money back guarantee. Nothing won, nothing lost. 😉

 

I think an additional data point and demonstration is a positive. And yes I'll be returning the unit.

[plissken]

 

 

[plissken]

Try watching the video Sandy and answering the question of what happens to the audio when both cables are pulled and by some magic the audio still plays.

[plissken]

6 hours ago, emcdade said:

How did you control for sighted bias against the Etherregen?

 

I didn't. My setup is showing that you don't play music off the cable. You play it out of RAM, or possibly the hard drive if the block of RAM holding music data gets written to the swap file.

[plissken]

15 minutes ago, m3lraaHnevetS said:

I guess what you are saying is that using a 500 USD DAC and a 130 USD headphone you can not hear the benefit of the etherregen, right?

So people in the same position as you might consider spending their money on a better DAC and or headphone, before getting an Etherregen correct?

Interesting, thanks for the data point.

 

"I guess what you are saying is that using a 500 USD DAC and a 130 USD headphone"

The ATH-M50X are $149, the AKG 701's were $449.

 

I picked up the Emotiva specifically because the headphone amp is killer and it can also act as a multi source pre-amp.

 

So as far as Headphone listening it's right up there with other high fidelity setups. Now if your personal preference isn't for the AKG's that's fine. And if it isn't for the DC-1 headphone amp that's fine also. You're welcome to that. But empirically they are both top shelf components.

 

"So people in the same position as you might consider spending their money on a better DAC and or headphone, before getting an Etherregen correct?"

 

What position would that be? Not being gullible? You care to place a wager using your own system? I can setup in your system what what I showed in the video and I can change cables out at any given moment. You'll be blind to that however. I'll provide the computer end point and you provide the USB cable and everything downstream of it.

 

I'm sure the above will shut you up.

 

I saw in your signature that you are using Paradigm S8's. I wouldn't go calling out other peoples setups with those speakers as my reference.

 

 

[plissken]

11 hours ago, wgscott said:

 

Maybe the power supply noise is stored in the buffer.

 

I've actually seen a 'fast transient' noise argument used. I honestly had a hard time not laughing.

 

 

Kind of scary.

[plissken]

2 hours ago, charlesphoto said:

It seems to me that expectation bias that objectivists so bemoan can perhaps work both ways - maybe some don't want to hear improvements just so they can prove a point. I'm not saying the OP is lying, because we all hear differently in our different systems and ears/brains, but it should be recognized bias can be both positive and negative. 

 

2 hours ago, The Computer Audiophile said:

This is absolutely true. 

 

100% couldn't agree more. But I'm making a few points. One, and I've shown it beyond a shadow of a doubt, I can host not only two different cables, but two different switches, all in real time without interrupting playback.

 

Now this may change with a dedicated streamer. I have a Marantz streamer that I will be putting on the bench shortly to see what kind of buffering we get.

 

My second point, and it's the more important, and also the one the more subjectivist side seems to have a hard time answering or even offering any counterpoint:

 

When I removed both cables and the sound is still playing: Does it get better, worse, or stay the same?

When I removed both cables and the sound is still playing: Is sound coming directly off the Ethernet Cable or from somewhere else. If from somewhere else what does the Ethernet cable have to do with what is currently being heard other than potentially delivering this audio prior. Perhaps minutes prior.

[plissken]

30 minutes ago, m3lraaHnevetS said:

okay, you convinced me, I am dumping those useless Paradigm's and start saving for a killer setup like your's 😉 

 

I am in Europe so not likely to do the test you proposed sorry

 

I guess the fact that you do not hear any changes when using the ER and/or pulling cables means even the ER can't polish a turd

 

I've heard the various versions of what you have. They've always sounded like Paradigm used woopi cushions for base extension.

 

You wanted to talk about setups remember?

[plissken]

52 minutes ago, m3lraaHnevetS said:

I guess the fact that you do not hear any changes when using the ER and/or pulling cables means even the ER can't polish a turd

 

It's clearly in the video. While I can explain things to you I can't understand things for you.

[plissken]

1 hour ago, RickyV said:

To me, and I am no expert, there is no doubt that expectation bias works both ways.

If you really believe something can’t work, will you hear it?

If you really believe something will work, will you hear it?

It is just from which way you approach the conundrum.

 

 

Totally agree. Again it's why I went and created a proof of concept video. This is something that can be done in  about any listening setup. Could run the cables to another room while the evaluation is ran.

[plissken]

52 minutes ago, soares said:

I really do not understand what you want to proof with your test!!! The data is already in the buffer of the dac/streamer and was already subject to the relocker process and electronic cleaning.

 

That's the entire point of the demonstration: The data is in a static buffer. You can even pause/play it. There is no clock on 'at rest' data. What is 'electronic cleaning'? How is it stored on in buffer or even disk. Does the eR perform this cleaning on PDF's? Job's I send to the printer? HD video?

 

The data in the buffer is the exact same no matter how it got there: Dial Modem, DSL, Wired Ethernet, WiFi...

 

52 minutes ago, soares said:

The  eR is so far one of the best bang for the buck I ever experienced. It’s not subtle. It’s night and day.

 

And I'm willing to bet you money that when you don't know what is in play, either a $640 'audiophile' switch, or some sub $100 switch that I can perform 802.3ad (my $69 Cisco SG-200/8 does this) your night and day difference will poof up in smoke.

 

52 minutes ago, soares said:

I suggest that you indicate somewhere here at AS the city where you live and probably one of our audiophile fellows would be kindly inclined to show you the difference between a normal switch, a Cisco (who knows, most of us were using them while waiting for the eR) and an eR

 

I've mentioned before I'm in the Ky/Indiana area. I would like for someone to do this in their own setup.

[plissken]

36 minutes ago, soares said:

Thank you for your reply. I rest my case. I hope that someone will be able to show you the difference. All the best. 

 

You shouldn't try and make technical points that you have no ability to defend.

 

Maybe you can use the eR to clean and re-clock to improve the amount of money in your banking account when you log on to their website.  Before you dismiss it you should try it first.

[plissken]

2 hours ago, RickyV said:

 

That doesn't sound very plausible, engineering a switch and not making any measurements. 

 

 I am looking forward to the publication of his white paper but somehow I have the feeling it will not be enough to satisfy some of us.

 

Find a single white paper that UpTone has published on any product they offer.

[plissken]

3 hours ago, RickyV said:

is @JohnSwenson he made the right measurements and engineered accordingly.

 

2 hours ago, 4est said:

Uptone has made it clear about Swensen needing to build something in order to measure what they are attempting to do.

 

Maybe you two should get together and make sure you have your 180 degree stories straight.

[plissken]

2 hours ago, sandyk said:

 

I would be pleasantly surprised if Plissken even knows how to correctly drive his AKG K701 headphones. The  ATH M70x doesn't have such a huge dip in the 5K to 10K area as the M50x either.

 

I wonder how the graphs would change with the eR...

[plissken]

1 minute ago, RickyV said:

I think John is still busy making the last measurements but superdad is releasing some info from it via the ER review from Austinpop.

 

Thanks.

[plissken]

7 minutes ago, RickyV said:

Nope both statement are correct 

 

We'll have to agree they are both written in English.

[plissken]

25 minutes ago, 4est said:

Sadly I think the OP's intent was for a pissing match by the looks it.

 

Here is my main goal and why I shot a video:

 

Did the sound change when I had both cables (port bundle) connected

The 2530 Switch only

The eR Switch only

No connectivity.

 

And the question is: If I can pull both cables out randomly and we still hear audio and it hasn't changed, which is what I'm willing to bet will be the outcome if you can't look, what mechanism is at play here? Since audio is cached and I'm able to randomly swap cables and we know that the sample you are currently listening to was copied over ahead of time how do you know which cable it arrived on?

 

What mechanistic or electrical behaviors or properties are getting stored with the buffered data?

 

I'm not making this a pissing match. I've asked very foundational questions.

[plissken]

45 minutes ago, sandyk said:

 I.O.W. , you are almost certainly unaware that the AKG K701 was designed to comply with IEC 61938.The now superseded International Standard IEC61938 specified that headphones should be driven by a 120 ohms source regardless of headphone impedance .

 

The IEC61938 looks to be a standard for audio system interconnect and with specificity for Phantom powered microphones.

 

So I'm not sure where you are pulling this from.

 

I have the Q701's

 

** Edit ** Found a Wiki

 

Output impedance

Many headphone amplifiers have an output impedance in the range of 0.5 - 50 Ohms. The 1996 IEC 61938 standard recommended an output impedance of 120 Ohms, but in practice this is rarely used and not recommended with modern headphones. High output impedance can result in frequency response fluctuations, due to varying load impedance at different frequencies. In 2008 Stereophile Magazine published an article that showed that a 120-Ohm output impedance could cause a 5-dB error in frequency response with certain types of headphones. However, the author of the article also states: "The ramifications for subjective assessment of headphones are more troublesome because it is usually unclear what assumptions the manufacturer has made regarding source impedance." ^[2]

More importantly, low output impedance can reduce distortion by improving the control that the source has over the transducer. This is often expressed as damping factor, with higher damping factors greatly reducing distortion.^[3] One company shows a 45 dB improvement in THD+N at 30 Hz for their low-impedance amplifier compared to a 30-ohm amplifier.^[4] For example, a 32 Ω headphone driven by a headphone amp with a <1 Ω output impedance would have a damping factor of >32, whereas the same headphone driven with an iPod touch 3G (7 Ω output impedance) ^[5] would have a damping factor of just 4.6. If the 120 ohms recommendation is applied, the damping factor would be an unacceptably low 0.26 and consequently distortion would be significantly higher. Conversely, the same iPod touch driving a pair of 120 ohm headphones would have a respectable damping factor of 17.1, and would most likely not benefit from the addition of a lower impedance headphone amplifier.

In addition to output impedance, other specifications are relevant to choosing a headphone amplifier — THD, frequency response, IMD, output power, minimum load impedance, and other measurements are also significant. However, most of these will be improved by lowering output impedance and hence improving damping factor.

 

 

[plissken]

1 minute ago, The Computer Audiophile said:

But I bet you have a guess 😁

 

On second read through I know what it sounds like. I did a bit of digging however. Also just visited the Reddit HP thread:

 

" Lastly due to it's low impedance, some believe you need a amp with a low output impedance to prevent damping effects... however, it has a fairly linear frequency to impedance graph... which make this less of an issue. The folk at rock-grotto believe the optimal output impedance for it to be 120 ohm... but that's really the only place I've seen that. "

 

So it's not as cut and dry as Sandy is making it out to be.