[Davesrose]

I very much enjoy Chord products. I regularly use a Mojo2 and TT2 and have decent experience with the Hugo2. In all three instances the sound is noticeably superior via the Innuos streamer than it is from any PC or laptop/MacBook I own.

If there is a perceptible difference in audio between these devices, how do you not know it’s the difference in which the audio is processed? The claim with ethernet is that it eliminates noise and improves the audio from the actual transmission. If measurements show there’s no difference in noise from a $20 switch vs $700 audiophile one, let alone difference in digital bandwidth, it makes you wonder if there’s a bias when someone says “I plugged this in my system and there was a night and day difference in sound”. There will continue to be a round robin with skeptics until there’s a blind test between $20 ethernet hub going off a PC vs audiophile ethernet off same PC.

 

[AussieMick]

If there is a perceptible difference in audio between these devices, how do you not know it’s the difference in which the audio is processed? The claim with ethernet is that it eliminates noise and improves the audio from the actual transmission. If measurements show there’s no difference in noise from a $20 switch vs $700 audiophile one, let alone difference in digital bandwidth, it makes you wonder if there’s a bias when someone says “I plugged this in my system and there was a night and day difference in sound”. There will continue to be a round robin with skeptics until there’s a blind test between $20 ethernet hub going off a PC vs audiophile ethernet off same PC.

Nah, it makes me wonder what measurements aren’t being made in order to detect why so many people hear differences. It’s the scientific principle; lots of us observe this phenomena, how can we figure out objectively what’s going on to make that change. The assumption that we already have an adequate suite of measurements to describe everything in audio is clearly not the right one.

 

[Davesrose]

Nah, it makes me wonder what measurements aren’t being made in order to detect why so many people hear differences. It’s the scientific principle; lots of us observe this phenomena, how can we figure out objectively what’s going on to make that change. The assumption that we already have an adequate suite of measurements to describe everything in audio is clearly not the right one.

If Amir's measurements handles signals in the Mhz, I don't see how there can be a difference in transmission (especially since the design for ethernet is to not be phased by noise, there's already methods in isolating it with the protocol, and audio bandwidth is not as demanding). The premise was that these devices eliminate noise: let's just assume it makes a difference, measurements in noise were identical for the $20 hub and the $700 audiophile hub. The scientific principle is to carry out a ABX test ($20 ethernet hub vs one off these devices off same PC and receiving audio setup), to make sure people can hear a difference that's not from bias. I'm not aware of any such demonstration confirming audible difference. Linus did a controlled test with staff members, and results were as you'd expect from probability. So again, the only way to appease a science skeptic is a controlled test: not sighted test that you're changing.

 

[runningwitit]

Bought it 2 days ago , i think zen stream and Lan isilencer are perfect match.
No need to Argue
This is only a US$100 toy, which is a lot cheaper than acoustic revive, jcat, sotm.
I think you should challenge the above manufacturers

What are the positive/negative changes made in sound with using one vs. two in the chain?

 

[AussieMick]

If Amir's measurements handles signals in the Mhz, I don't see how there can be a difference in transmission (especially since the design for ethernet is to not be phased by noise, there's already methods in isolating it with the protocol, and audio bandwidth is not as demanding). The premise was that these devices eliminate noise: let's just assume it makes a difference, measurements in noise were identical for the $20 hub and the $700 audiophile hub. The scientific principle is to carry out a ABX test ($20 ethernet hub vs one off these devices off same PC and receiving audio setup), to make sure people can hear a difference that's not from bias. I'm not aware of any such demonstration confirming audible difference. Linus did a controlled test with staff members, and results were as you'd expect from probability. So again, the only way to appease a science skeptic is a controlled test: not sighted test that you're changing.

I understand that, but you’ve just pointed to existing measurements as a rebuttal to me suggesting existing measurements aren’t getting it done. A/B/X testing would be fine with experienced listeners in their own systems. Let’s get ten high level audio design engineers to sit the test. From ten top flight companies. Would that do?

 

[Davesrose]

I understand that, but you’ve just pointed to existing measurements as a rebuttal to me suggesting existing measurements aren’t getting it done. A/B/X testing would be fine with experienced listeners in their own systems. Let’s get ten high level audio design engineers to sit the test. From ten top flight companies. Would that do?

You're rebuttal was that scientific measurements can't be accurate because people are claiming differences in hearing. These measurements have higher tolerances than human hearing and audio demands. For an ABX test, you need to be sure that you're testing one variable: since the question is if an ethernet device makes a difference, it would be switching between a $20 ethernet hub and one of these devices plugged in. You mentioned hearing a difference between a PC vs a dedicated streamer: but they have different stages in audio processing. To be consistent, seems the best setup would be to use a "noisy" PC as source and whatever audio setup you want on the other end. There needs to be results that prove there is an audible difference between ethernet connection: that's the only variable that's being claimed to have improvement. Certainly that would help convince the computer folks on this thread saying they're highly skeptical, and you can point to it when it's raised in other threads/reviews.

 

[AussieMick]

You're rebuttal was that scientific measurements can't be accurate because people are claiming differences in hearing. These measurements have higher tolerances than human hearing and audio demands. For an ABX test, you need to be sure that you're testing one variable: since the question is if an ethernet device makes a difference, it would be switching between a $20 ethernet hub and one of these devices plugged in. You mentioned hearing a difference between a PC vs a dedicated streamer: but they have different stages in audio processing. To be consistent, seems the best setup would be to use a "noisy" PC as source and whatever audio setup you want on the other end. There needs to be results that prove there is an audible difference between ethernet connection: that's the only variable that's being claimed to have improvement. Certainly that would help convince the computer folks on this thread saying they're highly skeptical, and you can point to it when it's raised in other threads/reviews.

Would you be happy with the panel of listeners I suggested, in their own familiar systems?
And for clarity, I didn’t say they weren’t accurate, I said we need tests we currently don’t have.

 

[Davesrose]

Would you be happy with the panel of listeners I suggested, in their own familiar systems?
And for clarity, I didn’t say they weren’t accurate, I said we need tests we currently don’t have.

No, because you're not testing one variable. You're saying we can get controlled results from people doing their own testing with their own setups? It makes it harder to be sure that they're doing a true ABX test with other people actually administering the test. Seems a controlled test where you have a group of listeners trying to detect an audible difference in ethernet connection with "noisiest" PC would be optimal. The claim is that noise in an ethernet connection is detrimental to audio components, and these devices can actually improve sound. Why not make the source something that's supposed to be detrimental and can achieve improvement?

Keep in mind that I'm arguing this from a standpoint of how you can scientifically prove there are audible differences (and then you can start convincing the people saying the way ethernet works, there is no difference with audio). The problem the argument "scientific measurements aren't good enough because people are saying they hear a difference", is that perceptual bias has been known for decades. One of the items in James Randi's $1 million challenge was being able to hear the difference between expensive speaker cables vs inexpensive.

 

[AussieMick]

No, because you're not testing one variable. You're saying we can get controlled results from people doing their own testing with their own setups? It makes it harder to be sure that they're doing a true ABX test with other people actually administering the test. Seems a controlled test where you have a group of listeners trying to detect an audible difference in ethernet connection with "noisiest" PC would be optimal. The claim is that noise in an ethernet connection is detrimental to audio components, and these devices can actually improve sound. Why not make the source something that's supposed to be detrimental and can achieve improvement?

Keep in mind that I'm arguing this from a standpoint of how you can scientifically prove there are audible differences (and then you can start convincing the people saying the way ethernet works, there is no difference with audio). The problem the argument "scientific measurements aren't good enough because people are saying they hear a difference", is that perceptual bias has been known for decades. One of the items in James Randi's $1 million challenge was being able to hear the difference between expensive speaker cables vs inexpensive.

Sorry, I wasn’t clear. You and I can travel to each panel member’s home system and test the single variable in a system they’re otherwise familiar with. Ten people sitting the test would mean ten different rooms and systems, but still only one variable changing for each test.

 

[AussieMick]

I’d suggest we then do the sane with power cables.

 

[Davesrose]

Sorry, I wasn’t clear. You and I can travel to each panel member’s home system and test the single variable in a system they’re otherwise familiar with. Ten people sitting the test would mean ten different rooms and systems, but still only one variable changing for each test.

From a practicality standpoint, how easy is it to have the science based test takers going to each audiophiles home? For example, we can't meet since I'm in the US. Seems a great way to get a sample pool is setup a test at a headphone meetup (where you have network/science folks chime in about how the test should be done). Since PCs are supposed to be bad for audio, lets have that as source (heck, if the venue has a server room even better: switch ethernet on one of the blade servers where there's lots of different noise in the room). I also think noisy PC as source is good because audiophile products are supposed to be better (more expensive should mean it can reject noise).

If during this meetup, you wanted to do a seperate controlled test of power cables, that's a seperate test. We've just brought up ABX testing for ethernet cables as this thread has brought up the invariable "how can an audiophile ethernet device make a difference?".

I'm an audiophile with some things: IE headphones and amps certainly have different sound signatures. When it comes to certain things like balanced vs unbalanced, I've noticed main factor with my home audio products is difference in gain (it's more for commercial applications that have very long cable runs). I like my iFi iCAN for its options and hybrid tube setup (which also is one of the few amps that doesn't pick up the radio station near me). I'm also involved with 3D animation and computer programming, so I'm used to needing high bandwidth networks. There's much more demand streaming a UHD rip that's 4K HDR movie plus 7.1TrueHD Atmos movie. Only issues I've found with a digital cord is if it has bad QC and interconnects fail, or if it's so poor that you hear static or there's noticeable frame drops.

 

[AussieMick]

From a practicality standpoint, how easy is it to have the science based test takers going to each audiophiles home? For example, we can't meet since I'm in the US. Seems a great way to get a sample pool is setup a test at a headphone meetup (where you have network/science folks chime in about how the test should be done). Since PCs are supposed to be bad for audio, lets have that as source (heck, if the venue has a server room even better: switch ethernet on one of the blade servers where there's lots of different noise in the room). I also think noisy PC as source is good because audiophile products are supposed to be better (more expensive should mean it can reject noise).

If during this meetup, you wanted to do a seperate controlled test of power cables, that's a seperate test. We've just brought up ABX testing for ethernet cables as this thread has brought up the invariable "how can an audiophile ethernet device make a difference?".

I'm an audiophile with some things: IE headphones and amps certainly have different sound signatures. When it comes to certain things like balanced vs unbalanced, I've noticed main factor with my home audio products is difference in gain (it's more for commercial applications that have very long cable runs). I like my iFi iCAN for its options and hybrid tube setup (which also is one of the few amps that doesn't pick up the radio station near me). I'm also involved with 3D animation and computer programming, so I'm used to needing high bandwidth networks. There's much more demand streaming a UHD rip that's 4K HDR movie plus 7.1TrueHD Atmos movie. Only issues I've found with a digital cord is if it has bad QC and interconnects fail, or if it's so poor that you hear static or there's noticeable frame drops.

I’m absolutely convinced that the people being tested should be industry heavyweights. If the hugely experienced design engineers can’t pick differences reliably I’d be satisfied. If it’s just anyone with any degree of experience the test won’t be as valid. Indeed, we’d want people positively biased towards believing there’s a difference. An inexperienced listener doing the test might not even know what to listen for, tainting the test.

I’m thinking of it from a “what would the cost no object best scenario be?”

 

[Davesrose]

I’m absolutely convinced that the people being tested should be industry heavyweights. If the hugely experienced design engineers can’t pick differences reliably I’d be satisfied. If it’s just anyone with any degree of experience the test won’t be as valid. Indeed, we’d want people positively biased towards believing there’s a difference. An inexperienced listener doing the test might not even know what to listen for, tainting the test.

I’m thinking of it from a “what would the cost no object best scenario be?”

How would you determine who is an experienced design engineer? For example, my first interaction in this thread was asking what can this device do if basic cat5 is adequate for audio applications. You gave me two podcasts from Darko Audio: a source that's known for focusing on marketing and interviewing enthusiasts/CEOs. So I've interacted with an audio engineer about the recent interest in DSD. Why was it ever a thing since you can't mix in it, and it has to be converted to PCM? Perhaps the over-riding factor was that CD patents were expiring, so Sony introduced SACD for a new format (I think Sony minimized cost impact given they have Columbia, and it also offered surround). Anyway, he highlighted how marketing can be very different from the engineering department. There was actually an event in which they included a lead engineer in the discussion panel. The take home the engineer gave was that there was no practical difference between CD standards from a mastering standpoint (so needless to say, they never brought him out again). The only study that found an audible difference was some perception in noise floor (but not something that applicable for reproduction). I collected SACDs due to different masterings. Like I do vinyl records: don't understand collecting new vinyl outside some for the offbeat holograms/coverart when the mastering is all digital.

I would hope that if you spend a lot on a digital audio device, it would reject noise. Seems it would be more pertinent to test from the least desirable device *that is supposed to have noise*. If you watch Amir's video, his only hypothetical about how an expensive audiophile ethernet hub would help is from a cheap music streamer that somehow doesn't meet ethernet standards. But why would you spend $700 to try to fix that vs buy a streamer that's competent. We're talking money that's cheaper than a PC.

So far you have indicated you think PCs are worst with noise. Why then go for "best scenario" as a source that has noise that needs to be eliminated. Now after a controlled test of PC being source, you want to test some high end systems, that could be interesting. You would think a high end system should not have perceived noise: if it has it, then you're actually paying more for a defect.

 

[AussieMick]

How about chief designers from CH Precision, Wilson Audio Specialties, Chord Electronics, Focal and Boulder (or any companies of their ilk) listening to the PC based system you describe?
Don’t forget that people who start audio companies ARE enthusiasts who are often the CEO. The idea behind getting these people is that they’re the ones who spend their days listening as a profession. If they can’t hear differences I’d doubt anyone else could either.
I wouldn’t trust an experiment like this if the participants weren’t expert. After sll, we’re not trying to establish if a difference is discernible by Joe Public, but if a difference is discernible AT ALL.

 

[Davesrose]

How about chief designers from CH Precision, Wilson Audio Specialties, Chord Electronics, Focal and Boulder (or any companies of their ilk) listening to the PC based system you describe?
Don’t forget that people who start audio companies ARE enthusiasts who are often the CEO. The idea behind getting these people is that they’re the ones who spend their days listening as a profession. If they can’t hear differences I’d doubt anyone else could either.
I wouldn’t trust an experiment like this if the participants weren’t expert. After sll, we’re not trying to establish if a difference is discernible by Joe Public, but if a difference is discernible AT ALL.

Are you really sure CEOs are spending their days listening and designing systems? Not really sure what an expert is if you think we have to find golden ears. (sorry, just my wording about how the test pool has to be exceptional). Heck, if we're saying the end all be all test for an audiophile ethernet connection with test pool, why not increase it? The premise is that it's obvious that an audiophile ethernet device eliminates noise and improves audio: but we need golden ears and audiophile components to test? Sure, if these companies are willing to participate in ABX, that's great. But also not sure why it has to just be exclusive to them if the claim is that it's demonstrably clear there's a difference. If it's demonstrable, why not a larger random sample?