[bfreedma]

You are repeating the comfortable beaten old path of dismissiveness by saying 'prove to us objectively that you are hearing what you are hearing' which is very similar to saying 'show the measurement proving that your whatever speaker sounds better than whichever other speaker', where their freq curves are highly identical, so is phase response, and the waterfall is not showing much difference either. Do you buy speakers and stuff based on TS or other parameters such as THD alone?

We're not talking wireless here, at least I am not, I will also repeat what I stated before; fiberoptic versus UTP on the same two switches in the same setup and there is an audible difference I cannot explain, but that does not invalidate the observation. Obviously if the precious measurements would have been any different the underlying cause would be closer to us, what s happening here is akin to someone with a yardstick calling out 'see the earth is flat because my yardstick does not need to bend to measure the ground we're walking on', he was right with that observation but in the larger scheme of things that observation ultimately proved not to be connected to the truth that the earth is round.

I'm not saying the measurements and protocols are incorrect, I've looked at packets lost etc (and did not see any lost in any working network) and my data files do not look brighter or darker or better in any way or shape through fiber optics. Still there is an audible difference and that observation is reported by a good number of people, some of which with very inquisitive nature, and to great (well huge) expense by some. To be clear, we're talking regular 'pro IT' components, stuff like stacking layer 2 switches, not potential snake oil bits and pieces with any audiophile label stuck to them. Why do files played from RAM sound better than from SSD , and why do they sound better when played from Intel Optane, the checksum is no different.... Do you really believe folks are buying Terabyte or so Optane PCIe discs at a price tag of a couple of grand each because they are gullible?

Yes, I believe that people are buying these expensive black boxes because they are gullible and lack the domain knowledge to understand why the manufacturers marketing claims aren’t technically possible.

Files played from RAM do not sound different than those played from SSD. Lack of your presentation of any evidence aside, that claim ignores the multiple buffers the data will still pass through, so not sure how anyone would see improvement from a source so electronically distant from the output,

Outrageous claims with no evidence will be dismissed quickly. And throwing comparing speakers in to this is the usual attempt at deflection. Of course different speakers can and will sound different. Speakers are also heavily impacted by room acoustics, which electronics are not. I never stated otherwise. I’ve never seen anyone here state otherwise. The discussion here is whether a specific product (audiophile network switch) can improve audio. It can’t because the 802 standard ensures it.

If you believe that the standard that currently enables trillions of bytes of data to successfully traverse global networks is incorrect or incomplete, put together the appropriate technical documentation. I was previously on several of the 802 standard subcommittees, so can quickly get it to the right people. It will need to be evidence based, so hard data documenting the improvements you suggest audiophile switches provide outside of the 802 standard will need to be included.

 

[Claypole]

Has anybody ever unscrewed the faceplate on the wall where they plug their phone/router into and had a look at the cabling? I have in my flat, and it's ancient Cat 3 UTP, terminated into an old cheap looking connector. I've also seen inside the junction board the end of my landing outside my property. It's a horrible looking mess in the same riser as the main electrical supply.

It seems a bit pointless to upgrade network components in my property when everything beyond the socket on the wall is clearly not 'audiophile' grade'. Or does all that other stuff not matter because I can't see it when I'm streaming music?

 

[bigshot]

Yes, I believe that people are buying these expensive black boxes because they are gullible and lack the domain knowledge to understand why the manufacturers marketing claims aren’t technically possible.

Absolutely. I think it's more the latter though. A lot of people lack the interest to do the research to understand how things work. They just want to buy something that someone in authority tells them works good. The problem is that authorities in home audio are employed directly or indirectly by the manufacturers. So they don't always give good advice. A lot of their advice is designed to upsell, not to solve real problems. Any time you spend a lot of money without making an effort to become an informed buyer, you're going to get in trouble... whether it's buying a used car, a fixer-upper house, or buying high end audio components.

 

[bfreedma]

Absolutely. I think it's more the latter though. A lot of people lack the interest to do the research to understand how things work. They just want to buy something that someone in authority tells them works good. The problem is that authorities in home audio are employed directly or indirectly by the manufacturers. So they don't always give good advice. A lot of their advice is designed to upsell, not to solve real problems. Any time you spend a lot of money without making an effort to become an informed buyer, you're going to get in trouble... whether it's buying a used car, a fixer-upper house, or buying high end audio components.

My best audio investment was Floyd Toole’s book on Sound Reproduction and Psychoacoustics. That $50 + time invested in reading put focus on where to spend money on impactful improvement rather than what’s being marketed.

​

 

[marcelnl]

Perhaps you want to invest some time in reading what is presented before jumping to conclusions that it's a conspiracy to upsell anything to anyone, I was talking about people buying/experimenting with regular off the shelf high endish pro IT equipment not marketed to audio in any way form or shape.

The speaker analogy gets dismissed because of room acoustics, but would that room acoustic not be identical no matter what two identical speakers are sitting in that room?
I'm happy to read that there is a standard making every network sound the same, can we please write a standard to ensure all audio sounds like the master?

 

[bfreedma]

Perhaps you want to invest some time in reading what is presented before jumping to conclusions that it's a conspiracy to upsell anything to anyone, I was talking about people buying/experimenting with regular off the shelf high endish pro IT equipment not marketed to audio in any way form or shape.

The speaker analogy gets dismissed because of room acoustics, but would that room acoustic not be identical no matter what two identical speakers are sitting in that room?
I'm happy to read that there is a standard making every network sound the same, can we please write a standard to ensure all audio sounds like the master?

i don’t know why you’re bringing speakers into the discussion. No one here stated that they sound the same - that’s just a red herring being injected into this discussion.

There are numerous standards to ensure the output signal is as close to the master as possible (once it hits the transducers, that becomes impossible). And nearly every time those standards are discussed, some people simply dismiss them by prioritizing their subjective listening experience over those standards. Add to that the marketing material used by manufacturers describing solutions in search of problems to solve and we end up with people believing audio somehow is “special” and not properly supported by known laws of physics and electrical engineering.

 

[bigshot]

Wow. That sure is a lot of irrelevant smoke thrown up at once! I’m just going to let that one lay on the ground and bake in the sun. I don’t see anything to be gained in replying. He’s all yours!

 

[marcelnl]

There are numerous standards to ensure the output signal is as close to the master as possible (once it hits the transducers, that becomes impossible). And nearly every time those standards are discussed, some people simply dismiss them by prioritizing their subjective listening experience over those standards. Add to that the marketing material used by manufacturers describing solutions in search of problems to solve and we end up with people believing audio somehow is “special” and not properly supported by known laws of physics and electrical engineering.

it's all a conspiracy making people buy stuff they don't need, with that stated there is no need to do an experiment...circular argument.
there indeed is no reason in replying.

 

[bigshot]

I can make you a great deal on the Brooklyn Bridge...

 

[bfreedma]

it's all a conspiracy making people buy stuff they don't need, with that stated there is no need to do an experiment...circular argument.
there indeed is no reason in replying.

I don’t feel the need to test anything that the manufacturer doesn’t bother to and where claims aren’t supported by known operational models. If the manufacturers want to convince me they’ve made a significant new discovery, it’s up to them to vet and publish that data.

The blind faith audiophiles put into marketing material while summarily rejecting existing vetted knowledge is, frankly, disturbing. Do people approach medical treatment with the same cavalier approach? I certainly hope not, but wonder why they aren’t consistent.

 

[bigshot]

The ones who argue indefensible positions like this are most likely people who have gotten ripped off themselves and their ego won’t allow them to admit it.

 

[sander99]

You are repeating the comfortable beaten old path of dismissiveness by saying 'prove to us objectively that you are hearing what you are hearing' which is very similar to saying 'show the measurement proving that your whatever speaker sounds better than whichever other speaker', where their freq curves are highly identical, so is phase response, and the waterfall is not showing much difference either. Do you buy speakers and stuff based on TS or other parameters such as THD alone?

From measurements generally three types of conclusions can be drawn with respect to (differences in) audibility:
1. The measurements indicate that the differences or imperfections are far below the known tresholds of audibility. That means we can safely conclude there is no audible difference or imperfection.
2. The measurements indicate that the differences or imperfections are clearly above the known tresholds of audibility. That means we can safely conclude there is audible difference or imperfection.
3. The measurements are close to the known tresholds of audibility. That means the only way to establish more objective information is by conducting well controlled level matched double blind ABX tests. One possible outcome is that there are audible differences or imperfections. Although it can never prove that there are no audible differences or imperfections, if enough tests are done you can have a high confidence that there most likely are no audible differences or imperfections. (And even more likely that even if there was an audible difference or imperfection it is so minimal that it doesn't matter.)

Measurements of loudspeakers will generally fall in category 2, sometimes maybe 3. And if there are clear differences it isn't always easy to say which loudspeaker is better. With loudspeakers and other transducers some subjectivity is maybe hard to avoid. Not so with electrical signals of which we know exactly what they should be.
The measurements of the influence of switches that I have seen (in the video posted earlier) fall in category 1. Even extremely so, leaving no shadow of a doubt.

In short: your comparison is totally irrelevant.

The speaker analogy gets dismissed because of room acoustics

No, because the order of magnitude and the complexity of the measurements and the complexity of how to interpret them are totally different. As partly explained above.

Why do files played from RAM sound better than from SSD , and why do they sound better when played from Intel Optane

They don't. (Unless something is done very wrong in the specific implementation in a specific case).

We make it very clear our main tool of measurement used while developing our products is continuous subjective listening tests, and over time we’ve learned exactly which parameters are technically critical.

That is a problem (but of course not specifically yours, but of almost all people in the business of selling, "professional reviewing", etc.). Especially if not done well controlled double blind etc. etc. Uncontrolled sighted subjective listening tests are generally worthless. Hearing is done mostly by the brain, and not (just) the ears. The influence of expectation bias etc. can make you hear night and day differences even if there is no audible difference. This holds for everyone, you, me, and everyone else. it can not be switched of by thinking you don't have a reason to be biased or something like that. Also this makes it harder to hear real differences if they were there. Because they would easily be overshadowed/masked by the imagined differences. This all is so easy to prove: just let people do a listening test with a switch (no not a network switch but a knob with two settings) that does nothing, but tell them that it does something. Many people will hear differences. Studies even showed that the people's brain activity changes in a similar way to a real change as when they are just told that this change is made.

We describe our products as we hear them

That is always a nice excuse to get away with many audio related claims.

supported with independent customer reviews pulled from trustpilot and clearly labelled professional reviews.

Who are also prone to expectation bias etc., or sometimes just parrot after each other. Let the customers say it really sounds better, then you don't have to do it and can suffice with "We describe our products as we hear them (and this switch sounds nicer to me than that one)" and you don't need to say: "There is a real objective audible difference".

Everything we sell comes with free shipping an a 30 day no quibble money back guarantee so customers can listen to our products risk free in their own system. Customers can return a product for any reason and receive a full refund.

But that doesn't help the customer who either is biased and hears an imagined improvement, or a customer who just wants to make sure he has the best possible sound and believes you and/or reviewers at your word, they both think they paid for better sound and they keep the product but they won't get audibly better sound.

We describe our products as we hear them

Back to this one. Legaly ok. But to me it doesn't really matter whether a seller is really deluded and really doesn't understand all this and really believes the impossible or that he deliberately wants to spread misinformation. I think someone who sells stuff should feel the moral obligation to objectively establish the real value of what he is selling. Yeah, yeah, I know it is an utopia, it is not how the world works unfortunately and certainly I don't want to make you personally responsible for that.

it's all a conspiracy making people buy stuff they don't need, with that stated there is no need to do an experiment...circular argument.

No, there are measurements showing beyond a shadow of a doubt that there can not be an audible difference (and that was to be expected based on how switches and networks work), and that is why an (additional) experiment is not needed, and we explained that people can hear differences when there is no audible difference and that is why uncontrolled subjective listening can not have any weight against such extremely unambiguous measurement results like we have seen in this case.

 

[teknorob23]

From measurements generally three types of conclusions can be drawn with respect to (differences in) audibility:
1. The measurements indicate that the differences or imperfections are far below the known tresholds of audibility. That means we can safely conclude there is no audible difference or imperfection.
2. The measurements indicate that the differences or imperfections are clearly above the known tresholds of audibility. That means we can safely conclude there is audible difference or imperfection.
3. The measurements are close to the known tresholds of audibility. That means the only way to establish more objective information is by conducting well controlled level matched double blind ABX tests. One possible outcome is that there are audible differences or imperfections. Although it can never prove that there are no audible differences or imperfections, if enough tests are done you can have a high confidence that there most likely are no audible differences or imperfections. (And even more likely that even if there was an audible difference or imperfection it is so minimal that it doesn't matter.)

Measurements of loudspeakers will generally fall in category 2, sometimes maybe 3. And if there are clear differences it isn't always easy to say which loudspeaker is better. With loudspeakers and other transducers some subjectivity is maybe hard to avoid. Not so with electrical signals of which we know exactly what they should be.
The measurements of the influence of switches that I have seen (in the video posted earlier) fall in category 1. Even extremely so, leaving no shadow of a doubt.

In short: your comparison is totally irrelevant.

No, because the order of magnitude and the complexity of the measurements and the complexity of how to interpret them are totally different. As partly explained above.

They don't. (Unless something is done very wrong in the specific implementation in a specific case).

That is a problem (but of course not specifically yours, but of almost all people in the business of selling, "professional reviewing", etc.). Especially if not done well controlled double blind etc. etc. Uncontrolled sighted subjective listening tests are generally worthless. Hearing is done mostly by the brain, and not (just) the ears. The influence of expectation bias etc. can make you hear night and day differences even if there is no audible difference. This holds for everyone, you, me, and everyone else. it can not be switched of by thinking you don't have a reason to be biased or something like that. Also this makes it harder to hear real differences if they were there. Because they would easily be overshadowed/masked by the imagined differences. This all is so easy to prove: just let people do a listening test with a switch (no not a network switch but a knob with two settings) that does nothing, but tell them that it does something. Many people will hear differences. Studies even showed that the people's brain activity changes in a similar way to a real change as when they are just told that this change is made.

That is always a nice excuse to get away with many audio related claims.

Who are also prone to expectation bias etc., or sometimes just parrot after each other. Let the customers say it really sounds better, then you don't have to do it and can suffice with "We describe our products as we hear them (and this switch sounds nicer to me than that one)" and you don't need to say: "There is a real objective audible difference".

But that doesn't help the customer who either is biased and hears an imagined improvement, or a customer who just wants to make sure he has the best possible sound and believes you and/or reviewers at your word, they both think they paid for better sound and they keep the product but they won't get audibly better sound.

Back to this one. Legaly ok. But to me it doesn't really matter whether a seller is really deluded and really doesn't understand all this and really believes the impossible or that he deliberately wants to spread misinformation. I think someone who sells stuff should feel the moral obligation to objectively establish the real value of what he is selling. Yeah, yeah, I know it is an utopia, it is not how the world works unfortunately and certainly I don't want to make you personally responsible for that.

No, there are measurements showing beyond a shadow of a doubt that there can not be an audible difference (and that was to be expected based on how switches and networks work), and that is why an (additional) experiment is not needed, and we explained that people can hear differences when there is no audible difference and that is why uncontrolled subjective listening can not have any weight against such extremely unambiguous measurement results like we have seen in this case.

Thanks for taking the time

 

[SoundAndMotion]

...that’s just a red herring being injected into this discussion.

Indeed this thread is rife with red herrings (or straw men, I mix those up).
But most are the bits lost or flipped, packets lost or modified, or galvanic isolation/IEEE 802 red herrings. I can't claim to be knowledgeable on all the marketing claims, and I haven't read all internet posts everywhere, but I have not seen claims made for which these red herrings are relevant.
Most claims I've seen are alleged "better sound", and a handful of "reduced jitter".

When your "observations" enable a viable theory to be proposed, ...

A viable theory has been proposed: EMI/RFI causes audible jitter in the DAC, or audible analog noise/distortion in the subsequent analog stages. For this to be true, one has to assume an unusually* large EMI/RFI source and/or an unusually* vulnerable analog stage.
I say "unusually", because I assume a usual amount of either would be noticed and called out, naming the culprit (as Amir did with a Schiit DAC, the Modi 2, IIRC, as being very vulnerable to computer activity).
But the problem can't be too unusual, if one believes all those who hear "better sound". But I cut you off:

When your "observations" enable a viable theory to be proposed, let alone supported by evidence, there will be something to discuss.

One *can* discuss the theory without evidence, but I agree with you that only evidence, in the form of simple, but competent, measurements, or competently-performed (blind) listening tests, will bring the discussion further.

@teknorob23 , @marcelnl I get your perspective. But as always happens, these discussions, debates, challenges fall into the "I'm not doing anything. It's up to you to do what I suggest first" quagmire.
To break through this, I'll propose: I'm willing to go through the hassle/cost of listening to an audiophile switch, abiding by your constraints, if you'll do a measurement or listening test, abiding by my (very reasonable/simple) constraints. Deal?
I'll admit that I wouldn't be surprised if I hear a difference, possibly explained within Amir's video (time: ~32:00). But I won't keep the device, unless I hear the difference, as an improvement, "blind".

Although it can never prove that there are no audible differences or imperfections, if enough tests are done you can have a high confidence that there most likely are no audible differences or imperfections. (And even more likely that even if there was an audible difference or imperfection it is so minimal that it doesn't matter.)

I can see the second sentence as probable, but the first is incorrect. Quantity does not lead to quality. Doing something wrong many, many times does not converge on the right answer. Failure to reject the null hypothesis does not prove the null hypothesis. Absence of evidence is not evidence of absence. Blah, blah, blah... many ways to say the same thing.
I believe the biggest problems are the streetlight effect (link) and McNamara's fallacy (link).

EDIT:
Thinking about the streetlight effect, there are some problems with Amir's video. If you are testing something that claims to reduce problems audibly, you need to have the problem in order to reduce it. If your setup does not exhibit the problem to start with, you can't measure it's reduction.
Yes, I know that proponents of audiophile switches make the improvement claim as a general statement. But even Amir quickly glosses over the fact that even if no bits (or packets) are damaged, a vulnerable DAC can pick up noise from a "purely digital" USB input (time: ~8:12). In the case of audiophile switches, showing no change in the measurements, shows nothing.

 

[marcelnl]

@SoundAndMotion : that is pretty much the first reasonable and open minded response in this conversation!

Again, I want to be clear that I am not talking about 'audiophile' switches or any magic marketing lingo. Edit: that does not mean I do not believe they can rightfully exist.

Simply add a switch between your ISP router and computer rig and listen, or better stack two switches or (even better, far larger difference) use fiber optics ports found in many of the more serious switches (used, I got two for like 50 quid), using SFP modules like the Finisar 1321 BTL modules (cheap second hand will do, tenner each) with some bog standard LC LC OS cable ( plastic will do, another 15 or so) and listen switching back and forth.

I did the experiment, blind be it without randomized script etc, with a couple of friends and we were able to separate the various configurations by preference, with fiberoptic being the most preferred option. After that test I added the Xilinx Solarflare network card, eliminating one switch, that IMO has effects within the computer so let's leave that aside (still those bits and bytes are not arriving in any different order). I hear what you are saying about EMI and or RFI, a DAC does not get more vulnerable using I2S so the difference should sit in the front end of things. It's just that I have a hard time believing that the many ultra high end DACs folks busy tinkering with network components all are affected that much by EMI and RFI that should not exist to begin with (the much praised galavnic isolation in most UTP components).