[digititus]

IFI are a business. No problem with them making any product. To buy or not to buy is an individual choice.
I haven't read any of the marketing material and do not know what they have published to support this product, but in the interests of company PR, they should at least try and show some data which shows how "noise" has been reduced with some empirical data.
If this product reduces electrical noise, then I will give it a pass. If it is doing some kind of data manipulation via TCP/IP then I give it a big LOL. Each to their own.

 

[evhvis]

Seconding. I've studied Ethernet and TCP/IP extensively and taught it at college. It is not possible for the product that iFi is selling to change an audio stream in any way. It's just not possible, sorry.

Don't think anyone has claimed that it changes the bits or packages. It is about removing noise that is not part of the signal which can affect the component that is connected to the network. How much the effect is depends on the amount of isolation on the final device(s) in the audio chain, how sensitive the end component is to noise, how much self noise is generated in the receiver depending on the quality of the electrical signal and how effective the lan iSilencer is.

This is not so much about ethernet, and definitely not about TCP/IP, it is about noise generated in computer chips in the receiving end and noise that is capable of traveling over cables and the micro-transformers in a normal ethernet setup.

 

[Wakalaaa]

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

 

[vext01]

Sounds don’t travel down the cable, electrical impulses representing data does.
Can you clarify what you mean?

Is your question: how can you send digital signals over analog copper wires? I mean, that's the linchpin of all digital electronics.

it is about noise generated in computer chips

Yeah, but why pick on ethernet when there are lots of other more electromagnetically noisy components in a typical computer system? A PC's PSU throws out tons of EMF. A wireless networking radio is a device specifically designed to emit EMF. Why does EMF from those components rarely cause us problems? Because (all but the worst) modern computers are designed with this in mind and are really well shielded.

This is only a US$100 toy

I think $100 is a lot of money to throw at something which probably doesn't do anything.

(I'm ducking out of this conversation until someone does some quantitative testing of the thing. As I've said before, if someone designs a rigorous and fair experiment and shows that this device does something audible, I'll be the first to concede. Until then, I'm highly skeptical).

 

[Davesrose]

Seems those that swear an audiophile ethernet device does something to the sound also haven’t done a blind test (and score 10/10). Really don’t see how audio data is different than other data that’s transmitted (jitter and noise in ethernet is irrelevant with TCP/IP)

 

[AussieMick]

It’s not about the data, people. It’s not about the data. Please stop saying it is.

The electrical storm in a box that is a PC or laptop is definitely detrimental to audio production. Again, it’s not about data corruption. “…rarely cause us problems” needs a definition of what you mean by problems. Does the data get where it’s supposed to go? Yes it does. That’s not the problem.

 

[teknorob23]

preventing predominantly high frequency noise from entering your streamer and consequently your dac is the main challenge one needs to overcome in improving the Sq of network source audio reproduction. It’s nothing to do with changing the 0s and 1s, it’s about removing rfi which otherwise travels with the Ethernet signal that infects the ground plane of the steaming device via the rj45 connection. This is an issue with audio as well as some medical and military/naval applications, but obviously rf and emi on the ground plane won’t effect how your printer works

It’s not about the data, people. It’s not about the data. Please stop saying it is.

The electrical storm in a box that is a PC or laptop is definitely detrimental to audio production. Again, it’s not about data corruption. “…rarely cause us problems” needs a definition of what you mean by problems. Does the data get where it’s supposed to go? Yes it does. That’s not the problem.

totally agree, it nothing to do with changing 0s and 1s, it's about high frequency noise mainly picked from or emitted by active devices connected to your network, which piggy-backs the data stream into your streamer. The noise in the Ethernet signal is at a very high frequency, way above audio frequencies. This high frequency noise is able to easily pass through digital and audio circuitry , and it keeps on going, flooding the ground plane of equipment and passes on to all 'infect' other connected equipment. Ethernet born noise eventually ends up affecting the timing of the DAC, which introduces phase inconsistencies in the conversion to analogue process, the effects of which human ears seem to be particularly sensitive too.

 

[Davesrose]

It’s not about the data, people. It’s not about the data. Please stop saying it is.

The electrical storm in a box that is a PC or laptop is definitely detrimental to audio production. Again, it’s not about data corruption. “…rarely cause us problems” needs a definition of what you mean by problems. Does the data get where it’s supposed to go? Yes it does. That’s not the problem.

If this theoretical PC has so much internal noise to be detrimental to its audio, then how can an ethernet device improve its audio? The video I link explains how ethernet is reliable digital transmission, already has isolation against noise, and digital devices have inherent designs to reject noise. Amir found identical noise values in regular ethernet device vs audiophile at high frequencies. The only way you can settle things with skeptics is do a blind test with regular ethernet connection vs one of these devices.

 

[adrianm]

preventing predominantly high frequency noise from entering your streamer and consequently your dac is the main challenge one needs to overcome in improving the Sq of network source audio reproduction. It’s nothing to do with changing the 0s and 1s, it’s about removing rfi which otherwise travels with the Ethernet signal that infects the ground plane of the steaming device via the rj45 connection. This is an issue with audio as well as some medical and military/naval applications, but obviously rf and emi on the ground plane won’t effect how your printer works


totally agree, it nothing to do with changing 0s and 1s, it's about high frequency noise mainly picked from or emitted by active devices connected to your network, which piggy-backs the data stream into your streamer. The noise in the Ethernet signal is at a very high frequency, way above audio frequencies. This high frequency noise is able to easily pass through digital and audio circuitry , and it keeps on going, flooding the ground plane of equipment and passes on to all 'infect' other connected equipment. Ethernet born noise eventually ends up affecting the timing of the DAC, which introduces phase inconsistencies in the conversion to analogue process, the effects of which human ears seem to be particularly sensitive too.

To this end, wouldn't it be preferable to just convert Ethernet to Optical and back to Ethernet? Using a clean power source for the converter ofc. Sure it's messier, but seems more effective.

 

[teknorob23]

To this end, wouldn't it be preferable to just convert Ethernet to Optical and back to Ethernet? Using a clean power source for the converter ofc. Sure it's messier, but seems more effective.

The problem is, that optical conversion is active and no matter how quiet its power supply the process emits high frequency noise of its own.

 

[adrianm]

The problem is, that optical conversion is active and no matter how quiet its power supply the process emits high frequency noise of its own.

Out of curiosity, what's your stance on Ethernet cables?

 

[teknorob23]

Out of curiosity, what's your stance on Ethernet cables?

in what sense?

 

[adrianm]

in what sense?

Well a lot of audio companies sell extra shielded ethernet cables, but that shielding basically acts as an antenna, defeating the purpose of galvanic Isolation. In cases like that, I imagine a device like this would make even more of a difference.
Curious if you see this as an issue or if there's a legitimate reason to recommend a shielded cable.

 

[teknorob23]

Well a lot of audio companies sell extra shielded ethernet cables, but that shielding basically acts as an antenna, defeating the purpose of galvanic Isolation. In cases like that, I imagine a device like this would make even more of a difference.
Curious if you see this as an issue or if there's a legitimate reason to recommend a shielded cable.

thank you for elaborating. i can only speak to my own experience where i've found that shielding is not needed and i can make a better performing cable without one. Choice and mix of dielectrics and of course the conductor Shielding makes sense when running vast lengths as it was designed to do industrial settings, but unless you live under power lines or your running 100s of metres of cable, theres not real needed for streaming.

 

[AussieMick]

If this theoretical PC has so much internal noise to be detrimental to its audio, then how can an ethernet device improve its audio? The video I link explains how ethernet is reliable digital transmission, already has isolation against noise, and digital devices have inherent designs to reject noise. Amir found identical noise values in regular ethernet device vs audiophile at high frequencies. The only way you can settle things with skeptics is do a blind test with regular ethernet connection vs one of these devices.

I was just using a PC as an example of an electrically noisy environment diminishing sound quality. With a PC or laptop I find some kind of filter on the outgoing USB to the DAC of significant help. I don’t know how much a filter on Ethernet into a PC would help, I’ve never tried.
However, I abandoned PC and laptop a while ago for a dedicated solution from Innuos. Extremely low noise such that filters on the output made no difference. To feed it from the wall I use an aftermarket AudioQuest Vodka Ethernet cable that sounds obviously different to generic CAT6 and 7 cables I have and also different to to a Cardas Clear Ethernet cable I have. Of those different flavours I prefer the AudioQuest.
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.