Ethernet switches for audio - Part B: why a regular switch will suffice.

[dbastin]

This thread is complimentry to another I created ...

Ethernet switches for audio - Part A: List of switches, related info & experiences

This Part B  is where people can share their views about why ethernet switches specfically made for audio should not make any difference to sound quality, and also the other side, why they should and do.

 

This thread is a place for these debates to occur and therefore minimise these debates cluttering up other threads, such as Part A.

 

Posts in this discussion may suggest or recommend particular switches, and discuss pro/cons of upgraded power supplies, power cables, grounding, optimised settings, tweaks, modifications, etc - but not ethernet cables. (there's another thread discussion about cables).

 

If you have compared with a switch listed in Part A, please share your experiences.

 

This has been restored from the Great Debate sub forum and its close monitoring by moderators. It was put there due to an incident some time ago, but we've been well behaved since then.

 

Please be civil - so this doesn't go back to the Great Debate, or shut down.

Edited May 15, 2020 by dbastin
clarifications to and enhancements of original post

[LogicprObe]

 

Well, you really want as little delay as possible for audio.

[Guest rmpfyf]

Because ultimately the only thing a nice switch can do, if you assume the data is all getting there, is not being any crap into the PC and have the interrupts generate packets nicely in a periodic way, or in another way that doesn't ass about with music playback.

 

Optical Ethernet handles the first point if it's a super issue (or decent isolation).

 

The latter can be made better by moving Ethernet interrupt and music playback to different CPU cores where resources exist, or using a good or tuned NIC. A mega switch is just a band-aid here.

 

And proper buffering and reclocking at playback makes all irrelevant where it exists.

 

This all assumes Ethernet during playback is needed - where we can do without, solutions are simple

[Decky]

SO, what is the "audiophiliac" assumption here - that switch data transfer can be slower than the DAC processing speed? A better PSU - sure - OK - another 5V or 12 V linear PSU will not hurt anyone but even that - how will that exactly influence audio processing downstream? That is not even taking into account capabilities of modern FPGA chips with sizable DSP trickery on their disposal. 

 

And yes, I have read quite a bit about AQ-SWITCH SE and I just do not buy into their pseudo-scientific bedtime story. I put it on the same shelf as my shakti stones and quantum fart purifiers. Unless....they sell it with a blue laser LED light - then I will spend $1000. 

Edited December 22, 2019 by Decky

[Guest]

15 minutes ago, Decky said:

SO, what is the "audiophiliac" assumption here - that switch data transfer can be slower than the DAC processing speed? A better PSU - sure - OK - another 5V or 12 V linear PSU will not hurt anyone but even that - how will that exactly influence audio processing downstream? That is not even taking into account capabilities of modern FPGA chips with sizable DSP trickery on their disposal. 

 

And yes, I have read quite a bit about AQ-SWITCH SE and I just do not buy into their pseudo-scientific bedtime story. I put it on the same shelf as my shakti stones and quantum fart purifiers. Unless....they sell it with a blue laser LED light - then I will spend $1000. 

 

Don't, just dont

[Guest]

OSI model. That's all I'm saying fellas.

 

 

[Guest rmpfyf]

2 hours ago, TDK said:

OSI model. That's all I'm saying fellas.

 

If you assume the data gets there - which is completely reasonable to assume - the rest is timing at the output.

 

Ethernet just does it's thing IMHO. Everything audiophile to these ends is a band-aid for a deficiency upstream. Some nicer and more effective than others, but band-aids nonetheless.

 

4 hours ago, Decky said:

SO, what is the "audiophiliac" assumption here - that switch data transfer can be slower than the DAC processing speed? 

 

No.

 

Though your coda had me remembering times when the cool kids would spend $100 getting their Nokia's outfitted with blue LEDs. Good times.

[BigJim]

Part B: why a regular switch will suffice: Perhaps if you have a highly resolving system and you can hear the supposed difference an audiophile switch makes, then go for it. My streamer cost around $500, I doubt a switch that costs more than that is going to make a difference to my old ears. There's always going to be something better, newer and more shiny. I'd rather enjoy some more music.

 

[F18]

A switch physically itself won't affect the sound quality at all. This is down to the physics of how switches work at a logical software level and that you're transmitting digital data 0's and 1's. If the packet fails the transmission it gets re-transmitted. If anything the only way I can see it may improve delays etc. is setting up your own QoS and traffic rules on a managed switch. This is why I group Ethernet cables into the same category, they don't do anything besides assisting the above. The only way a switch is going to make a difference is if the designer incorporates there own set of protocols of how digital network transmission works - which nobody in the audio industry has the knowledge to.

Edited December 22, 2019 by F18

[Guest rmpfyf]

I love the ongoing confusion around data getting there and sq 

[Brett1968]

I think there is a lot of confusion and misinformation about what in the telecoms world is called network synchronisation.  ie. all oscillators in the network are synced to a master oscillator.

 

In the context we are talking about for a home network I don't think it really matters that much in regards to network transmission, but what is important is having a digital to analog converter that is electrically isolated and has a stable and accurate oscillator.

 

Are there any DAC's that can take an external clock input?

[Ittaku]

8 minutes ago, Brett1968 said:

Are there any DAC's that can take an external clock input?

Plenty, but what has that got to do with ethernet? The timing on ethernet switches is wildly different to anything used on the DAC.

[Guest rmpfyf]

1 minute ago, Ittaku said:

Plenty, but what has that got to do with ethernet? The timing on ethernet switches is wildly different to anything used on the DAC.

Insisting on this repeatedly doesn't make it a universal truth.

[Ittaku]

24 minutes ago, rmpfyf said:

Insisting on this repeatedly doesn't make it a universal truth.

Insisting it's not also does not.

[Guest rmpfyf]

Just now, Ittaku said:

Insisting it's not also does not.

 

I see what you did there though I don't insist. If you can pull a cable and hear no difference, any amount of Ethernet upgrades won't yada yada. My refrain here is consistent and not insistent as you suggest.

 

Plenty of people here suggest they hear differences. I err on the side of respecting as much. 

 

Understanding how Ethernet can affect SQ involves more than simply understanding Ethernet.

[Ittaku]

1 minute ago, rmpfyf said:

 

I see what you did there though I don't insist. If you can pull a cable and hear no difference, any amount of Ethernet upgrades won't yada yada. My refrain here is consistent and not insistent as you suggest.

 

Plenty of people here suggest they hear differences. I err on the side of respecting as much. 

 

Understanding how Ethernet can affect SQ involves more than simply understanding Ethernet.

But you are talking about something completely different. You are saying that the ethernet has an influence on the sound quality. Where did I dispute that in what I said about timing?

[Guest rmpfyf]

11 minutes ago, Ittaku said:

But you are talking about something completely different. You are saying that the ethernet has an influence on the sound quality. Where did I dispute that in what I said about timing?

 

43 minutes ago, Ittaku said:

Plenty, but what has that got to do with ethernet? The timing on ethernet switches is wildly different to anything used on the DAC.

 

 

Wildly different, but not unrelated.

[Ittaku]

2 minutes ago, rmpfyf said:

Wildly different, but not unrelated.

The question I was answering was whether DACs take external clocks. I hate to get this pedantic... I knew I shouldn't have even chimed in on this thread. I even tried to steer clear of making reference to sound quality effects...

Edited December 23, 2019 by Ittaku

[Guest rmpfyf]

3 hours ago, Ittaku said:

The question I was answering was whether DACs take external clocks. I hate to get this pedantic... I knew I shouldn't have even chimed in on this thread. I even tried to steer clear of making reference to sound quality effects...

 

Suggesting that your answer were so specific would be being very economical with truth

[Brett1968]

4 hours ago, Ittaku said:

The timing on ethernet switches is wildly different to anything used on the DAC.

How Is it different?  Aren't they both oscillators that should run at a stable know frequency?

[dbastin]

On 23/12/2019 at 4:34 AM, F18 said:

The only way a switch is going to make a difference is if the designer incorporates there own set of protocols of how digital network transmission works - which nobody in the audio industry has the knowledge to.

I'd speculate that some companies making switches for optimal audio have or obtain the expertise to achieve that objective.

 

Take Melco/Buffalo for instance ... check out what they're in to ... https://en.m.wikipedia.org/wiki/Melco

 

I would think, aside from tweaking existing switches, development of a switch optimised for audio would involve experts in all its aspects.

Edited January 6, 2020 by dbastin

[Guest rmpfyf]

37 minutes ago, dbastin said:

I'd speculate that some companies making switches for optimal audio have or obtain the expertise to achieve that objective.

 

Take Melco/Buffalo for instance ... check out what they're in to ... https://en.m.wikipedia.org/wiki/Melco

 

I would think, aside from tweaking existing switches, development of a switch optimised for audio would involve experts its all aspects.

 

Would think the next stage is a NIC for audiophile purposes. 

[Ittaku]

1 hour ago, rmpfyf said:

Suggesting that your answer were so specific would be being very economical with truth

You almost seem smug in your attempt to make me spell out what I believe, when what I said was just a string of facts - ethernet clocks are unrelated to DAC clocks and one never drives the other. You know how I feel about ethernet and hifi and are intent on trying to extract it from me here based on that comment, but I have no interest in being dragged down into that debate. Stick to what I said, and try to find any science or evidence that ethernet clocks have anything to do with DAC clocks instead.

[PositivelyMusicallyGeared]

If there is a problem, it needs to be clearly defined first before any answer should be attempted.

 

It is an almost impossible task to answer this broad question in order to convince - it is even futile as it is required of the to-be-convinced to understand the OSI model and relevant network protocols first.

[Guest]

22 hours ago, rmpfyf said:

Ethernet just does it's thing IMHO. Everything audiophile to these ends is a band-aid for a deficiency upstream. Some nicer and more effective than others, but band-aids nonetheless.

 

 

What ethernet does is not down to opinion I'm afraid!