Melco £1999 ethernet switch

Precisely!

You are wasting your time here. Not worth it. Leave it alone. Not worth your mental well-being

Has anyone tried listening through a 10,000 rpm hard drive? It increases the pitch quite a bit but adds a lot more energy and dynamics. All slower rpm hard disks just lack the punch for me now. Don’t get me started on SSD’s! Far too digital for my liking!

3 Likes

Ethernet is a transport protocol, covering the data link layer. It includes error detection/correction. Everything is transported bit perfect. No alterations.

An Ethernet switch is not altering the signal (ie. binary data) it is transporting, which could be anything: Data about sound or spam email.

You don’t need to listen to and try an Ethernet switch to tell it is affecting the SQ in any way. Because that is how Ethernet works.

3 Likes

Add to that: The base clock is 25MHz. Way outside human hearing. UTP cabling means that it’s ground isolated.

Siemons put out a paper called the ‘Antenna Myth’ and T.I. has an excellent white paper on 10/100 Ethernet PHY’s and radiated emissions.

I don’t need to purchase every hammer on the market to know it’s going to hurt when I smack my thumb with it.

1 Like

Ethernet switches are trivial to blind test.

If you don’t already have one, buy a $15 Gigabit switch from Amazon. Hook both it and the EtherRegen up to your home network, placing them both behind a curtain, so that you can’t see which one is hooked up to your Naim streamer.

After familiarizing yourself with how each of them “sounds”, have a friend

  • flip a coin behind the curtain
  • unplug the ethernet cable from the switch it’s plugged into and
    • if the coin came up heads, plug it into the $650 switch
    • if it came up tails, plug it into the $15 switch
  • record which one it got plugged into on a sheet of paper

Meanwhile, you record on your sheet of paper which switch you think is hooked up. When you’re done, tell your friend to flip the coin again.

Repeat this 20 times. Then compare the two sheets of paper.

I think we both know what the result will be.

7 Likes

I don’t need to stand on the moon by myself to know that the earth is not flat. Please stop blaming well educated people because they don`t believe in audiophool voodoo.

5 Likes

I copied a Word document from my file server to my PC the other day. When I opened it on my PC my writing had lost some impact and clarity and air between the words. Will this switch fix that?

5 Likes

I can find a 1000 people that think the Earth is flat.

2 Likes

As I see it, it’s mainly because of “answers” like this that you’re getting what you perceive as “a steam of bile”. Your strategy of deliberately misunderstanding well-written posts is highly unfair. (You knew full well @hwz was talking about the Melco switch! Why didn’t you respond to his [IMO very convincing] line of argument?)

Why are you “sharing” your “perspective” if you don’t want others to comment on it? This is a thread about whether super expensive ethernet switches make sense or not. Seriously, what were you expecting?

Nobody in this thread has ever claimed that pasta and lobster taste exactly the same…:wink:

6 Likes

I like turtle

Nor has anyone here ever claimed that all ethernet switches taste the same…

Keep your fish, I’ll have RibEye.

2 Likes

I wonder what will happen when Audio Precision or someone can measure these things… :grinning:

Thanks and it’s even EASIER than that… Go ahead and get the cheap $15 GBe switch from Amazon.

Connect the NAIM streamer to it. I believe NAIM can buffer something like 30-60 seconds of 16/44.1.

Simply have someone plug / un-plug while the participant is blind to this fact.

If they can’t tell when the cable is unplugged then you don’t even need to bother with getting a fancy switch.

Yeah, but no true audiophile would sully their rig with anything less that 24/192. :slight_smile:

With my (slightly more involved) test, the true audiophile can use as high-res a source as he wants, and take as much time as he needs, to identify the switch being used.

This point I keep reading about buffers - when you unplug an ethernet cable/switch etc… as a valid test to tell these differences or bunkum. Once you unplug the cable/disconnect the switch - its immaterial about the network at that point as the data/music bits are already in the buffer of the dac/streamer aren’t they?
Is the valid test, once the buffer stops, to then swap what ever cable device or not at that point - i.e. start playing the music again and see if you can hear any difference?

1 Like

I don’t believe your logic work for this. In the case of one manufacturer they speak about ‘leakage currents’. They are referring to a pervasive and constant analog effect making it over the cable and introducing ‘phase noise’ or ‘clock float’ of the DAC section.

So this is prevalent only when the cable is plugged in. So it doesn’t matter if you have 60 seconds of buffered audio (some streamers) or the entire track buffered like you can in JRiver.

Just by virtue of being plugged in you are compromised. They will even maintain that if you are playing a file off of local non-volatile storage that you are still compromised if any data is going over the Ethernet port because they can’t argue that leakage current only exist when you are transmitting data that contains audio vs data that contains something like a 4GB ISO or Family Photos.

So, I guess the noise is not part of the data stream, it’s in addition to the data stream.


Correct. If a manufacturer made the argument that noise is in the data they would get denounced even more voraciously then they currently do.

So they have to argue the layer 1 of the OSI model angle and fortunately with buffered systems it’s easy enough as I showed in my video in another thread here to create redundant links that allow us to make swaps with non-stop/hit-less fail over.