How can you have jitter reduction in a store and forward Ethernet switch that is connected to another Ethernet PHY or switch that you have no control over or know the properties of?
Store and forward kind of gives a bit of a clue.
Ethernet devices need a tight clock to even work. That gets the physical layer 1 working. Anything tighter than 25ppm is a waste of time as you are also reliant on the link partner.
The whole argument is a total and utter waste of time. Snake oil.
I thought we had finished here. ā¦and it gets dragged up again 
Itās worth keeping the USB cable short and get a decent one (shielded) as HS USB uses signalling levels of about 400mV, but I wouldnāt worry about Ethernet cabling so long as you donāt get packet loss.
Until somebody really puts it in audio system and hears difference.
Trust real goals must be hidden.
Someone needs to save the poor souls from the evil, right? They cannot let this happen
Iāts not possible to say how much a switch and ethernet cables affects your system, but here are some things that can be affected that might (and often will) affect sound quality at the end:
Looking at this, its not so hard to understand why a switch with linear electronic and accurate clocks actually can (and will) sound better in most cases.
I have myself tried using ethernet straight from ānormalā switch, using a UpTone HiFi switch, using fiber and fiber media converter and using an extra Pink Faun LAN isolator, and in my system they all sound different. I rank them as follow (from best to worst), but these are subjective opinions and other might rank them differently:
Once you hear differences from an HiFi equipment, and can explain those differences with science (in this case quite easy science), its hardly snake oil. But having said that, networking improvements does not give the same type of benefits as for example better speakers, DAC or amp does.
Technology moves forward; so it is with developing switches. ethernet cables, etc for audio. At one time lamp cord from the hardware store was considered good enough for speaker cable. I think weāve moved on from that.
Thatās true.
OTOH, per Kurt at BJC, the only thing that is important in speaker wire is gauge, any other selling point is snake oil.
OTOH if I donāt run the right kind of cable with my vintage Naim amp, it could blow up!
LOL! You realize that famous āKurtā, the man, the myth, the legend, is there to sell you his cables. That is the Blue Jeans cables. What do you expect him to say? When I go to buy a Toyota, I fully expect the Toyota salesman to pitch me on a Toyota. Not on an Audi or BMW. I would be very disappointed if he did not pitch me on a Toyota 
He makes cables. He can make any cable he wants, for any price he wants.
Yeahā¦ sure. He can certainly walk on water
I uses Flymo cable for speakers cable years ago to great effect and it sounded great. Iād almost say it was tuned to the system. I loved the orange colour too.
Iād need to see so much more detail backing up all those generalisation Iām afraid. A decent endpoint DAC should eliminate all of the theoretical issues.
Thatās what designers do and what we pay for.
If you are talking about the nth degree of subtle slight difference that you have to sit for hours comparing to notice, then I askā¦ Whatās the point? Buy more music, go to gigs (when we can) and enjoy more wine (add drink of choice here)
Personally i am not an audio designer and have no interest in being one. Thatās what I pay system designers for. Decent auditions over the years means I personally donāt yearn for this constant search for improvement. Iām already there.
The topic intrigues and baffles me in equal measure. Iām intrigued to know, when will it be done, what then?
At this rate, it will never be done and all the time we are left feeling dissatisfied with the incredible systems most of us have already, life passes and we forget to enjoy the music.
Thatās correct, a āperfectā DAC would only need the bits and sound the same no matter what comes before it. To bad it donāt exist!
Personally, I have a RME ADi-2 DAC and a mR (both run on LiFePO4 20Ah batteries) and in my case switch and other networking tweaks does matter. But how much you hear from tweaks like that also depends on how transparent your system it, and how good room acoustics you have.
I also did tests like you have. I also like you found that everything I tested sounded different. Some I liked and some I didnāt.
My best set up so far is using two switches (cisco) second hand from Ebay. and a wireless modem supplied bu EE (UK).
Fiber from modem to cisco 1 close to server & windows pc with Roon server. Cisco 2 by end point and 40 meters between 1&2 of cat 6ā¦ Sounds great.
Some of these subtle differences are a little more profound than a quick listen reveals. For some people half the fun is tinkering, half actually listening to music.
Plenty of people over on the Naim forum hearing differences from switches and cables with their āpoorly designedā ND555 $20k streamers. Maybe itās the other way around - the differences canāt be heard if youāre satisfied with a cheap Chinese made DAC that one thinks itās all they need. Anyway, others gain in audio satisfaction, your loss. But we all have different priorities in life so letās leave it at that.
Well my poorly made Meridian DSPSE set up sounds fine as is. I am at a stage where curiosity for improvements has just gone. Iām off the upgrade train and itās a great place to be.
All the changes I hear are through Meridian SEs, expensive switches improve the SQ of the 218, fibre optic also makes a difference, endpoint power suppliesā¦Yes they sound fine, I want them to sound finest!