AQVOX audio grade switch

4 hours ago, The Computer Audiophile said:

If you want to pay for the empirical testing, the testing can be done at my place. It will take about 30 seconds for me to show the results that a deaf person could hear. 

I'm sure that the SQ changes will be obvious in your rig, because, it is a "revealing setup". Said with zero sarcasm, because I have heard how the craziest things can alter what one hears, profoundly, once a certain standard is achieved. But the next step is, what is causing the variation?

You believe it's the USB C to Ethernet adapter, and you're highly likely correct. But this doesn't mean it's sending corrupt data; rather, in some fashion it's disturbing the waveforms in some area of your system to a degree that the conversion to analogue or following circuitry is impacted enough to be quite audible. The why this should be so is just electrical engineering not done well enough, somewhere ... but what's really important is the how to fix it. I tend to work on the principle of letting areas that by nature are electrically messy just do their thing - and make the circuitry that really counts for getting SQ right, so robust that it can ignore the noise, wherever it comes from. So, say the server and the adapter are really badly behaved - but still work, as electrical devices ... okay, leave them be, and just isolate their crappiness by buffering, regenerating, converting, filtering, whatever, until you get clean signal to work with - this also gives a head start if you change anything in the messy areas, and upgrade, etc.

4 hours ago, fas42 said:

The why this should be so is just electrical engineering not done well enough, somewhere

In my case, it’s a combination of hardware and software causing it. 

12 hours ago, The Computer Audiophile said:

I think we are talking about something different. I’m totally with you on your statements. The issues I’m talking about involve network audio, but I don’t believe they involve something fixable by a network device. 
 

This is a tough discussion because we don’t have a concrete example the problem, at least what I’m talking about, caused at the music server. I’ve heard it, but I can’t say, do XYZ and you’ll hear it. 
 

Would make for a great discussion in person with hardware to test!

Protocols might matter too. I know that I noticed increased latency when using USB hard drives — years ago because I don’t use them anymore for this reason. I used to run latency and throughput tests on various network file protocols including AppleTalk — which at one time was actually great but gains in SMB have made SMB super efficient for file access, better than NFS… a lot of that can be hardware dependent and do doubt the USB protocol might mess things up in terms of packet size/timing — and thunderbolt 4 which uses the same cable as usb4 — wouldn’t be expected to have that problem. 
 

If the SQ issue goes away using Thunderbolt Ethernet adapter that would lend support to the idea that it’s a packet timing issue.

There other place where these issue are apparent are with video eg via a blu-ray player vs via a video library. The blu-ray players are clearly optimized to deliver smooth video — and they are dirt cheap without fancy power supplies ;)