Testing Ethernet switch


If you have bought an "audio" Ethernet switch, don't bother with this thread 

If you question Ethernet switches, here is one test of one brand. 

Search You-Tube   Linus Tech Tips  Aqvox

https://www.youtube.com/watch?v=NMFQ3YvR3Eo&t=914s

 

tvrgeek

Simpleton Linus should stick to building computers

Watch these measurements/tests from alpha audio and get educated on this topic a bit more. Yes, they do have audible differences.

Youtube Video - Does a Bad Network Switch Ruin Your Streamer?

If you don’t have a highly resolving room+rig, don’t bother, pat yourself on the back after watching that Linus video, get any crap switch and enjoy the music.

Nevertheless, an audiophile grade switch need not be as expensive at all. That’s the doing of the vultures in this industry.

Deep, you make an extraordinary claim with nothing but your belief and some other you-tuber who has no evidence either.  Extraordinary claims need extraordinary evidence. 

As I know how Ethernet works, I know of no way using the science of this universe that suggests layer 1 or 2 can have any effect on your system sound. A poorly designed power supply may transfer some noise, but as that is inherent, a client device that does not deal with that is a bad design. Don't blame the switch. An overloaded network could starve a path of data as IP is not a real time protocol, but a managed switch can mitigate that as well by using QoS.   A defective switch could cause high error rates, but that is not what this is talking about.  IP ( Internet Protocol) will completely resolve any errors if using TCP as it is a guaranteed delivery. If using UDP, a best effort, well you are not attempting a robust system and any error is your fault and the switch just reads the packets, figures out their destination and reforms packets to send to the correct client. Bits is bits. There is no level, slope, timing, jitter, or magical influences. 

If you have any concept of how a switch or cable ( often talked about magic) can effect your sound, please present it.  We use Ethernet for life-threat, live-fire, banking ,and nuclear control so if you have discovered some mysterious error, please let it be known. Until it leaves the client buffers in order and stripped of the headers, bits is bits.  It has been that way for 30 years and overseen by engineers far smarter than any of us. You do know packets don't even have to arrive in order don't you? 

I do agree, the scam of these expensive switches that are just a couple better power supply caps and some encapsulation is taking advantage of folks looking for magic who do not have the technical background. Not everyone does. Last time I looked you don't need a class in networking to be a brain surgeon. 

Again, our brains lie to us. They hate to admit we have been scamed and will reinforce belief harder and harder.  Sound is real. What we hear is the lie our brain tells us.  If you believe it sounds better, your brain will convince you it does. Just it has no bearing on the actual sound. 

There are a lot of improvements we can do to our systems. Some well known, some innovative. Some plausible but ineffective. Then there are those that are impossible. Where should one spend their time and money? Possible or impossible. 

If you believe in the impossible and want the very best, you can always buy a Cisco Catalyst.  I'm sticking with my $35 Netgear. 360 Mb/s download to any of my clients.  IGb/s aggregate client to client as my internet is only guaranteed to 300Mb/s.  It can stream 4K video without error and streams radio to my PC streamer just fine. 

Don't believe me. I did not invent the Internet. ( actually I do know someone who did invent some of the IP stack. He was a DoD contractor I supported when I was a systems admin)  Please go read about how it works. Look at what each of the seven layers of the IP stack do. Then explain how anything but an actual dropout can effect your sound.