No facts or testing are valid but at what point is hearing not the real point in this debate.
I mean no one here harm but I will be honest in my comments
I have tried all permutations of playback on my system and they are all identical. Local USB disk playback, wired, wired over powerline and wireless. I recently ditched powerlines across my network and upgraded my WiFi to use Ubiquiti Unifi access points as its a vast improvement over domestic WiFi kit. No difference in SQ but it’s rock solid.
1 Like
Not sure if it matters to anyone - but I have a 3 node Google wifi mesh network and 6 Google mini’s, a couple of chromecast audios (one plugged into a DAC via Optical) a JBL Link 20 portable chromecast speaker… and… I can stream 24bit/96k to all of them in one cast group from Roon with no dropouts and all in perfect Sync… Welcome to 2018. 
How you might ask? Using an Android tablet with system wide casting turned on.
Android tablet - Settings > Display > Cast > select cast device or group
Fire up Roon on the tablet and set zone to local tablet… miracle of miracles… Roon audio to whole house via Chromecast in perfect sync @ 96/24.
3 Likes
The thing with hearing is that it’s not at all what it’s cracked up to be. There are numerous scientific tests that show a preference for sound that’s as little as 0.1dB louder, where we reliably detect changes of 0.5dB only.
Any hearing based testing is therefore quite difficult to accomplish correctly, and one of the more error prone ways to establish the impact of anything in an audio system, where the changes are not immediately obvious.
When the claim is that a system “more resolving” is needed to detect the impact of wired versus wireless transmission of audio data, which has exactly zero scientific foundation (with RAAT), it just makes the whole thing very magic, because the claim is essentially that unmeasurable changes yield audible effects.
I don’t suppose you see the difference on the TV screen between Netflix streamed over wireless or wired, or can tell if a photo on your computer was downloaded via WiFi? It’s the same thing really (with the caveat I’ve mentioned above about Airplay - see GitHub - shivasiddharth/shairport-sync: AirPlay audio player. Shairport Sync adds multi-room capability with Audio Synchronisation under Synchronisation, Latency, “Stuffing” - and this only applies in certain situations).
1 Like
Very good points indeed. And yes WiFi stinks at times on my tv when the bad with drops lol. But you’re point is solid
But having said this truthfully I can tell reliable but there is so many other magical objects in Audio. And I’ll bet in general a well setup WiFi does sound well. But for me I would never use it for a only use purpose
I keep my Audio network isolated from the rest of the home with its own port setup dhcp and ip config this is a must for all as well
If what’s digitally sent is is the same as what’s digitally received, which an underlying protocol ensures, the perceived SQ would be exactly the same between magical carrier pigeons and the most expensive Ethernet cable in the world.
On a reliable network with speed and capacity, wired or wireless, the digital zeroes and ones on one network are the exact same zeroes and ones as another.
This Rule is nonsense. My wifi tests consistently faster than wired Ethernet. You just need a good connection. I listen to tidal high resolution all day long - never had a drop out.
I would say it is best to have the core connected to the DAC given that there is little to no buffering on that side of things.
I think many would disagree with this. Once concern is that the audio processing in the core pushes the CPU (etc.) to generate potential EMF/electrical noise that can pollute the USB connection to the DAC.
Thus, the idea is that pushing the wave file across the network (wired or wireless is just a connection quality question) to an endpoint that is doing nothing but passing on the wave contents to the DAC, and is not thus generating so much potential electrical noise, eliminates that concern.
1 Like
If the DAC is not well designed and well built then I would agree. I think a good DAC should not need special soft glove treatment. I agree that many DACs are disappointing in this regard particularly with a USB connection, but the most recent round is getting way better.
I don’t agree that you should change the entire architecture because of inadequate noise isolation in some DACs but I understand that is what many must do to actually get the performance they were promised.
Which new DACs would you recommend as having good immune USB connection?
Isn’t one the Nucleus’ SP that it can to be part of the kit in hifi rack? So this would be exception to this rule?
Check those reviewed by Stereophile. They do the standard AES jitter J test which is quite challenging. If a device can pass that test with flying colours then any jitter is most likely inaudible with ordinary music.
Jitter isn’t electromagnetic interference, which is what’s being discussed here.
Why do you feel the need to convince or educate people? People can listen and make their own minds up, without your assistance.
I recommend Ethernet for use with our sonicTransporter Roon servers and our Rendu Roon Ready players for one reason. It always works!
Even cheap WiFi routers work great sometimes. And that’s the key word sometimes. WiFi can work great for months then your neighbor get a new wifi router running on the same frequency as yours and BOOM you start getting dropouts in your music (and you have no idea why)
Those mesh also work really well most of the time and have much better throughput then cheap wifi but again if they fail they are complex so it can take a tech support call to the router manufacture to get them working again.
My recommendation is use Ethernet if you can. If you are worried about any noise over Ethernet use optical networking and if you really need WiFi get a good mesh network.
6 Likes
Recently, SOtM has released a firmware update for its SMS-200 device, which implemented network switching 100/1000 with a recommendation to use mode 100 for better sound quality. What do you think about it?
“Also we tested when setting network speed to 100BASE-T, there was certain sound improvement compared to when setting network speed to 1000 BASE-T, so we’ve added a function to be selected the network speed between 100Base-T and 1000Base-T and it is to make the better sound quality according to the system and network environment. But when upgrading the system or installing or uninstalling the player app, the 100BASE-T network speed are not enough and it could bring malfunction with 100BASE-T speed. So when upgrading the system or installing or uninstalling the player apps, please change the network speed to 1000BASE-T. And the Librespot and HQplayer NAA should be uninstalled first and then install it again to have the latest version.”
1 Like
That’s interesting. A few years ago a guy who sold 100Mbps audiophile switches with modified clocks also said 100Mbps sounded better than 1000Mbps, but later his switches also become 1000Mbps.
I guess it all depends on the situation wether WiFi outperforms Ethernet or visa versa. In my particular setup I have invested quite some money and time to optimize my network and WiFi. I use an Auralic Aries G1 which is (according to the manufacturer) optimized for using WiFi. I did some A/B testing and prefer WiFi over Ethernet in my setup. Using WiFi enables way more micro details to be audible. Also thanks to a stable network I have yet to encounter dropouts or other WiFi related problems.
YMMV though and I can totally understand that there are many situations (home wise and hardware wise) where Ethernet outperforms WiFi.
Sorry, it turned out two discussions:
1 - ethernet vs wi-fi;
2 - ethernet 100 vs ethernet 1000.
Think of it like a pipe. With ethernet, the pipe is always the same fixed diameter. With WiFi, your pipe is constantly changing diameter, multiple times a second, sometimes its bigger than a ethernet pipe, sometimes its smaller. Sometimes its much smaller, or even pinched off completely. Now think of the flow of water coming out of the wifi pipe vs the ethernet. It will be stopping and starting, spurts and surges of flow, then nothing. The flow out of the ethernet pipe will be a constant steady stream.
Ethernet is a reliable and dedicated link. Wifi is very much a shared medium subject to all sorts of influencing factors that change its performance and availability. If your wifi works, then great, you got lucky, that’s all. Ethernet will work as long as it’s plugged in.
Regarding the 100 vs 1000 issue, ropieee sets the lan port to 100 vs 1000 on the rPi3B+ because it causes issues to the USB controller when running at GigE.
100Mbps is at least an order of magnitude greater than the needed capacity to stream hi resolution audio, two orders of magnitude just aren’t needed.