Sure, I suspect you clearly understand, but reading the thread there is confusion elsewhere, so as I usually do I’d thought I’d try and help out and explain at a high level what does actually happen. You can’t reply easily to a group of messages.
I actually found the discussion on groove cutting / modulation on records interesting… we seem to disproportionately talk about data networks with streaming and many data or networking concepts get rather mixed up and confused. It’s nice to shift the discussion to a possibly equivalent topic for turn tables and that is groove modulation, channel separation and cartridge tracking for records. We swap network engineering with mechanical engineering.
The issue with using a wifi connection is that you are trying to avoid introducing RF noise in to the system, so using RF as your network link is going to undo that - it will actually exacerbate the problem!
Wifi works in the GHz frequency band, and its power levels are quite benign, and just one in a multitude of electromagnetic waves around us. But of course, there is always something one can be concerned about.
In a specific device, wifi may possibly indeed be worse, but then it would still be one possible solution to the cable problem if it was implemented in a suitable way.
I really think it’s unlikely. Mobile phones in the room are probably more affecting with respect to WiFi.
Sure a WiFi RF unit can provide a varying load to the powersupply, and if not carefully decoupled in a poor design could audible artefacts.
I would say fast Ethernet is more liable to cause issues as it has an effective carrier around 31 MHz which is more likely to cause feedback interference in audio circuit componentry.
This isn’t conjecture, Audioquest conducted testing and were able to definitively show an improvement. Simple enough to test this yourself.
For Ethernet (referring to 1Gbps and greater) there is no carrier frequency, data is pulse-train. I can provide a technical reference showing the power spectrum if you need it.
Not necessarily so. I tested my Atom with a few tracks a few years back via WiFi, wired and local via USB stick. Played them through and recorded the output via the pre outs into a digital recorder.
I then edited the different leading silences for the tracks so I could have them all in sync in my audio editing softwares timeline. Then applied a sum and difference to them and it showed zero difference between all 3 versions of the same tracks.
I went one step further and adjusted phase on one version so it was 180 degrees inverted and thus out of phase and played it in sync with one of the others that was in phase. I got silence which is what I would expect as they cancel each other out. Any differences would be audible. I then did this with all the files one by one comparing each to the other. Zero difference.
Your interpretation is entirely up to you, all we are doing here is ensuring the correct information is disseminated and I can assure you as an engineer that passing a signal via RF is not a buffer to RF interference.
Optical isolation perhaps, albeit that’s another rabbithole.
Can you define the test & the test equipment used by AQ.
Continuity testing is easy & all ethernet installers do this.
Whereas conformance testing requires special test equipment
I would be great, and as an engineer should not be too hard, if you stuck to what I actually said:
I.e., it was not that RF from, e.g., wifi can’t have any influence (though I have yet to be convinced that it’s audible), but specifically that RF that’s somehow coming from the internet that magically traveled through the router would not be transmitted further by using wifi locally. (And nowhere did I say to necessarily use the wifi module in the streamer)
