If one thinks noise is an issue, they can and should test for that. Whether they think the problem is noise getting inside the ethernet cable or the noise getting out of the ethernet cable.
I personally haven't found noise as discussed above, to be a problem in my setup with my Bryston BDP-1. I use an 18 feet AES cable so I can move the BDP-1 far away from my rack so no ethernet cable goes near the rest of the components. I don't notice a difference when the BDP-1 and the ethernet cable is in the rack or far away. In fact, I've tried piling 100 feet of Cat6 UTP on my DAC and interconnects, and noticed no difference in sound. I suppose my interconnects and gear is well shielded. It very well may make a difference in other people's setup.
Similarly, I can play music off a USB flash drive, and use the ethernet connection for only interface control. I don't notice a difference in sound by the plugging and unplugging of the ethernet cable. If there is noise present, the unplugging of the cable should make it better.
However, I do find a difference when streaming through Roon based on the length of the ethernet cable or the device/power supply used. Based on my experiments, I haven't been able to find conducted and radiated noise as an issue. Aside from shielded and grounded ethernet cables which are breaking the isolation, I suspect in a calm home environment, the differences in sound may be with how the signal is dealing with the PHY. Signal integrity could be at play here.
It's hard to conclude anything without measurements, but still there are tests that people can do at home to test out various hypothesis.
I personally haven't found noise as discussed above, to be a problem in my setup with my Bryston BDP-1. I use an 18 feet AES cable so I can move the BDP-1 far away from my rack so no ethernet cable goes near the rest of the components. I don't notice a difference when the BDP-1 and the ethernet cable is in the rack or far away. In fact, I've tried piling 100 feet of Cat6 UTP on my DAC and interconnects, and noticed no difference in sound. I suppose my interconnects and gear is well shielded. It very well may make a difference in other people's setup.
Similarly, I can play music off a USB flash drive, and use the ethernet connection for only interface control. I don't notice a difference in sound by the plugging and unplugging of the ethernet cable. If there is noise present, the unplugging of the cable should make it better.
However, I do find a difference when streaming through Roon based on the length of the ethernet cable or the device/power supply used. Based on my experiments, I haven't been able to find conducted and radiated noise as an issue. Aside from shielded and grounded ethernet cables which are breaking the isolation, I suspect in a calm home environment, the differences in sound may be with how the signal is dealing with the PHY. Signal integrity could be at play here.
It's hard to conclude anything without measurements, but still there are tests that people can do at home to test out various hypothesis.