So there seems to be more proof for point 2. A too stable and clean ethernet signal will make the sound quality decrease as well.

I have removed 4 Ferrite cores and I have left only 2 directly between the JCat netcard and the DAC. The sound has cleared up in 30 minutes

That´s crazy... Not for you, since you are used to this i guess. Not sure why i even try comment on this...:-) but are you sure it has anything to do with ethernet at all? My instinct want to think it is related to something else since the sound is getting "dark"? But i guess you are pretty dialed into this by now, and have drawn a lot of conclusions at this point.

Do you think that ferrit setup is even reducing noise? I never really understood what it does. I thought it only was an affective shield but it is so effective in my setup that i question if it does even more?

Goodmorning Tobias, Yes indeed I'm pretty dialed into this subject. Actually the ethernet was the very last thing that I suspected to be the cause. The list of things that I have tried before I came to the conclusion that it must be the ethernet is very long, the tiniest things sometimes not even related to audio are tried extensively. But also a complete swap of the audio system and setting up a second system are on the list.

I'm positive now it's the ethernet, because I now know exactly what I can do to solve the problem every time. I even can get the second system dark, when I put it on the same network. Adding noise into the ethernet makes things better when the system is dark. Adding more filters on the powernet or like I have done with the ferrites on the ethernet cabling is making it worse. When adding filters on the powernet then the ethernet system will benefit from this as well, but with the effect that it will sound dark in time.

Yes they do. The trick is that you actually build an inductor. The more windings you create with the ethernet cable around the ferrite in the same direction the more inductivity you will create. It's not a shield, but just like with a transformer the inductor will work against the signal, it doesn't like changes in the signal at all. So actually sine waves are continuous changes in the signal. So the inductor will try to flatten the sine waves, but due to the small amount of windings it will only flatten the spikes that are on the sine waves. Spikes are also changes in the signal. See it as a smoother.