There is a bit of a rubber band effect with Entreq in my experience, spaz. I actually saw it happen when I did my amateur measurements just using the "second sound card" technique. I'd pull the plug and the noise floor did not drop right away. It slowly dropped, then remained somewhat steady after 15 minutes. Same thing happens when it re-connected. Entreq I believe says some time connected is needed for the full results to appear and I do have to say I literally saw this happen with my eyes, not my ears. I think anyone wanting to test Entreq would really want to have the box and cable connected for a full 48 hours to do "with" testing and 24 hours disconnected for "without" testing. I wasn't so pedantic in my tsting because my test rig playback sound card only has a quoted S/N of 117dB at 48 KHz to begin with (and 114 at 96) and I wasn't going to sit there for days when the workstation was needed for other stuff.
Oddly enough, in my first experienced that was a failure (the one with the "antenna" interconnect), I could see the high frequency noise floor change immediately when the Entreq was connected / disconnected. But after building a "proper" interconnect, I then witnessed the rubber band effect.
I actually did the test a few times to make sure what I was seeing was repeatable and representative and was not due to some other factors unrelated to Entreq.