vibration and noise management - surprise surprise

Has anyone had a similar experience?

PREFACE: I talked to the Audiopoint people. My idea was to put audiopoints underneath a slab of marble with points stuck in the rug, and my Plinius amp on top of the marble. They suggested INSTEAD that I put the marble directly on the rug and the points underneath the Plinius, claiming that the marble plus rug work as a mechanical noise ground plain, and that the audiopoints will conduct Coulomb friction away from the amp without overdamping the unit.
I did not try this.

THE STORY: Brought home new QUAD 989 speakers. My listening room has thick carpet surrounded by a wide border of ceramic tile. Put my speakers on the rug, points dug in deeply. Sounded TERRIBLE. Thin and nasal highs, anemic bass. Immediately moved them onto the ceramic tile floor. This provided a great improvement, however, now the speakers were too close to the wall and the back wave from the speakers was compromising the imaging.

THE SURPRISE: Then I remembered the AUDIOPOINT discussion.
I went to the garage and pulled out several large heavy ceramic tiles. I laid them on the rug where I really wanted to locate my speakers and placed the speakers ON their points ON the tiles ON the rug. Everything improved: imaging, resolution, noise floor, tonal accuracy. Since this is a multi-parameter optimization problem, I may never know if the tile on rug thing was special or whether the tile simply gave me 95% of the hard floor effect while the superior placement reduced interference and the surrounding rug gave me better attenuation of the back wave.