Sorry, just caught your comment about all layers of sheet rock glue together - No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No -
Just in case I didn't make that clear; I don't think that's a very good idea...
OK, now an actual
REASON
When you place multiple layers of mass against each other but do NOT glue them, they retain their individual characteristics for the most part; one of those is their "coincidence" frequency. This is, for all practical purposes, that material's RESONANT frequency, and at that frequency the material will be WEAKEST in its ability to stop sound (Transmission loss, or TL) - If you instead GLUE the multiple layers together, you get closer to the resonant frequency of one, THICKER, piece of that material.
In the case of gypsum, 5/8", or 15mm, gypsum has a coincidence of around 2500 hZ - 15/8", (3 layers) or 45mm, has a coincidence of around 800 hZ - both "dips" (this phenom is called the "coincidence dip") cause a drop in TL of about 7-8 dB at coincidence.
The lower in frequency this "dip" occurs, the closer it gets to the harder-to-stop bass frequencies, and so the more audible its effect on wall performance.
There is also some speculation that three individual layers of the same material, since each would be slightly different in flexibility, may spread this coincidence dip a bit in frequency; no proof here, just speculation on the part of some very experienced friends of mine...
Brian Ravnaas, a prolific poster on several acoustics related sites, is involved with a company making "green glue", a very highly damped inter-layer adhesive that is said to improve wall performance when gluing multiple layers together; however, if this is done at all it should (IMO) ONLY be done to ONE side of a wall (one leaf, not both) and possibly, in the case of 3-layer construction, ONLY between 2 of the 3 layers.
The UP side of gluing is that it will increase stiffness, which arguably can help LF performance - the DOWN side is, you're putting all your eggs into ONE basket (only ONE coincidence frequency for that leaf) AND, because of the increased mass, you're also LOWERING that frequency closer to the bass range.
Bottom line is, if you're trying for best bass containment you should use multiple layers NOT glued, heavy mass, wide air gaps, tight caulking, full insulation fill for best damping of surfaces.
BTW, I'm short of time for this but you might check the kinetics site to see if that 70 dB figure is STC; I have a feeling it is. If so, your LOW end isolation using the RIM system won't be as good as if you set your slab on 2x6's over EPDM with fiberglass or rockwool fill - the increased air gap will do wonders for the lower range isolation, in combination with the weight of 4" concrete (approx. 48 PSF excluding the plywood)
Remember, STC is a VOICE isolation system, centered on 500 hZ; for drums/bass, the TL at 50 hZ or so is 'WAY more important, so you need to calculate individual TL values instead, and bias your choices toward LF performance... Steve