Uh-oh... it sounds like we may be in trouble. It sounds like you know whereof you speak (heaven knows I don't, and I don't think the Owens Corning franchise does, either). If I describe the original plan, as I understand it, would you be so kind as to tell me where it falls short? And what, if anything, we can do to fix it, short of breaking the contract altogether?
WALLS. The Owens Corning system is basically made up of a thick slabs of rigid insulation covered with fabric. The slabs attach to metal frames that are in turn (in most places) attached directly to the concrete basement walls. As far as I know, there is no space between the back of the OC wall and the concrete wall.
The absorption properties of the OC walls are supposed to be quite good; we would like things to be quieter for our son the drummer as well. My assumption (flawed I presume) was that relatively little of the remaining sound would not pass through the concrete walls, so that part of the room (most of it) would be taken care of.
The main wall I was worried about originally was the new dividing wall, which would consist (on the back side) of open insulation between metal studs. I figured we could reduce the sound considerably by attaching drywall (or Homasote, or the QuietRock suggested by nibhaz) to the studs. I don't know exactly how wide the studs are, but RC is at least a possibility there.
The ceiling issue is more complicated that I originally suggested, because a beam and heating duct run across the middle of the room-to-be. Owens Corning proposes to box these out with pine and insulation above. At this point, though, the only heat would come from the duct running across the room, so the project manager proposed a panel that could be slid away from the register when the room was cold. (It could be closed during drum practice.)
We aren't looking for absolute quiet here, but if this isn't going to make a significant difference, then we'd best get out while we can. Is this approach salvageable at all?