Got the Vive setup. Everything for the most part went as planned. The magnetic mounts have a little less stability along the thin edge than I'd like, but nothing that impacts actual use. (Measures out to .59/.15/.11 max/std/rotation deviation for tracking jitter which is on par for those reporting direct wall mounts.)
Revisiting my mounts for a bit. I decided to redesign them because I wasn't happy with the fact that they could rock along the thin edge. They wouldn't move with normal operation, and would fall back to the same position if nudged, but just knowing it was possible kind of bugged me. So I went ahead and modeled a completely custom design that didn't use any of the provided hardware. Support along both axes, twice the magnetic holding power, predefined angles, and a cleaner look.
Holds rock steady now, and amusingly enough could be used on the floor on a makeshift setup as the legs are large enough to stabilize the lighthouse even without the magnets.
Unfortunately, ironically enough, my jitter performance is marginally
worse now than my initial measurements. I think the old design had a level of vibrational decoupling as drop ceiling rails are less rigid the further you go from the walls or cross sections. I can't compare directly though as they're destroyed now to salvage the magnets and metal shielding rails. (The originals weren't designed to disassemble cleanly.) Perhaps I'll try putting some sorbothane between the lighthouse and the bracket itself and see if that leads to any improvement. Jitter aside, relative controller positioning seems more consistent as I rotate around now, but that could be due to the sync cable being used instead of relying on optical sync. I'd get a message every once in awhile about the lighthouses losing sync before, so I figured I'd run the sync cable while I was printing out the new brackets.
It's amazing (and somewhat annoying) how sensitive everything is. The more concerning part is I'm not sure how to address the inherent weakness of the Lighthouse producing its own vibration in future iterations. Any device (including the Rift's cameras) will be subject to external vibration, but the lighthouse also has to contend with its own on top of that. Theoretically something like a micromirror array would eliminate its own vibration, but they don't have the sweep angle needed, and I'm pretty sure they're also binary in position.