Olsen and Archuleta and someone else published some papers a few years ago simulating 3D motion using the 3D geometry for the sedimentary basin under Los Angeles, for an earthquake on the San Andreas fault (probably San Bernadino segment, but I forget). The motions were much worse than expected at the low frequencies that affect tall buildings the most. So, could be some bad things from a relatively distant quake, just so long as it is a big quake (probably bigger than 7.5). This scenario quake would be (much?) closer to LA than Landers was. Chris