P. Qian sent me a PDB model that caused Chimera to crash when calculating an MSMS surface (menu Actions / Surface / show). He got it not to crash by shifting the coordinates of all atoms (by moving and saving the PDB with "use untransformed coordinates" turned off, and reloading that file). This is a little nicer than changing atom radii as we've advised in the past. Not sure how often a simple shift (or rotation) avoids the crash though. Tom