#28
by StratComm
In theory, it would, I'm just not sure that PCS actually aquires that data. I forget where I have set it in the past, so I'll take a look.
EDIT: It needs to be set in modelview, PCS doesn't have the interface for it and apparently does not convert it correctly. MOI is, according to what Bob dug up not long ago, dependent on the mesh as well as mass, etc, but it still should be in the same order of magnitude as the original. Yours was, well, not. Off by as much as 10^40, in fact.
These are the values of the original vs your high-poly one, for the record:
Original Sobek:
0.00000000031218144736 0.00000000000004268307 0.00000000000049032301
0.00000000000004268307 0.00000000030472743773 -0.00000000003332326429
0.00000000000049032306 -0.00000000003332326429 0.00000000169464808852
High-Poly Sobek: (Trailing zeros removed for space)
227906650112.0 4371236168191097700000000.0 274952335980479310000000000.0
0.00000000003865366707 262233136196483020000.0 0.00000000000014932656
0.0 40.099060058593750 13386286.0
On closer observation, it would appear that the MOI tensor is symmetrical about the diagonal, so the fact that whatever garbage is in your sobek's MOI isn't could really be causing some wierd stuff. I don't know where it came up with that stuff (really, 2E26 and 0 being on opposite corners?) but it is nowhere near appropriate for MOI data.
I've changed it (haven't yet had the chance to test it in game) and will be happy to send it to you if it saves you time in the copy-paste phase.