3dsmax physx 3.4.0 plugin floating point precision

[yuebingzero]

using points (-40.0, 0.0, 0.0), (-20.0, 0.0, 0.0), (0.0, 0.0, 0.0) to test. The exported fbx has exactly those three points. But exported apx has points (-39.9999962, 0,0, 0.0), (-19.9999981, 0.0, 0.0).

for our real mesh, point errors are acceptable, but errors of computed normal are very obvious.

This plugin is not open source, we don`t know how to fix this.