It is believed that DD mesh with CRIMS data for a thin walled part will results in more accurate prediction of warpage than 3D mesh (Correct me if I am wrong).
Does it mean, without CRIMS data in the material 3D and DD will have the same accuracy of warp results ??
It is not appropriate to use 3D mesh when part is thin walled, instead you can use midplane or DD mesh.
CRIMS will always give you better results with DD in Thin walled components.