I started profiling my lenses, and have set up the shooting exactly like laid out in the shooting guide (combination of tilt and moving the camera for the different framings, 9 frames per image set). Still, I usually get 1-2 grid detection failures per image set, even with the centered chart. In a shooting session I shoot multiple aperture positions in every camera position, and some grids were detected correctly at F2.8 and failed at F8, without any change in camera position or lighting.
Is there any way to help the grid detection (by telling it "here is your missing point")? Or would it be advisable to shoot several images with identical settings at the same camera position, hoping at least one passes grid detection?