The restoration of cross sections in extensional settings is commonly carried out with simple shear constructions which use either vertical or inclined shear geometrical construction techniques. These techniques assume plane strain and that the X and Z principal finite strain axes are contained within the plane of the cross-section. In 2-dimensional balanced cross sections the displacement vector is, of necessity, assumed to be in the plane of the section. In 3-dimensional restoration the main components are the horizontal displacement vector (heave), the deformation plane that contains the X and Z principal strain axes in plane strain deformation and the vertical/inclined shear angle (shear pins) parallel to which the hanging wall deforms during translation. During 3-dimensional restoration it is possible to vary the orientation of the displacement vector and shear pins in both azimuth and plunge. We can therefore test the sensitivity of artificial models, or natural examples, to variation in restoring parameters. The examples used here show that restorations are extremely sensitive to the shear angle chosen for the hanging wall, and in the case of an oblique ramp model and a natural fault example, to the movement direction assumed for the restoration. Using map view visualisation of faulted hanging-wall surfaces it is possible to obtain a good estimate of the original slip vector which would not be apparent from 2-D sections by matching hanging-wall and footwall cut-offs. Using this 3-D approach it is possible to minimise errors in restoration which result from erroneous restoring slip vectors. Any information on the angle of simple shear deformation of the hanging wall during deformation should be used when restoring the 3-D surface or cross-section.