When modelling transforming is in space mode (the default) each modelling transformation is specified with respect to the most recent set of space axes (sometimes called current or local axes). Subsequent changes to the transformation matrix are post-multiplied.
When in picture mode each modelling transformation is specified with respect to the picture (or screen) axes. Subsequent changes to the transformation matrix are pre-multiplied.
The picture axes never change and have their origin at the bottom left-hand corner of the picture, with X horizontal and to the right, Y vertical, positive upwards, and the Z axis perpendicular to the plane of the picture, in the direction of the user.
