Abstract: The invention provides active-matrix display devices with non-linear two-pole switching elements such as MIMs or three-pole switching elements such as TFTs having a smaller difference in charge current at a different image content by using a suitable. ##EQU1## This leads to less burn-in of the image to be displayed. Moreover, a satisfactory transmission characteristic is obtained at a higher value of .epsilon..sub..perp.. For TFTs it is then possible to use smaller auxiliary capacitances due to the higher values of (.epsilon..sub..perp.). The latter aspect may in turn be utilized to realise a larger aperture.

Abstract: When an active matrix LCD is driven while using inversion per n rows (n.gtoreq.2), stripe effects occur. In the case of double line inversion this leads to stripes in the picture. This can be largely obviated by supplying a different selection voltage to at least the last row of the rows in a group of n rows.

Abstract: In operation of an active matrix display device comprising an array of display elements (12), for example liquid crystal elements, each connected in series with an associated two terminal non-linear switching device (30), e.g. a MIM, between row and column address conductors (22,24), and row and column driver circuits (40,43) for applying selection signals to each row conductor in turn and data signals to the column conductors, the data signals are applied for part only of the row address period and a row selection signal commences prior to the data signal and while a reference potential is applied to the column conductors whereby during a row address period a display element is initially charged to a level approaching the lower end of the display element's operational range of voltages and thereafter charged to the required level according to the data signal. Vertical cross-talk is reduced and peak current density through the non-linear devices is kept low, thereby avoiding the risk of damage.