Eight patients have received subretinal implants, p owered and controlled via a subdermal cable that en ters the body retroauricularly and ends in a thin intraocula r foil, placed transsclerally between the retinal p igment epithelium and the neuroretina. The tip of this foil carr ies two distinct arrays, a Multiphotodiode Array (M PDA) with 1500 electrodes, each electrode being controlled by an adjacent photodiode and an amplifier within a 3 x3x0.1 mm chip, as well as a second one with 16 electrode s on a 1.2 x 1.2mm array, for direct stimulation (D S) directly controlled from outside. Safety of the approach has been investigated by means of angiography, OCT and fundus photography. Efficacy has been investigated in term s of threshold voltages to elicit phosphenes and th e ability to perceive visual information mediated by the MPDA. Patients were tested for 4 weeks and spatial as well as temporal characteristics of repetitive multielectrode- stimulation were investigated. The implant was well tolerated in all eight patient s. OCT and angiography showed minor changes which were transient in most cases. Comparison of fundus photo graphs showed good subretinal stability of the impl ant without significant movement of the implants in relatio n to retinal structures. Threshold values have been determined for different spatial and temporal stimulation sett ings and were within safety margins. We were able t o analyze the dependency of thresholds on pulse form (biphasi c vs monophasic voltage controlled pulses), pulse d uration, pulse repetition frequency, number of pulses in a p ulse series, and on the pattern of electrode activa tion in several blind subjects. It was demonstrated that the s ubretinal MPDA yields visual percepts in a light de pendent fashion and the minimum illumination level to elici t these percepts was determined.