Retinal prosthetics consistently demonstrate the ability to restore limited visual perception to those blinded from outer retinal degenerative diseases. However, the performance of retinal implants is highly inconsistent and high-density arrays have proven only marginally better than much sparser ones suggesting that improving the overall quality of elicited vision may require more than just a high density electrode array. Existing devices are also implanted subretinally and epiretinally raising the possibility that electrode location also contributes to percept quality. Here, we compared the responses to stimulation from subretinal and epiretinal electrodes in the same cell. Use of a 4×4 subretinal electrode array allowed us to also compare responses to different numbers of electrodes activated simultaneously. Surprisingly, responses showed minimal dependence of both the electrode position (epiretinal vs. subretinal) as well as on electrode size (one vs. up to nine electrodes). However, when charge density considerations are implemented, such as those that are necessary during clinical use, the responses arising from smaller electrodes were less effective. This finding may help to explain the inconsistency between theoretical visual acuity and achievement in clinical testing with the high density implanted arrays.