The production of near-photorealistic images of buildings is becoming increasingly common. The software to produce reasonably sophisticated images being available at affordable prices and the increasing power of generally affordable computers have contributed to this trend. It is also probably the case that the run-of-the-mill architectural practice sees the competition producing this kind of image with a superficially beguiling quality and follow suit. What we ask in this paper is whether we should be more thoughtful about the kind of image used? Should the kind of image chosen to suit the stage of the design that it applies to and the nature of the human agents viewing the image? Of course, in posing the question we imply our answer, that it should. What we do in this paper is to illustrate why we feel it should and what the consequences are for the education of architects who are about to join the world of practice.