The synthesis of a real image out of a virtual scene (model) is called "rendering". This process involves the perspective, lighting conditions, reflections, colors, materials and many more characteristics of the objects to be displayed. Some may be shaded by a given viewpoint, others can be blurred, still others will be transparent to show the background which appears distored. All such effects are processed by digital rendering. It is the visualisation out of a set of prepared virtual data. It can also pretend a 3D-perception of a 2D-image. The software enabling this requires algorithms based on various disciplines like geometry, light physics and cartography. The outcome is a rendered image that can be used in animations, video games, simulators, movies etc.

Fulldome movies employ digitalised astronomical scenes. One virtual flight to the stellar objects has to consider the viewpoint of the observer, in particular. "Real-time rendering" is a goal to show as much information as possible to enable dynamical illusion of the objects. The important factor is speed and not quality of the images. This is mainly used for games and simulations. On the other hand, "non real-time" means that the rendered images are stored on a hard disk. They show complex scences at higher quality and will be retrieved when necessary -- as being done for movies. Sometimes, segments of an image content are outsourced to a storage, e.g. simulations of rain, fire etc. In this case, the rendering process of a single image can take several hours or even days.