A method of and apparatus for generating a graphical image of at least part of an object provided within a hierarchically divided region of space from a plurality of stored graphical representations, of the differently sized sub-regions of the space is described. The method comprises: determining a viewing distance between the object and a viewpoint from which the object is being viewed; and using the determined viewing distance to select, from the plurality of stored graphical representations, a graphical representation of a sub-region of the space in which at least part of the object resides, the size of the selected sub-region corresponding to the calculated viewing distance and determining the level of detail of the graphical image of at least part of the object to be generated. The method has particular application in the computer games field where creating three-dimensional simulations of the world and objects within it is highly desirable.