The Virtual Shadow Map (VSM) technique, unlike cascaded shadow maps, employs clipmaps to divide the shadow maps. By allocating higher detail clipmaps for the nearest objects and lower detail clipmaps for objects in the distance, VSM optimizes the rendering process. Additionally, the clipmaps are further divided into fixed-size pages, and these pages are allocated only for objects within the camera frustum. This approach ensures that memory is efficiently used for visible objects, resulting in per-pixel accurate shadows while maintaining a relatively low memory budget. Furthermore, a PCSS implementation is available for generating soft shadows.