Adaptive partition mechanism with arbitrary tile shapes for tile-based rendering GPU architecture

摘要

A method and apparatus related to an adaptive partition mechanism having an arbitrary tile shape for a tile-based rendering GPU (graphics processing unit) architecture is described. In an embodiment, a primitive intersection cost value for each atomic tile of the image is determined based at least in part on the size of the vertex element, the length of the vertex shader, and the number of vertices of the primitive of the image. Other embodiments are also disclosed and claimed.