Patent 11983809 was granted and assigned to Rockwell Collins on May, 2024 by the United States Patent and Trademark Office.
A method for order independent transparency is described. The method includes rendering objects into an A-buffer by a GPU. A CPU performs dynamic memory allocation to allocate the A-Buffer based on a size of the previous frames A-Buffer. By allocating the A-Buffer based on a previous frame, a GPU-to-CPU synch point may be removed. Depending upon a size of the A-Buffer, not all objects may fit into the A-Buffer. The method may additionally include creating an R-Buffer with the identification of partially stored objects and removing the partially stored objects from the A-Buffer prior to blending.