Efficient Rendering with Tile Local Storage
Authors: Marius Bjorge, Sam Martin, Sandeep Kakarlapudi, Jan-Harald Fredriksen
Abstract: The most significant difference between mobile GPUs and their desktop equivalents is the limited availability of sustained memory bandwidth. With advances in bandwidth expected to be incremental for many years, mobile graphics must be tailored to work efficiently in a bandwidth-scarce environment. This is true at all levels of the hardware-software stack. We showed previously that deferred rendering could be made bandwidth efficient by exploiting the on-chip memory used to store tile framebuffer contents in many tile-based GPUs. We refer to this memory as tile local storage. In this talk, we build upon this research to demonstrate the versatility and effectiveness of tile local storage with real world content. We show how key rendering challenges can be met efficiently by use of tile local storage, and present an updated extension that has cross-vendor support.
Download here: Efficient Rendering with Tile Local Storage