This document summarizes research on adapting CPU cache optimization techniques for general purpose graphic processing units (GPGPUs). It first discusses related work on CPU and GPGPU cache architectures and optimization techniques. It then presents the conceptual design of selecting CPU techniques and analyzing their adaptation to GPGPUs. Two common CPU techniques, stride-one access and blocking, are adapted and experimental results show their effectiveness on a GPGPU, with blocking providing better performance than non-blocking approaches. The research contributes techniques for programmers to optimize GPGPU cache performance.