Reducing power by optimizing the necessary precision/range of floating-point arithmetic
IEEE Transactions on Very Large Scale Integration (VLSI) Systems - Special section on low-power electronics and design
I3D '01 Proceedings of the 2001 symposium on Interactive 3D graphics
Power-Delay Characteristics of CMOS Multipliers
ARITH '97 Proceedings of the 13th Symposium on Computer Arithmetic (ARITH '97)
The design of a low power asynchronous multiplier
Proceedings of the 2004 international symposium on Low power electronics and design
Minimum triangle separation for correct z-buffer occlusion
GH '06 Proceedings of the 21st ACM SIGGRAPH/EUROGRAPHICS symposium on Graphics hardware
Design of Low Power MAC Operator with Dual Precision Mode
RTCSA '07 Proceedings of the 13th IEEE International Conference on Embedded and Real-Time Computing Systems and Applications
Lossless compression of variable-precision floating-point buffers on GPUs
I3D '12 Proceedings of the ACM SIGGRAPH Symposium on Interactive 3D Graphics and Games
Power efficiency for software algorithms running on graphics processors
EGGH-HPG'12 Proceedings of the Fourth ACM SIGGRAPH / Eurographics conference on High-Performance Graphics
Technical Section: Energy-aware hybrid precision selection framework for mobile GPUs
Computers and Graphics
Hi-index | 0.00 |
In this work, we seek to realize energy savings in modern pixel shaders by reducing the precision of their arithmetic. We explore three schemes for controlling this reduction. The first is a static analysis technique, which analyzes shader programs to choose precision with guaranteed error bounds. This approach may be too conservative in practice since it cannot take advantage of run-time information, so we also examine two methods that take the actual data values into account - a programmer-directed approach and a closed-loop error-tracking approach, both of which can lead to higher savings. To use this last method, we developed several heuristics to control how the precisions will change over time. We simulate several series of frames from commercial applications to evaluate the performance of these different schemes. The average savings found by the static and dynamic approaches are 31%, 70%, and 62% in the pixel shader's arithmetic, respectively, which could result in as much as a 10--20% savings of the GPU's energy as a whole.