Association for Computing Machinery
In this paper, the authors describe an alternative circuit design methodology when considering trade-offs between accuracy, performance and silicon area. They compare two different approaches that could trade accuracy for performance. One is the traditional approach where the precision used in the datapath is limited to meet target latency. The other is a proposed new approach which simply allows the datapath to operate without timing closure. They demonstrate analytically and experimentally that on average their approach obtains either smaller errors or equivalent faster operating frequencies in comparison to the traditional approach.