OptimEye replaces the "blind sweeps" of Tx settings traditionally used to find optimal settings with an automated process performed by the simulator to provide the best answer directly. This single simulation automatically adjusts Tx & Rx equalization settings in AMI models to maximize Rx eye height and then report the best Tx/Rx settings and design margin. OptimEye runs faster, reduces the chance for error and produces the best possible results. In short, it buys you time, sanity and design margin.
OptimEye uses unique co-optimization algorithms tailored for AMI analysis and built right into QCD's simulation engine. It uses information contained in special optimization control files (OCF file) to understand the capabilities of a particular Tx/Rx combination and predict what the best combination of settings will be. OptimEye then uses additional information in the OCF to simulate the Tx/Rx models with what it expects are the best Tx/Rx settings. If the actual simulation matches expected results, both the simulation results and Tx/Rx settings are reported. If the actual simulation results vary substantially from what OptimEye predicted, OptimEye updates its prediction and the process iterates until it converges.
In SiSoft's testing, OptimEye results are frequently better than results from other methods, and within millivolts of cases where extensive effort was put into creating an optimized result through other methods. Most importantly, OptimEye generates these results automatically, bypassing the extensive time and effort put into optimizing through blind sweep, design of experiment (DOE) and manual analysis methods.