This paper describes the use of multiple go-path measurements, in combination with algorithmic test sequences, to aid in improving test efficiency and accuracy and diagnostics. As an electronic device or circuit is tested, the output of the unit under tests (UUT) may be considered as a function of the input. Through the use of multiple tests designed to exercise system capabilities in evaluating UUT performance, the characteristic behavior of the UUT can be established. Test results obtained automatically can be used in conjunction with evaluating software in classifying good and failed UUTs. Also, failing UUT behavior can be further classified to distinguish faulty lower-level UUT assemblies and components based only on go-path measurements. Software engines can actually discover new information about a circuit's performance, thus aiding in human understanding of the unit under test (UUT). UUT data rates and data volumes are generally beyond human processing capability in our ability to do fine differentiation, however neural networks and other software routines perform well under these conditions. Analysis of UUT performance may require the examination of small changes in multiple, and varied UUT signal parameters (e.g. voltage and current) that may be within current algorithmic test tolerances, yet are representative of circuit failure characteristics. These slight variations are not understood using typical circuit diagnostic routines. Software engines like neural networks can see slight variations in signals and make decisions based on these variations. One of the major problems with theoretical circuit diagnostic analysis routines is the narrow focus used to evaluate problems. Checking for a specific logic level or using the upper/lower limit criteria does not focus in on unique failures like circuit skewing. This paper discusses techniques that can provide a viable solution to many of the testing problems that we experience today. Also, data capturing is discussed.