To compare the performance of a convolution-based metric, to that of the van Rossum metric and the Victor-Purpura metric, we will evaluate all three metrics on computer-generated neural data of multiple kinds. One of the data arguments will be a systematically altered version of the other; we will simultaneously plot the metrics versus an ``offset'' parameter that controls how much alteration occurs. The data include Poisson spike trains and simulated Hodgkin-Huxley neuron data (standard model...