Cascade performance is examined when the degree of enrichment per separative unit is arbitrary, but not small. An approach is developed that is applicable to new technologies, which project significant enrichment per unit. Analysis shows that ideal operation and minimum separative work do not occur concurrently and that neither is a reliable guide for optimizing costs for a small cascade. Cascade performance is studied in a systematic manner that keeps the feed assay, the product assay, and the amount of product fixed. Many small cascade configurations are examined, and the results are combined with a hypothetical operating curve, thereby generating a simplified relative cost evaluation. Three costs are considered: feed, mechanical pump power, and laser power. No single configuration optimizes all three, but the analysis does indicate the superiority of a few of the examined configurations.