Interferometry applications (e.g., radio astronomy) often wish to optimize the placement of the interferometric elements. One such optimal criterion is a uniform distribution of non-redundant element spacings (in both distance and position angle). While large systems, with many elements, can rely on saturating the sample space, and disregard “wasted” sampling, small arrays with only a few elements are more critical, where a single element can represent a significant fraction of the overall cost. This paper defines a “perfect array” as a mathematical construct having uniform and complete element spacings within a circle of radius equal to the maximum element spacing. Additionally, the largest perfect non-redundant array, comprising six elements, is presented. The geometry is described, along with the properties of the layout and situations where it would be of significant benefit to array application and non-redundant masking designs.