It has been recently shown^19,20 that specifications of a Germanium detector simply in terms of its relative efficiency, defined according to the familiar IEEE standard²¹, does not always predict how the detector will perform in an actual "real world" measurement situation. In these "real world" applications, the samples are Marinelli beakers, filter paper or bottles. These geometries are very different to a point source at 25 cm from the end cap, as defined by the standard. This means that in almost every case, specifying a detector requirement in terms of the IEEE "point source" relative efficiency is at best imprecise, and may be misleading. The IEEE relative-efficiency measurement does not distinguish among the different Germanium crystal sizes and shapes that are made. Germanium crystals can be made with diameter equal to length ("square"), diameter larger than length ("over-square"), or diameter less than length ("under-square"). The efficiency is affected by both the front surface area or solid angle (diameter) and the stopping power (length). Three detectors with very different shapes (over, under, and square) could all have the same relative efficiency at 1332 keV, but would have very different efficiencies for other samples. This is well illustrated by Figure 32, which

represents the 3 different detector geometries, all of which have the same relative or absolute efficiency for a point source at a distance (source A), but which very clearly have significantly different absolute efficiency for the counting of thin disk source B (e.g., filter paper) on the endcap, where diameter of B is greater than the smallest diameter detector, but less than the other two larger diameter detectors.