The precision obtainable by instrumental neutron activation analysisdepends strongly on the quality of the calibration of the gamma-ray spectrometerused for measuring the irradiated samples. Even when relative standardizationis employed, practical experience has shown the importance of detector calibration.This is especially relevant when radionuclides of varying activities are involved.The problems often encountered are those of intolerable high dead time andthe occurrence of summed peaks in the gamma-ray spectra. A simple but accuratesemi-empirical formula is presented that could effectively predict the efficiencyof a detector at any source-to-detector distance. Experimental data obtainedat 0, 50, 100, 200 mm source-to-detector distances are used to demonstratethe usefulness of this formulation. In addition, nine elements were analyzedin the NIST Orchard Leaves using absolute standardization including four elementsSm, Ta, Au, and La for which no certified values could be obtained.