Abstract: In coordinate measurement, the larger the number of samples the smaller the error is to be expected. This paper raises the questions: whether it is possible to lower the number of samples and raise the accuracy at the same time. The answer, based on a sequence of numbers and their distribution, is affirmative. The basis for such a sequence, its computation, the analysis of the error, and comparision with uniform and random sequences are examined, along with experimental results of applying such a sequence to the measuring of the surface roughness.

Key words: Distribution accuracy, Error inspection, Number theory, Sampling, Sequence