For all inputs and calculations, GoldSim uses double precision floating point arithmetic. However, in order to reduce the size of the output files, when results are stored, they are only saved in single precision format. This means that:
- Stored results are accurate to no more than about seven significant figures, even though the calculations were carried out with up to 13 significant figures of accuracy.
- Results that are smaller than about 1E-38 have reduced precision, and results that are smaller than about 1E-45 show as 0.
- Results that are greater than about 3.4E38 show error values.
Note that all GoldSim calculations are done internally in SI units, so that the numbers mentioned above refer to values when expressed in those units.