The RD-14M large LOCA test, characterized by a reliable set of experimental data, was selected for an international standard problem exercise (SPE) entitled “Intercomparison and validation of computer codes for thermal–hydraulics safety analyses”. The activity was performed within the frame of International Atomic Energy Agency’s (IAEAs) Technical Working Group on Advanced Technologies for Heavy Water Reactors (TWG-HWR). In this study, the recently improved fast Fourier trans- form based method (FFTBM) was used for accuracy quantification of RD-14M large LOCA test B9401 calculations of six participants using four different thermal–hydraulic codes. In addition, developing the capability to calculate the accuracy as a function of time-continues-valued accuracy, did further improvement of FFTBM. Namely, in the past only single val- ued accuracy parameters for selected time windows and time intervals were calculated. The objective of the study was to demonstrate that the new FFTBM is a powerful tool for quantitative assessment of thermal–hydraulic codes. For demon- stration, the test from the facility simulating heavy water reactor was used. The blind accuracy analysis was completed based on solely experimental and calculated data. However, short discussions were held with the representative from Italy (co-author, here) regarding phenomenological windows, variables and void fraction weights selection. In general, the open accuracy analysis confirmed the results obtained in blind accuracy analysis. The main conclusions from accuracy analysis agree with the conclusions from the SPE intercomparison report, which was written independently. Finally, the results suggest that the accuracy of the best calculations of the RD-14M test is comparable with the best calculations of light water reactor experiments.