Background: This study aims to investigate the effect of reference dose calculation grid size (RDCGS) on gamma passing rate (GPR) for patient-specific quality assurance of intensity-modulated radiation therapy (IMRT) and volumetric modulated arc therapy (VMAT). Materials and Methods: A total of 20 patients were retrospectively selected. Both IMRT and VMAT plans were generated for each patient. Reference dose distributions for gamma analysis were calculated with RDCGS of 1–5 mm at intervals of 1 mm. Dose distributions were measured using MapCHECK2 and ArcCHECK dosimeters. Both global and local gamma analyses with gamma criteria of 3%/3 mm, 2%/3 mm, 2%/2 mm, and 2%/1 mm were performed with various RDCGS. Results: As the RDCGS increased from 1 mm to 5 mm, the average global GPRs with 2%/2 mm for VMAT with MapCHECK2 and ArcCHECK decreased by 9.3% and 5.9%, respectively. The average local GPRs decreased by 14% and 11.7%, respectively. For IMRT, the global GPRs decreased by 4.8% and 6%, respectively, whereas the local GPRs decreased by 10.5% and 8.6%, respectively. The effect of the RDCGS on the GPRs became larger when performing local gamma analysis as well as when applying small distance-to-agreement (DTA). As the RDCGS increased, the average changes in the GPR per mm of DTA change increased regardless of the type of radiotherapy, detector, or gamma analysis. Conclusion: For an accurate verification of the IMRT and VMAT plans, it is recommended that the reference dose distribution must be calculated with the smallest possible RDCGS.