The validity of the current practice of using tank pressure curves to discriminate airbag inflators is evaluated. Sled test results of two inflators, which have similar tank pressure curves, are first compared. Significant differences in airbag performance have been observed which suggests that the two inflators are not similar. The gas dynamics of airbag inflation are then reviewed to develop theories to explain the phenomenon. The dual-pressure method, which was previously developed for modeling airbag inflators, is found to be useful in this analysis. The analysis clearly shows that the inequality is due to the difference in gas temperature among inflators. We find that the higher the gas temperature the faster the gas venting and leaking will be. This is why different airbag performance is obtained from inflators which have similar tank pressure curves. Based on both theoretical and experimental evidence, we conclude that knowing the tank pressure curve alone is insufficient to determine the capability of an inflator for both driver and passenger airbag systems, as long as vents and/or leaks exist in the systems. The inability to discriminate among inflators early in the design, could lead to late changes in the design of the restraint system when the actual performance of the airbag is measured during sled or barrier tests. Finally, four new methods to discriminate inflators are discussed.