Analyses are made to compare the two approaches of using the transient method to determine permeability of rocks: the simplified version and the numerical version. The simplified version assumes that when the sample volume is much smaller than the reservoir volume, the fluid storage in a rock sample can be neglected and the pressure decay in the upstream reservoir can be approximated by an exponential function of time. The numerical version uses finite difference method to solve the differential equation of pressure decay and matches the observed pressure decay with the calculated decay curves. The pressure decay calculated by the numerical version is not generally a simple exponential function of time, as suggested by the simplified version. Our analyses show that the numerical version fits to the observed data very well. The permeability value determined by the simplified version tends to be greater than that determined by the numerical version. The difference in apparent permeability between the two depends on many factors such as rock properties (porosity and compressibility), sample size, reservoir volumes, etc. A relative fluid storage is used to illustrate the difference of these two approaches. For a relative fluid storage value of greater than 0.03, the difference in permeability between the two is more than 30%. For a system with a fluid storage of less than 0.01, these two versions agree with each other well.