We theoretically study dynamic faulting, taking account of thermoporoelastic effects including inelastic porosity change; the porosity is assumed to increase with increasing fault slip. We first derive the analytical solutions for the changes of fluid pressure, temperature, and fault slip, assuming 1-D fault model. As found in our previous study, the evolution of these quantities is controlled by a single nondimensional parameter S _u , a quantity proportional to the rate of increase in inelastic porosity multiplied by the thickness of shear heating zone. If we assume that temperature elevation remains below the melting temperature of fault rocks, S _u is found be close to zero or greater than unity. Our 2-D simulation shows that pulse-like slip appears when S _u is greater than unity, which is a necessary consequence of the emergence of slip strengthening, while slip weakening caused by thermal pressurization arises when S _u is close to zero. The applied shear stress is entirely released when S _u is close to zero, while it is released only partially when S _u is greater than unity, so that the latter will be more reasonable according to seismological observations. The fault tip stress increases significantly with the fault extension in classical singular fault model. In contrast, the fault tip stress is only weakly dependent on the size of extending fault in our model when S _u is greater than unity, so that the fault growth is more vulnerable to spatial perturbation of model parameters. Our present study indicates the importance of slip strengthening for the understanding of dynamic faulting.