The in situ hydromechanical behaviors within a fracturing fracture (induced by hydraulic fracturing) without a proppant support was usually estimated using artificially splitting fractures (induced by Brazilian splitting). This causes a deviation of fracture permeability from the actual one, but it is not fully understood how strong it could be. We investigate this issue by series of laboratory gas flow tests under effective confining pressure varying from 1 to 40 MPa and inlet gas pressure varying from 0.1 to 2.3 MPa and by a numerical simulation based on reproduction of fracture geometry and matching the experimental data. The obtained results show that (1) fracturing fracture surfaces are relatively smooth with fewer asperities and are prone to be in contact in response to the loading; (2) the permeability of fracturing fractures is lower than that of splitting fractures, and the error between them grows with the increasing loading, implying the actual permeability evaluated with splitting fracture is somewhat overestimated; (3) laminar and transitional flows dominate in both fractures, but a nonlinear effect is stronger in splitting fractures; (4) the onset of nonlinearities occurs at a relatively large Reynolds number (>10) because of the larger amount of inertial losses in fracturing fractures; and (5) the tortuous fracture propagation path is dependent on rock heterogeneity, and such dependence is stronger in the Brazilian splitting test.