Abstract: In the majority of previous research, the limited diversity of joint surface morphologies and the restricted number of samples have resulted in a narrow range of obtained fractal dimensions. As a consequence, the fractal differences manifested in the shear mechanical behavior of rock joints have not been prominently evident. The current paper utilizes standard deviation(σ) and fractal dimension(D)as controlling parameters to generate randomly rough joints and investigates the shear mechanical behavior of joints with different fractal characteristics through Constant Normal Load(CNL)shear tests. Based on the geometric characteristics of the generated rock joints, the overall variability of jointing increases with the augmentation of standard deviation, while the frequency of variability fluctuations amplifies with the increase in fractal dimension. Utilize 3D printing technology was employed to fabricate ten different joint surface specimens, consisting of two standard deviations and five fractal dimensions. Shear tests were carried out using the TAJW-2000 rock triaxial shear rheometer. The results indicate that, under constant normal load, the shear strength of the rock joints increases with the increase of fractal dimension. Moreover, larger standard deviation leads to a higher increase in shear strength. Additionally, as the shear rate increases, the shear strength of the rock joints decreases. Through the analysis of the variation in shear strength ratio(η_v)with increasing fractal dimension for different standard deviation joints defined based on shear rates, it was observed that the shear strength ratio exhibits a decreasing trend as the fractal dimension increases. The impact of rate variation on the morphology of complex jointing is found to be more pronounced. The residual strength increases with the increase of normal stress. When the standard deviation is relatively large and the shear rate increases, there is a noticeable reduction in the residual strength, especially under high normal stress, the significant reduction will be observed. Based on the power-law relationship between the directional root-mean-square of the first derivative of each rough rock joint(Z′ ₂) and the shear strength, a predictive model for the shear strength of the rock joints under different shear rate conditions is proposed.