Verification test with simulated data: a comparison of the E-mode excess mass (intensity scale) obtained from the 1024 deg² mock data with two different method. The lens pairs are selected from the separation interval close-θ and redshift bin high-z (z̅_d ≈ 0.52). The lens positions inside the map are indicated by crosses. The shape and sampling noise of the sources are negligible here. Top panel: explicit stacking of the convergence on a grid; only applied to the simulated data. Bottom panel: reconstruction with a shear stack as applied to our CFHTLenS data. The contours show the relative difference between the two methods (top) and (bottom) in steps of 10%, 20%, etc. for regions where the signal magnitude is above 5 × 10⁻⁴.