The use of the true-amplitude weighing function in the modified diffraction stack operator is essential in obtaining a reconstructed source pulse that is proportional to the reflection coefficient of the target reflector. To determine the true-amplitude weighting function, the 2×2 ray transformation submatrix, designated as matrix B, as well as the number of ray-branch caustics should be accurately computed. An alternative method to compute the matrix B and the number of ray-branch caustics is presented. By using a basic equation in Bortfeld's theory, matrix B can be obtained without any numerical differentiation.