__m256i A2 = _mm256_set_epi32(color2[0][ind + 3], color2[0][ind + 2], color2[0][ind + 1], color2[0][ind + 0], 0, 0, 0 ,0);
            __m256i A2_255 = _mm256_set_epi32(255 - color2[0][ind + 3], 255 - color2[0][ind + 2], 255 - color2[0][ind + 1], 255 - color2[0][ind], 0, 0, 0 ,0);
            __m256i max = _mm256_set_epi64x(255, 255, 255, 255);

            __m256i col1 = _mm256_set_epi32(color1[2][ind + 3], color1[2][ind + 2], color1[2][ind + 1], color1[2][ind + 0], 0, 0, 0 ,0);
            __m256i col2 = _mm256_set_epi32(color2[2][ind + 3], color2[2][ind + 2], color2[2][ind + 1], color2[2][ind + 0], 0, 0, 0 ,0);

            __m256i amul1 = _mm256_mul_epi32(col1, A2_255);
            __m256i amul2 = _mm256_mul_epi32(col2, A2);
            amul1 = _mm256_add_epi64(amul1, amul2);
             amul1 = _mm256_div_epi64(amul1, max);