I've developed a workflow supporting diffuse/specular separation for isolating specular information from cross-polarized and co-polarized photography, this used in photogrammetry. I started this in Gimp, but then had moved to Photoshop, was surprised to see the difference blend behaving very differently from that in Gimp. I can post examples if need be, but I ended up going with subtract blend, which I believe is correct anyway (color+specular from co-polarized - color only from cross-polatized = isolated specular). I'm just curious why the difference with difference.

  • \$\begingroup\$ Two different dev teams with own understanding of the meaning and mathematics behind different blending modes. And this is the result. \$\endgroup\$ Jul 11, 2022 at 16:44
  • \$\begingroup\$ Why does Colonel Sanders' fried chicken taste different from Mrs. Winner's fried chicken? Because they use different methods to cook it. \$\endgroup\$
    – Michael C
    Jul 11, 2022 at 20:39


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.