sprite generation gone wrong

Discussion in 'Objects' started by Raeven, Jan 8, 2017.

  1. Raeven

    Raeven Active Member

    To be clear, the trouble is in my BLEND file, not in the FreeSO Sprite exporter plugin.
    I am not sure how, but I managed to corrupt my blend file and this is only one of the things wonky about it. I know that I need to start over again to solve my actual issue but I am curious to see if anyone knows the cause of this issue for future troubleshooting.

    Two of the views of my two-tile object is rendering properly, but the other two do not seem to know where it's tile ends one the one side.
    NOTE: Clip to Cage is checked properly.
    [​IMG]

    Again, not looking for support on the plug-in, just trying to learn what this might be a symptom of.
    Last edited: Jan 8, 2017
    JWofles and simsfreak like this.
  2. Raeven

    Raeven Active Member

    Found the cause!
    And there was no way for anyone else to find it because I am the cause :D

    A few days ago I'd experimented with the the Z buffer graphic (nearclip.bmp) to see if I could get it to work better in FreeSO (it works quite well in TS1 but FreeSo's world's Z is a bit different) and when I saved the Z buffer my graphics program's settings must have been wrong. (GIMP. depending on settings the palette saves in a different order or something. kinda annoying, really) . Never would have noticed except I didn't have a thumbnail preview in the file when I saved it.

    I still don't understand exactly why it would work properly in 2 views but not be able to figure out where a tile ends in the other two, but extracting the original nearclip.bmp solved the issue so I am resisting the urge to be fascinated by this.
  3. RHY3756547

    RHY3756547 FreeSO Developer Staff Member Moderator

    Near-clip will only affect the output if geometry is coming through the front of the tile, like on the left example.
  4. Raeven

    Raeven Active Member

    Oh! Does that mean those BMPs do something other than what I thought they do?
    I thought that they define the "front wall" and the "back wall" of each tile and your Bender-fu then calculates all of the shades of grey between the two. That is to say that the two BMPs acted as a min and max and all of the blander math worked out the color for each part of the model between that range.
    Your post (along with them being named "clip") sounds a bit like the range is already there and these act as a cutoff points. Like a barrier.
    (I am not sure I am describing the difference between the two properly, though)
    That is (would be?) really interesting. and neat.

Share This Page