poly_mode question (and the hunt for jaggie reduction)


    May 04 2013 | 11:49 pm
    okay, I'm returning to an old old issue of mine, but since I'm using jitter this coming week I feel I must ask:
    This is a picture of the back of some cubes, poly_mode 0 0, drawn with jit.gl.multiple controlling jit.gl.gridshape: http://i.imgur.com/DNt1lWm.png
    ...lovely.
    And these are the same cubes, but drawn with poly_mode 0 2: http://i.imgur.com/kEJlsxP.png
    so as you can see, poly_mode 0 2 is actually able to draw a cube - and even though you can't get rid of the points, they're still quite attractive here. It's as if "poly_mode 0 0" isn't... isn't being drawn properly. I know I'll be told something about openGL specs, but... I mean. It's cubes, right?

    • May 06 2013 | 7:43 pm
      any ideas?
    • May 06 2013 | 7:53 pm
      have you tried increasing @dim ?
    • May 06 2013 | 7:56 pm
      yeah, and decreasing, too.
    • May 06 2013 | 8:08 pm
      how about posting your patch?
    • May 06 2013 | 9:58 pm
      yeah, had to spend ages cutting it down (I would take a "how to feature freeze" class in a heartbeat):
      Hopefully this is pretty self-explanatory. Do ask if there's anything odd going on. thanks again.
    • May 06 2013 | 10:22 pm
      instead of messing with poly_mode, try @cull_face 1, to cull the back-facing polygons.
      also, you are sending a negative scale to your gridshapes, which is why the shading is screwy, but maybe that's intentional.
    • May 07 2013 | 12:45 pm
      huh, my reply never made it here... strange.
      Thanks for looking into this Rob - cull_face doesn't help, it seems - this is my main patch generating with @cull_face 1, @poly_mode 0 0: http://i.imgur.com/TrU28E4.png
      - in this example it looks almost intentional... It happens regardless of dim, and regardless of overlaps with other objects on screen.
    • May 07 2013 | 12:46 pm
      and yeah the weird scaling wast just me messing around, it happens with... unipolar scaling too ;)