Advertisement
Lordmau5

Untitled

May 25th, 2016
235
0
Never
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
text 3.34 KB | None | 0 0
  1. <+Lordmau5> YUP
  2. <+Lordmau5> THAT IS IT
  3. <+fry> so, what was it? :P
  4. <+Lordmau5> pushAttrib **before** pushMatrix
  5. <+Lordmau5> https://github.com/MightyPirates/OpenComputers/blob/master-MC1.9.4/src/main/scala/li/cil/oc/client/renderer/tileentity/CaseRenderer.scala#L18-L24
  6. <+Lordmau5> these lines here
  7. <+fry> wat
  8. <+Lordmau5> I switched pushAttrib + pushMatrix around, same at the bottom where it's popAttrib and popMatrix
  9. <+Lordmau5> and that made it work
  10. <+Lordmau5> *shrugs*
  11. <+fry> wat
  12. <+fry> that's not a thing that can be
  13. <+Lordmau5> but apparently it is
  14. <+fry> I'll say that the caching issue in the IDE is more probably
  15. <+fry> *probable
  16. <+Lordmau5> https://i.lordmau5.com/1464205654-932
  17. <+Lordmau5> check that recording, fry
  18. <+Lordmau5> (don't worry, has no sound)
  19. <+fry> well there you go
  20. <+fry> you're not just switching attrib and matrix calls
  21. <+Lordmau5> ?
  22. <+fry> you're effectively moving RenderState calls up further
  23. <+Lordmau5> wat
  24. <+Lordmau5> I am moving them into the matrix
  25. <+Lordmau5> his code has pushAttrib **BEFORE** the pushMatrix
  26. <+Lordmau5> my fix has it **AFTER** :P
  27. <+fry> no, matrix has nothing to do with it
  28. <+Lordmau5> attrib does?
  29. <+fry> move 3 RenderState calls above
  30. <+fry> (above the pushAttrib)
  31. <+Lordmau5> works as well
  32. <+Lordmau5> okay, then it's that :P
  33. <+fry> and here's the relevant issue: https://github.com/MinecraftForge/MinecraftForge/issues/1637
  34. <+Lordmau5> urgh
  35. <+Lordmau5> so what's the best approach to get this fixed?
  36. <+Lordmau5> the interesting thing is, that I do my "enableBlend" + "enableCull" call *after* pushing the matrix + attrib in my overlay renderer
  37. <+fry> don't use push/pop attrib at all
  38. <+HenryLoenwind> does anyone by chance have the transforms for "head" that let's an item render the same as it was in 1.8.9 as "builtin/generated"?
  39. <+fry> or, if you do, use GL calls and not flStateManager
  40. <+fry> *gl
  41. <+Vexatos> good to know >_>
  42. <+Lordmau5> I disabled the attrib thing in both, my mod and the class that it's preferring through my IDE
  43. <+Lordmau5> it still renders like a charme
  44. <+Vexatos> "Never directly use GL11. Always use GlStateManager" ~ fry like 4 months ago.
  45. <+fry> HenryLoenwind: assets/minecraft/models/item/generated.json
  46. <+fry> Vexatos: I'd like to hear my exact quote please :P
  47. <+Vexatos> something along those lines >_>
  48. <+HenryLoenwind> that gives me a different size as before
  49. <+Vexatos> you were yelling at me for not using GLStateManager yet
  50. <+fry> and my general advide is to not use push/pop attrib
  51. <+fry> *advice
  52. <+Vexatos> but it's useful >_>
  53. <+Vexatos> when doing a lot of attrib configuration
  54. <+fry> it's not
  55. <+fry> since it doesn't work as you expect
  56. <+Lordmau5> I wonder if there's a fix planned on Mojang's end
  57. <+HenryLoenwind> GlStateManager.pus/popAttrib is completely and unrecoverable broken. I'd recommend to ASM it so that it throwns an exception when anyone tries to use it
  58. <+Lordmau5> or if it's up to one of us to take a crack at it
  59. <+fry> HenryLoenwind: vanilla uses it
  60. <+Lordmau5> gg, vanilla uses it's own broken system
  61. <+HenryLoenwind> at one location where it (hopefully) doesn't do any damage
  62. <+HenryLoenwind> item frames
  63. * insaneau ([email protected]) has joined #simplymodding
  64. * DerpServ gives voice to insaneau
  65. <+Vexatos> fry, how does it not break there
  66. <+HenryLoenwind> because nothing in there leaves a changed GL state behind so it's a nop
Advertisement
Add Comment
Please, Sign In to add comment
Advertisement