WorldTreeRenderer #538
WorldTreeRenderer #538
Conversation
|
crashing at aabbrenderer::clear when I press T, not sure why (this = stack seems corrupted) |
|
this method just clear a std::vector... can you uncomment its one line and print out this instead to see if we're working inside a valid object? |
|
fixed it |
|
Yes and no. What really is missing is m_initialized(false) in the WorldTreeRenderer constructors initializer list. AABBRenderer is an OpenGL-Dependant component and should be created lazy to keep tests save. |
|
ahh, I didn't read the code to the bottom :D |
|
please add your hotkey to doc/howto.md, will do the same for the flycam |
|
I take back my previous comment, the solution by @xchrdw is nearer to the truth than my proposed. AABBRenderer is a component written by us, it takes care of laziness itself. Updated howto as well. does this work for all of you now? |
| m_voxelRenderer->afterDraw(); | ||
|
|
||
| World::instance()->particleEngine().draw(camera); | ||
|
|
||
| if (m_worldTreeRendererEnabled) { |
There was a problem hiding this comment.
the worldTree gets rendered to the normalZ buffer. you should either draw correct normals or set the drawBuffers to Color or Color and Emissiveness.
There was a problem hiding this comment.
I like the glow effects on the lines so emissiveness could stay ;)
There was a problem hiding this comment.
I don't quite get your first comment. Why am I rendering to the normalZ?
There was a problem hiding this comment.
you call
m_framebuffer->setDrawBuffers({ BufferNames::Color, BufferNames::NormalZ, BufferNames::Emissisiveness });
and render afterwards. The fragColor output gets copied to all of them
There was a problem hiding this comment.
you mean because my shader doesn't explicitly write to location 1 and 2 the value from location 0 gets there automatically?
There was a problem hiding this comment.
yes. at least on my pc I see the lines in the normalz buffer. I think the best way is to call m_framebuffer->setDrawBuffers({ BufferNames::Color }); before rendering the grid. (maybe add emissiveness)
There was a problem hiding this comment.
I added dummy values to normalz (directly into the camera) and emissiveness.
I don't really want to hack around anything (=disabling drawbuffers) until we have discussed the structure of gameplayscene (see #537)
There was a problem hiding this comment.
why do you consider setting different drawbuffers hacking around? I don't have a problem with debug info not getting written to some postprocessing buffers.
There was a problem hiding this comment.
because it would mean that gameplayscene and the pp are tightly tied and cannot be divided from each other at some future point
| vec4 color = vec4(0.3, 1.0, 0.3, 1.0); | ||
|
|
||
| fragColor = color; | ||
| normalz = vec4(0.0, 0.0, 1.0, 1.0); |
There was a problem hiding this comment.
fix indentation for color and use gl_FragCoord.z for the normalz.w buffer. I think there should either be valid data or it should not be written to it.
There was a problem hiding this comment.
which component of normalz should gl_FragCoord.z go to? x, y, z or w?
There was a problem hiding this comment.
aah! normal AND Z buffer, not normalized Z buffer :D
| /** | ||
| Provides the geometry of a unit-cube | ||
| Not that the lifetime of this class has to be >= the lifetime of the program/vao | ||
| it is bound to, otherwise their behaviour is undefined |
|
looks good to me |
|
|
This draws all active leafs of the WorldTree as wireframes. Interesting to watch and (in combination with the flycam?) useful to debug and optimize the worldtree.
(De)Activate it by pressing T
Note that while I placed the WorldTreeRenderer inside the GamePlayScene, I'm not really sure if it belongs there. See #537