Topic: renderer with LOD

Hello,

this is my first visit on this forum and I hope it will not be the last. That depends on the answer to my question.

I intend to continue the development of Tuxracer, the wellknown game with the cute penguin. The project

  • anael
  • Administrator
  • Offline

Re: renderer with LOD

we can't see your message,
but I guess you want LOD for your game, there is no build'in lod feature,
it has to be done manually for now, but there is a view elimination for big scenes (it only draw what visible).

Re: renderer with LOD

Oh, I don't know what happened with my message, but your answer was right. The scenes in Tuxracer are very large (about 500,000 or more faces) and the render algorithm should be optimized. The limitation of the view range might help for the time being, combined with clever fog settings.

However, anytime a good LOD algorithm should be implemented in my opinion. Which program module does the rendering work? Or does Maratis use an external rendering code?

  • anael
  • Administrator
  • Offline

Re: renderer with LOD

For now the best is to divide the scene in multiple objects so that they can be detected out of the screen,
the engine is Maratis one, called MEngine (itself using Maratis core MCore), it is a virtual engine, it uses OpenGL on desktop and openGLES on phones (iOS). It could theoretically uses other things like directx, console systems or software engines.

Games are coded in lua script or/and c++ pluign (using the engine functionalities)
If you really need LOD, you can always do a little c++ code that hide/show different parts of a mesh (called subMesh) according to the camera distance. There is a decimate modifier in Blender that can be used to generate manually multiple LOD models.