Original Post
I'll make this as short as possible. I'm making a GUI from scratch for fun. I have my rendering (OGL) code completely seperated from all objects in the scene, so that I can plug in renderers at will. Loading a mesh normally goes like this: -Create Triangles, load into Mesh, -Copy mesh triangles to root of Octree, divde octree(distribute triangles) During renderer startup: -Translate triangle octree into vertex array octree During rendering: -Render vertex arrays in vertex array octrees My first thought is that this is all too much work for displaying GUI elements. My question to everyone is: Do you have a seperate rendering path for 2d or GUI elements? Or do you just put it in the scenegraph and let the renderer render everything using one algorithm?