I’ve been studying OpenGL graphics programming and I’ve determined I’d fairly wish to make a easy recreation utilizing OpenGL.

Drawing one thing in OpenGL, for sure, is sort of the endeavour. However it normally follows the next steps:

1) Use an array to stipulate the place the vertices are of what I need to draw, in addition to

2) Create a vertex array object VAO to retailer the vertex buffer objects and retailer how the info is formatted, and so forth.

3) Generating vertex buffer objects VBOs and aspect buffer objects EBOs (if relevant). Binding the related buffer knowledge to them (ie, passing within the vertex knowledge to the VBO and vertex drawing order to the EBO).

4) Instructing OpenGL on how the info is definitely formatted throughout the array I outlined at the beginning (the vertex coordinates, texture coordinates, and so forth) utilizing glVertexAttribPointer.

5) Binding the textures if needed, enabling the vertex and fragment shaders, binding the vertex array, after which lastly drawing.

6) Unbinding the vertex array

Since I’m virtually at all times following these 6 steps after I need to draw issues I assumed I’d be capable of summary it. I assumed I might do that by passing in a batch of vertices to a ‘render’ class each render name which might do these steps each time for me so I can render issues extra simply with out the good massive spiel that takes 20 traces of code each time I need to draw a field.

This could be immensely useful. For instance, suppose I needed to attract a heightmap. All I’d need to do could be to move within the vertex places/textures to bind/shaders to make use of, and so forth, and it’d draw it for me! Possibly even alongside some timber/rocks that I might move in to that one render name.

However I’ve realised that there’re lots of technical limitations with this and I’m starting to query whether or not that is even a essentially appropriate strategy to OpenGL rendering.The purpose why I feel that is essentially incorrect is as a result of it nonetheless does not actually keep away from the large spiel of code that I’ve to undergo each time I need to draw some vertices – I simply create a ‘intermediary’ class inbetween, which is a bit ineffective and pointless.

What are the very best practises in relation to making a simplified and easy-to-use rendering ‘pipeline’ like what I’m making an attempt to explain?