You will have to write the GPU shader programs that do everything from transforming your own application data into screen-space coordinates, to calculating the exact colour of every pixel on the screen incorporating lighting and shading algorithms that you implement yourself (fortunately linear algebra makes this stuff a lot simpler than it sounds!) So OpenGL will not do any inherently 3D stuff for you – most OpenGL commands and types are capable of describing 3D positions, directions and transformations but you have to do the grunt work yourself. The next thing to know about modern OpenGL is that these days it does very little legwork for you other than allowing you to run a program on the GPU. Anyone looking to learn to use OpenGL will face a constant battle with finding relevant documentation for their chosen version on their chosen platform with their chosen extensions. In its various incarnations OpenGL spans almost 20 years and at least 7 major revisions, including the embedded versions. It is up to the many various vendors to implement the API spec as best they can. This is because it is an API specification and not a product library per-se. The most important thing a programmer should know before deciding whether to learn OpenGL is that OpenGL is very low level, poorly documented and extremely crufty. If you have any recommendations on how this could be more beginner-friendly please tell me.Īlso thanks Gregg Tavares for pointing out my various errors! Things I wish I knew when learning OpenGL

It’s mostly stuff that I found difficult to research plus a little summary of the differences between OpenGL versions.

This article is a culmination of all the little notes I took while learning OpenGL over the last several months.

Cmake opengl es 2.0 glcreateshader