Original Post
Hello,
I was reading some posts across the internet and it seems the the modelview and projection matrix, along all they're modification functions, such as glTranslate, rotate, etc became deprecated and removed in the most recent openGL versions.
I'm still using opengl 3.0 version, so I don't get compilation errors using them, but seeing that they are no longer in use in current openGL versions I would like to follow that philosophy.
Since this is Computer Graphics the matrixes will still be needed, even if it's not in the openGL matrix stack. After some more reading seems that the general idea is to implement the functions by myself (or using someone elses library), and pass the matrix to my shaders.
The thing is, I don't plan to use shaders everytime I want to render something, so my question here is:
-Should I start using shaders everytime I want to render something, even though it's just making simple calculations as would be done without my shader implementation, or in these particular cases I should pass my own calculated matrixes to openGL?
I'm doing this thinking from a performance point of view.
I was reading some posts across the internet and it seems the the modelview and projection matrix, along all they're modification functions, such as glTranslate, rotate, etc became deprecated and removed in the most recent openGL versions.
I'm still using opengl 3.0 version, so I don't get compilation errors using them, but seeing that they are no longer in use in current openGL versions I would like to follow that philosophy.
Since this is Computer Graphics the matrixes will still be needed, even if it's not in the openGL matrix stack. After some more reading seems that the general idea is to implement the functions by myself (or using someone elses library), and pass the matrix to my shaders.
The thing is, I don't plan to use shaders everytime I want to render something, so my question here is:
-Should I start using shaders everytime I want to render something, even though it's just making simple calculations as would be done without my shader implementation, or in these particular cases I should pass my own calculated matrixes to openGL?
I'm doing this thinking from a performance point of view.
