Hi!

I am a programmer-hobbyist, who enjoys casual game programming and some slightly more useful programming for personal work. What I end up wondering every time I am writing some code is whether the amount of classes and protected members I use is appropriate. For example, this is one of the classes I usually have in my game projects:

Code:
//on a side notice: this class is further inherited by several other classes
class GLObject
{
    public:
        GLObject();
        ~GLObject();
        
        virtual void glRender() {}
        
        virtual void timeStep();
        
        Vector getPosition() const {return Position;}
        void setPosition(Vector NewPosition) { Position=NewPosition; }
        
        Vector getSize() const {return Size;}
        void setSize(Vector NewSize) { Size=NewSize; }
        
        Vector getVelocity() const {return Velocity;}
        void setVelocity(Vector NewVelocity) { Velocity=NewVelocity; }
        
        Vector getAcceleration() const {return Acceleration;}
        void setAcceleration(Vector NewAcceleration) { Acceleration=NewAcceleration; }
    
    protected:
        Vector Position, Size, Velocity, Acceleration;
};
The problem with this is that I have made all member variables protected. This prompts for two functions for each variable (a "set" and "get" function). What I really am worried about is that those functions (especially the "get" one) get called *really* often.

Does this have any effect on speed or does it not? Would it be better to keep all of those members public, delete all the accessory functions and access variables directly? Or does this make the code more C-like? Is the latter necessarily a bad thing for C++ code? Am I so to speak overusing the classic programming approach?

Any help and thoughts are highly appreciated!
Thanks