lerp implementation for a "tween"

Question

I have this as my lerp function:

  Vec2f lerp(float t, Vec2f a, Vec2f b){
      return (1-t)*a + t*b;
  }

And I have the following code below, which I was hoping that would result in a "tween":

    Vec2f a(0,0);
    Vec2f b(3,4);
    Vec2f c(5,4);
    Vec2f d(5,0);

    if( i < 100 ){
        if(i <= 30){
            ball.pos = lerp(i, a, b);
        }
        else if(i > 30 && i < 80){
            ball.pos = lerp(i, b, c);
        }
        else {
            ball.pos = lerp(i, a, d);
        }
    } i += 1;

But what I get is a "discontinuous tween", that is instead of starting at the last point where the lerp from A to B ends, it starts somewhere else, and some goes for the other lerps. What am I doing wrong?

Answer 1

t has to be between 0 and 1 in your interpolation function, but you are passing values between 0 and 100. Change your calls to lerp(i/100.0f, a, b) and so on. (It is important that you specify the 100.0 as a floating point literal, NOT an integer literal!)

As DarenW correctly points out, you have to cover the range from 0 to 1 for every segment for the desired effect, i.e. in your case, lerp(i/30.0f, a, b), lerp((i-30)/50.0f, a, b), etc.