I usually would not consider my self a ‘beginner’ when it comes to OpenGL… however, I recently tried to use normal OpenGL clip planes and I simply do not understand how they are supposed to work. Specifically what the plane equation needs to look like, depending on how your view and projection looks like etc.

The documentation states

equation is transformed by the inverse of the modelview matrix and stored in the resulting eye coordinates

Now that right there confuses me already. Usually you multiply eye coordinates with the inverse modelview to get their object coordinates. But in this case, we have to construct a plane equation, that is multiplied by the inverse modelview and the resulting plane vector is used against the eye coordinates of the vertices to determine whether they are in or out.

Let’s say I have a Ortho projection for 2D rendering, something like

```
glOrtho( 0.0, 100, 0.0, 100, -1.0, 1.0 );
```

and a viewport like this

```
glViewport( 0, 0, 100, 100 );
```

Now I want to clip away anything that is below 20 and above 80 on the x-Axis, what plane equations do I have to use?

As far as I have tried a plane equation like this: -1,0,0,80 would work for the latter, but 1,0,0,20 would not work for the former.