Development Journal of a Raytracer

The second thing I have started is to add the scattering stuff to the new medium implementation. There are quite some changes I have to make to integrate it properly, but it looks good. If the current compile runs through I have quite a bit already done. I will have the light arriving at a point on the sight line through the liquid, I will then calculate the angle between the line of sight and the light and calculate the phase function. That multiplied by the scattering factors and multiplied by the integral of the absorption function over the complete line of sight should be good enough for a first approximation.

And then we will see, if it really looks like apple juice.

After yesterdays work I could finally start to use the new components. It is currently more than simple, but at least it starts to work. Currently the editor just displays the UV-map at the time it is started. I also started to add the ability to give the lines a color, up to now they were always white. As soon as that works I will be able to mark the partial mapping which is currently being worked on.

Then I will have to rewrite a bigger part, currently the vector graphics widgets retain the information. I will change that to the model OpenGL uses. Whenever there is a need to refresh the display, it will fire off a callback to get the data evaluated and drawn again. I will use that mechanism to keep the display updated, when something changes (I obviously have a mechanism in place to notice that, because I already tell the GL-renderer to update) it will tell the widget it needs to redraw.

Basically I copy the design of OpenGL. I really like the beauty and simplicity of OpenGL, it is a very well thought out system to draw. 2D is a bit simpler, as there are some things I won't need, and I won't add the more complicated stuff (like display lists or vertex arrays or textures).