Working on transmission we had something which to us looked correct.
However tonight we realized that it was only “correct” looking by a double fluke error in our code.
In trying to fix the problem we’ve found error after error in our transmission algorithm we’re working to fix.
Here’s what we have thus far:
As we fix it we’ll update this article. Hopefully we’ll finish it soon. 
Our raytracer has been significantly expanded.
First, we added reflection, which was cool… but everything looked like glass balls.
Second, we added distributed rays for reflections which gave us wonderful blurry reflections
Third, we applied this distribution to shadow calculations to get ourselves some soft shadows.
Finally, we added simple pre-rendering of a rotation around a scene using a distance from to position a point a distance
from our camera to rotate around.
Take a look:
With fuzzy shadows:
Animation even:
(The jerky motion is an artifact of the video screen-cap and not of the raytracer)
And yes, it has procedural texturing!
Texturing mapping from a file is on the way, but who knows when it’ll be done…
Obligatory photo!
Dear Dr. Phong,
As Computer Science students wetting our feet in the realm of 3D ray-tracing, I can’t help but feel we’re standing on the shoulders of giants. You’ve provided us with this amazing approximation of synthetic lighting. We love it. We implemented it for our ray-tracer.
Thank you for everything.
Love,
Team More Bloom
P.S. Obligatory Screen shots!
One Light Source:
Two (Multiple) Light Sources:
Well, not too much to look at, but enough to get us excited. We’ve got rays tracing!
Below you should see a render with no reflection and no shadow detection.
What it does have is super sampling!
First with no super sampling:
And with super sampling:
