The first performance consideration, for this project, was with respect to thread divergence. As you can see in the above image, from the course notes, the means by which a bsp tree processed is similar to the reduction function example from the course notes [[File:Cudasplitblock. png|500px]]
[[File:Cudasplitblock.png|500px]]
[[File:Fig04-z-curve.png|500px]]
[[File:Fig04-z-curve.png|500px]]
[[File:Fig06-numbering.png|500px]]