ABSTRACT
In practice, it is very frequent that people jump directly to conclusions. So if you did, you might as well continue reading because in this case, your choice of starting with the conclusion rst indicates what you are aŸer: solutions to a particular problem, not necessarily an exhaustive overview. is book provides a su›cient amount of details to understand and imple-
ment the major techniques. We describedmany techniques in order to give a good overview of the state-of-the-art solutions. But why did we present so many methods? Why not just the best solution? In fact, some readers might even feel a little overwhelmed by the variety of possibilities. e simple reason is that there is no ultimate shadow algorithm at the current moment in time. We have many solutions that all address certain situations and needs. Further, it can make perfect sense to combine several methods to yield a system that provides an even better behavior. We saw such a situation in Chapters 3 and 4 where reparametrization of shadow maps and partitioning lead to a better overall behavior. We feel it is therefore of large benet to have a good overview to nd the most adequate solution for a particular situation. In the following, we will investigate several application scenarios and point out
which algorithms might be of high interest to you.
