Treaps, which combine the properties of binary search trees and heaps, originally appeared under the name of randomized binary search trees by Aragon and Seidel in 1996. Priorities are i.i.d. uniformly distributed random variables on the unit interval $[0,1)$. Treaps have good expected time complexity for several operations, including insertions, deletions, and searches under the former assumption that priorities are i.i.d. uniformly distributed, which implies that few rotations occur when inserting or deleting keys. However, one must have to store about $O(\log_{2}(n))$ random bits in every node for a treap of $n$ nodes. Several attempts to partially derandomize treaps occurred, including seeded pseudo-random sequences, which allows for computing priorities on the fly; some additional statistical assumptions must be made on the type of pseudo-random sequences, too. In this talk, I will discuss a simple C implementation and related excellent empirical results about the number of rotations required for insertions and deletions when elements of a deterministic equidistributed sequence represent priorities.