Random numbers can be problematic in certain applications such as graphics and game development because they have a tendency to clump together. This can cause issues in tasks like path tracing, Monte Carlo integration, and randomly placing objects in a game. To address this problem, low discrepancy sequences can be used to generate random numbers that are more evenly distributed. These sequences have a lower discrepancy, meaning that the density of points is more balanced. Different techniques, such as subrandom numbers, irrational numbers, and Halton sequences, can be used to generate low discrepancy sequences. These sequences can improve the convergence and overall quality of the random sampling in various applications.
https://blog.demofox.org/2017/05/29/when-random-numbers-are-too-random-low-discrepancy-sequences/