|
|
GPU Accelerated Adversarial Search
Brehovský, Martin ; Bošanský, Branislav (advisor) ; Bída, Michal (referee)
General purpose graphical processing units were proven to be useful for accelerating computationally intensive algorithms. Their capability to perform massive parallel computing significantly improve performance of many algorithms. This thesis focuses on using graphical processors (GPUs) to accelerate algorithms based on adversarial search. We investigate whether or not the adversarial algorithms are suitable for single instruction multiple data (SIMD) type of parallelism, which GPU provides. Therefore, parallel versions of selected algorithms accelerated by GPU were implemented and compared with the algorithms running on CPU. Obtained results show significant speed improvement and proof the applicability of GPU technology in the domain of adversarial search algorithms.
|
|
|
Real Time Visualization of Chaotic Functions
Teichmann, Antonín ; Elek, Oskár (advisor) ; Wilkie, Alexander (referee)
Fractals are a fundamental natural structure that has fascinated the sci- entific community for a long time. To allow for better understanding of fractals, visualization techniques can be used. The focus of this thesis is real-time rendering of fractals that are similar to the Mandelbrot set or the Newton fractal. Detailed exploration of these fractals is complicated due to their recursive-manner which leads to the fact that rendering them is com- putationally demanding. Existing solutions do not work in real-time or have low visual quality. We want to change that and allow high-quality real- time rendering. During our analysis of the problem, we generalize fractals to chaotic functions. To achieve high-quality rendering with low overhead, we introduce a method for adaptive super-sampling of chaotic functions. To achieve real-time performance, we show how to use sample reuse, foveated rendering, and other techniques. We implement a parallel, GPU-based, high- quality renderer that runs in real-time and produces visually-attractive views of given fractals. The program can visualize any given chaotic function. This way, we open the realm of real-time visualization of chaotic functions to the public and lay a basis for future research. 1
|
|
|
GPU Accelerated Adversarial Search
Brehovský, Martin ; Bošanský, Branislav (advisor) ; Bída, Michal (referee)
General purpose graphical processing units were proven to be useful for accelerating computationally intensive algorithms. Their capability to perform massive parallel computing significantly improve performance of many algorithms. This thesis focuses on using graphical processors (GPUs) to accelerate algorithms based on adversarial search. We investigate whether or not the adversarial algorithms are suitable for single instruction multiple data (SIMD) type of parallelism, which GPU provides. Therefore, parallel versions of selected algorithms accelerated by GPU were implemented and compared with the algorithms running on CPU. Obtained results show significant speed improvement and proof the applicability of GPU technology in the domain of adversarial search algorithms.
|
guest ::
