The main objective of this project is to parallelize Lindenmayer systems algorithm with CUDA toolkit. Implementation is performed both on Central Processing Unit (CPU) and Graphics Processing Unit (GPU). Speed gain from parallel implementation is measured on a GTX 660 graphics processor.
Tests and Results
All tests have been performed on a PC with Intel Core i7-3770 CPU @ 3.40 GHz and 8.00 GB RAM. The graphics processor is Nvidia GTX 660 with 1152 CUDA cores and 823 MHz Base clock. The operating system is Windows 8. Tests have been done on both CPU and GPU to compare the computation times. Algorithms tested with 6 different axiom and rule pair.
An axial tree generated with L-system algorithm on CUDA
