摘要：En el artículo se presentan los pasos para simular y visualizar multitudes masivas variadas y animadas en tiempo real, usando el procesador gráfico (GPU). En particular, se discutirán los métodos para la simulación de comportamientos, nivel de detalle, animación y generación de personajes variados. Dada la arquitectura de estas técnicas, se pueden extender a clusters de GPU o en sistemas de cómputo de alto rendimiento (HPC). Estos sistemas son escalables casi linealmente si se incrementa el uso de nodos, sin embargo se deben combinar técnicas de simulación y rendering paralelos. Sin embargo, dados los avances tecnológicos recientes como plataformas de cloud gaming, estas técnicas están listas para funcionar en dichas plataformas
其他摘要：We discuss several steps in the process of simulating and visualizing large and varied crowds in real time for consumer-level computers and graphic cards (GPUs). Animating varied crowds using a diversity of models and animations (assets) is complex and costly. One has to use models that are expensive if bought, take a long time to model, and consume too much memory and computing resources. We discuss methods for simulating, generating, animating and rendering crowds of varied aspect and a diversity of behaviors. Efficient simulations run in low cost systems because we use the power of modern programmable GPUs. One can apply similar technology using GPU clusters and HPC for large scale problems. Such systems scale up almost linearly by using multiple nodes. One must combine parallel simulation and parallel rendering in the cluster with interaction and final rendering in lighter clients. However, in view of the latest developments such as the new family of mobile multicore chipsets and GPU-based cloud gaming platforms, the pieces are almost there for this kind of architecture to work.
关键词：Simulation; real-time crowds; rendering and animation;Generación; simulación; animación; visualización; multitudes; tiempo real
