Evolutionary algorithms are effective, robust methods for solving many practical problems such as feature selection, electrical-circuit synthesis, and data mining. However, they can take a long time on some difficult problems because they need to perform several fitness evaluations. Parallelizing these algorithms is a promising way to overcome this limitation. The authors propose to implement a parallel EA on consumer graphics cards. Experiments demonstrated that this parallel EA is much more effective than an ordinary EA, achieving between 1.25 to 5 times greater speed using a current-generation graphics card. Because most personal computers have graphics cards, and these computers are easy to use and manage, more people will be able to use the parallel algorithm to solve their real-world problems.