This quasi-experimental study attempted to investigate which mode of language games, paper-based or computer-based, can better expand the English vocabulary size of Form Four students especially at the 2000 word-level. Seventy students were involved in this study i.e. thirty-five in each group. The study was conducted in three stages. First, the participants sat for the Productive Vocabulary Levels Test (PVLT) as a pre-treatment measurement. Then the Experimental Group 1 played the computer-based games and the Experimental Group 2 played the paper-based games for seven weeks before sitting for the first post-treatment parallel PVLT. After that, both groups switched treatments before they were assessed again using the second parallel PVLT. The statistical analysis of the pre-and post-treatment test scores were done using the t-test. The results show a significant gain for both modes of games, but the computer-based games had a higher mean gain. The study provides evidence that the computer-based games had better influence on students' vocabulary enhancement than the paper-based games.