The use of digital technologies in teaching and learning processes has opened new and wide frontiers in mathematics education. Research in education moves towards teaching and learning (and assessing) twenty-first-century skills and computer-based testing seems an interesting possibility for assessing such skills. From this perspective, comparing computer-based with paper and pencil testing is an important issue for the educational community. It could allow us to better understand differences, innovations and weaknesses of classical assessing strategies and methods. The purpose of this study is to provide an example of the differences that can be found in migrating a task from one environment to another through a qualitative analysis of task and students' solution processes. The sample involved in this study includes grade 6 and 8 students. The analysis shows that the environment migration process is not neutral: the tasks present differences. Students' performances in the tasks are the same, but differences can be observed in solving processes. In particular, students of our sample differ in the mathematical resources applied, and in the assessing processes. In addition, it seems that students activate similar problem-solving processes as they become more familiar with the task and the environment in which it is administered.