One of the main goals of Virtual Reality is to provide immersive environments that take participants away from the real life into a virtual one, this is how Cave Automated Virtual Environment (CAVE) came about many years ago. Nowadays there are many of this kind of room-sized systems providing a superior Virtual Reality experience and are used for research into a wide range of disciplines including archaeology, architecture, art, biology, engineering, geometry, geology, medicine and healthcare, meteorology and physics. Nevertheless, for a good Virtual Reality user experience, it is necessary to have a processing system optimized for visual computing (based on CAVE-related features, Interaction, Application, etc.). In this work we propose an evaluation methodology for our Cave-like multi-VRmedia System. The proposal is based on three generic criteria: Performance, Usability and Relevance. The strategy tries to prove how assertive a system is when it comes to solving a problem.