Depth perception is one of the key issues in virtual reality. Many questions within this area are still under investigation. Among these last, one can find the intriguing and pending question related to distance misestimation. This phenomenon is arising when head mounted devices are used to display virtual worlds and the factors leading to this fact are still unidentified. In this paper we describe an experiment confirming distance underestimation from another point of view. The approach we developed is based on a very simple task: subjects had to compare relative depths of two virtual objects. The experiment was performed with both head mounted devices and immersive wide screen displays allowing us to consider that subjects use the same visual cues to estimate depth. To avoid motoric effects, subjects were seated and their estimations were only verbal. Likely, to avoid the well known apparent size effects (namely, the size-distance invariance), the experiment was performed under conflict cases: the viewed objects had the same apparent sizes with different depths or the same depth but different real sizes. The obtained results show significant differences between the two devices and confirm the distance misestimation phenomenon for head mounted devices.