2001 年 66 巻 541 号 p. 57-62
The evaluation of the visual impressions of architectural spaces is often carried out with various simulation systems. In this paper a newly developed VR system (MOON) and some other simulation systems are compared for their practicability as a medium for evaluating architectural spaces. The evaluation by semantic differential method results essentially in little difference in bases for evaluation between those in real and simulated spaces. And it suggests that the resultant values are likely to have hierarchical relations rather than linear ones to the distance between the place where subjects made evaluation and the main object to see in the space.