In recent years, computational aesthetics, which connects science and art, is becoming a new interdisciplinary field that brings technology into art creating. With advanced CAD technology and artificial intelligence, computational aesthetics has been widely used in many design fields such as architecture and industry design. Famous design companies such as Foster+Partners, Zaha Hadid and BIG have already adopted computational and parametric design method as a competitive way to enhance creativity. But there are still some argues that this new aesthetics is superficial or even cold. In terms of generative art, as an advanced digital art, reveals a new aesthetic power, which is impressive. In this context, the purpose of the study aims at comparing computational aesthetics with aesthetics to reveal their differentiation and correlation that benefit for designer to fresh the cognition of this new approach vice versa. As methods of this research, firstly previous researches and literature are collected and comparatively analyzed on the concept and relationships of aesthetics and computational aesthetics, as well as the assessment method. Secondly, through case studies, under the characters of these two aesthetic approaches in design field, three representative cases are selected to be deeply investigated by collecting and analyzing relevant articles, books and online materials to explain the respective features. Then, after comparison and further analysis, explaining how computational aesthetics as a rational aesthetic expression can augment the limit of human aesthetic experience and the design thinking process. The final results of this paper point that aesthetics dominated by emotional mind belongs to a top-down thinking process, while computational aesthetics because of its algorithmic generated basis, which dominated by logic mind, it belongs to a bottom-up thinking process. It is expected that the findings based on this research can provide certain guidance for designers to make good use of them.