SBArt4 – Breeding abstract animations in realtime

Tatsuo Unemi
Department of Information Systems Science, Soka University, 1-236 Tangi-machi, Hachioji, Tokyo, 192-8577 Japan
IEEE World Congress on Computational Intelligence, WCCI 2010


   title={SBArt4–Breeding Abstract Animations in Realtime},

   author={Unemi, T.}


Download Download (PDF)   View View   Source Source   



SBART was developed in early 1990’s as one of the derivatives from Artificial Evolution by Karl Sims. It has a functionality to create a movie from a bred image through post-processing. The innovation of graphics processing unit (GPU) in these years improved the calculation performance to be fast enough to realize breeding animations in realtime on the personal computer. SBArt4 utilizes the advantage of GPU by compiling each expression in genotype into a type of OpenGL shading language. Even when it renders each frame of the animation in realtime, it achieves enough speed for users to evaluate the product of an abstract animation immediately. Though there were a number of problems because of architectural difference between CPU and GPU, almost compatible functionalities with the previous version have been implemented including reference to an external image and integer-based bitwise operation. Through experimental executions on some different hardware configurations, it was certified that it runs fast enough on recent consumer machines though some older machines are not powerful enough.
No votes yet.
Please wait...

* * *

* * *

HGPU group © 2010-2021 hgpu.org

All rights belong to the respective authors

Contact us: