Darwin’s Demons: Does Evolution Improve the Game?
It is widely assumed that evolution has the potential to make better video games. However, relatively few commercial games have been released that use evolution as a core game mechanic, and of these games only a very small sub-set have shown that evolution occurs as expected and improves game play as intended. Thus, there remains a critical gap between studies showing the clear potential of evolution to improve video games and studies showing that evolution did improve game play in a commercially released game.
We have developed Darwin’s Demons, a space shooter inspired by old style arcade games, with the added feature of evolving enemies. In August, 2016 Darwin’s Demons was Green-lit for sale on Steam, a standard benchmark for commercialization of games. In this paper we present and test four hypotheses that form the basis for the claim that evolution occurs and improves game play in Darwin’s Demons. More generally, these hypotheses can be used to confirm that evolution meets the intended design goals for other evolutionary games.
Our results support the hypotheses that evolution makes Darwin’s Demons get progressively more difficult over the course of a game, and that the fitness function, player choices, and player strategy all affect the evolutionary trajectory during a single game. This suggests that in Darwin’s Demons, the enemies adapt to the player’s decisions and strategy, making the game interesting and increasing its replayability.
KeywordsEvolution Games Procedural generation
- 1.Hastings, E.J., Guha, R.K., Stanley, K.O.: Evolving content in the galactic arms race video game. In: 2009 IEEE Symposium on Computational Intelligence and Games, pp. 241–248. IEEE (2009)Google Scholar
- 2.Muñoz-Avila, H., Bauckhage, C., Bida, M., Congdon, C.B., Kendall, G.: Learning and game AI. Dagstuhl Follow-Ups 6 (2013)Google Scholar
- 3.Risi, S., Lehman, J., D’Ambrosio, D.B., Hall, R., Stanley, K.O.: Combining search-based procedural content generation and social gaming in the petalz video game. In: Aiide. Citeseer (2012)Google Scholar
- 4.Risi, S., Lehman, J., D’Ambrosio, D., Hall, R., Stanley, K.: Petalz: search-based procedural content generation for the casual gamer (2015)Google Scholar
- 6.Stanley, K.O., Bryant, B.D., Karpov, I., Miikkulainen, R.: Real-time evolution of neural networks in the nero video game. In: AAAI, vol. 6, pp. 1671–1674 (2006)Google Scholar
- 7.Jallov, D., Risi, S., Togelius, J.: Evocommander: a novel game based on evolving and switching between artificial brains. IEEE Trans. Comput. Intell. AI Games (2016)Google Scholar
- 9.Liapis, A., Martınez, H.P., Togelius, J., Yannakakis, G.N.: Transforming exploratory creativity with delenox. In: Proceedings of the Fourth International Conference on Computational Creativity, pp. 56–63. AAAI Press (2013)Google Scholar
- 10.Cachia, W., Aquilina, L., Martínez, H.P., Yannakakis, G.N.: Procedural generation of music-guided weapons. In: 2014 IEEE Conference on Computational Intelligence and Games (CIG), pp. 1–2. IEEE (2014)Google Scholar
- 11.Hoover, A.K., Cachia, W., Liapis, A., Yannakakis, G.N.: AudioInSpace: exploring the creative fusion of generative audio, visuals and gameplay. In: Johnson, C., Carballal, A., Correia, J. (eds.) EvoMUSART 2015. LNCS, vol. 9027, pp. 101–112. Springer, Cham (2015). doi: 10.1007/978-3-319-16498-4_10Google Scholar
- 12.Schumacher, J.: Species (2016)Google Scholar
- 13.Schrum, J., Miikkulainen, R.: Constructing game agents through simulated evolution (2015)Google Scholar
- 14.Avery, P., Togelius, J., Alistar, E., Van Leeuwen, R.P.: Computational intelligence and tower defense games. In: 2011 IEEE Congress on Evolutionary Computation (CEC), pp. 1084–1091. IEEE (2011)Google Scholar