Creative Assembly is collaborating with IGGI, The EPSRC Centre for Doctoral Training in Intelligent Games and Game Intelligence, on two exciting PhD proposals with the opportunity for placement at Creative Assembly UK, and mentorship from industry professionals.

Students are invited to apply here now.

The project proposals include:

HIGH PRECISION BATTLE PREDICTIONS IN RTS GAMES

The goal of this project is to investigate how to build a model for a precise and fast battle simulator from past played battles, which can be used to provide insights and more personalized predictions based on the player’s play style and army  composition. The simulation should provide instantaneous battle results and details such as how many units were killed or captured on both sides and how much damage each unit took. It should also balance between overly favourable or very unfavourable outcomes, represent basic playing strategies, and take into account highly valued units, arbitrary priorities of the armies, different battle scenarios and the types of units involved.

MODELLING PLAYER BEHAVIOUR IN TOTAL WAR

In Total War games -both in campaign and battle mode- players have various play styles and different ways of using in-game mechanics (such as spell systems, or economy), features (such as units, armies or maps) or game controls (such as interaction with UI elements or camera). The goal of this project is to gain insights into our players and their play styles, utilizing our in-game metrics data and player modelling techniques. We would like to explore this area mainly (but not limited to) with two goals in mind:

• Player-like AI: Creating player-like opponent AI using these models would have benefits in improving and play-testing our current AI. We can additionally utilise such AI for creating opposing strategies that would lead us to acquire more dynamic personalities in game.

• Data informed game design: We would like to understand the impact of our gameplay features on our players and inform the game design direction with this understanding. Carefully created and analysed player models could lead us to discover relations between play styles and in-game mechanics, features or controls.