Next-generation tools for next-generation games, an EngD studentship in collaboration with Reflections (a Ubisoft computer games studio)

Fully-funded Doctoral Studentship

Applications are invited for an Engineering Doctorate (EngD) in the Centre for Digital Entertainment (CDE), a collaboration between the University of Bath and Bournemouth University and Reflections, an Ubisoft Studio in Newcastle.

RESEARCH AREA
With next generation gaming consoles on the horizon, players' expectations are set to rise once again. This means game developers will need to up their game even further, resulting in:

More in depth, and social online gaming experiences.
Even larger and more varied and detailed play spaces.
Increased texture and geometry resolution.
Better artificial intelligence resulting in more immersive gameplay.

Development costs will increase to unsustainable levels without powerful tools to ensure content creators can build assets quickly, and iterate rapidly on the target hardware. In addition to make effective use of the next-generation of hardware will require innovative approaches to multi-threading and GPGPU (General Purpose computation on Graphics Processing Units) usage.

There are many tools available already to assist artists and designers to generate textures, geometry, cities and larger environments procedurally with varying degrees of artistic control. Typically, artistic control is provided by manipulating the input parameters or seeds to the procedural tool, but changes to the generated data are often lost when the artist wants to fine tune the results in Photoshop, or the 3D package of their choice.
We would like to explore this further, to see how artists can be given more control over the procedural generation process in an intuitive fashion.

Processing power is increasing at a faster rate than I/O performance. With ever larger playing environments becoming commonplace, streaming environment data from the disc with sufficiently high detail is becoming a bottleneck. Moving the procedural generation of content into the run-time space is one solution to this problem, and we’re already seeing this with run-time tessellation of geometry and displacement mapping.
We would also like to investigate other avenues of run-time procedural content generation.

The ability to iterate on game content rapidly is one of the keys to AAA quality. However there are increasing demands on the system when using procedurally generated content, pre-processing of the environment to improve lighting fidelity, automated navigation mesh generation for path-planning, geometry conditioning the list goes on. Typically changes to the environment require offline pre-processing steps to regenerate this content, which hampers productivity, either by introducing latency into the iteration steps or requiring the content creators to work with non-final content – lighting, navigation data and so on.
We would like to investigate ways of using distributed cloud computing to improve iteration times for content creators allowing them to work with near final quality content on the target platforms.

REQUIREMENTS
Applicants should hold a first degree, or equivalent, of at least an Upper Second (2.1) in a relevant subject, such as, but not limited to, Computer Science, Maths, Physics, Software Engineering, Computer Arts, Animation, Digital Effects, Design or Interactive Media. Knowledge of programming languages, such as C++, Python, or industry standard software is a distinct advantage.