|
The Game Development Research Group (GDRG) focusses on research into computational issues
arising from problems in computer and video game development.
The Director is Ian Parberry.
GDRG is housed in the Laboratory
for Recreational Computing, whose mission is research and education in computer games.
Students can get credit in research in the GDRG by signing up for one of the following classes with Dr. Parberry, although it is also possible to volunteer for projects without signing up for credit.
-
Undergraduate: Directed Studies or Special Topics.
-
Master's: Directed Studies, Special Topics, or Thesis.
-
PhD: Directed Studies, Special Topics, Independent research, or Dissertation.
If you are interested in joining this research group, please contact
Ian Parberry.
Some of the research projects currently underway are described below. You
may join one of these projects, or you may propose your own. But first, take a moment to
examine GDRG's research philosophy
and some publications from recent projects
so you'll know what you're getting into.
Procedural Generation
The so-called next-gen computer games require a much heavier burden of content,
in the form of models, images, layouts, and design than previously allowed. This
means that game companies need much larger teams of artists and designers to
fully take advantage of next-gen technology. To assist this endeavour and take
the brunt of the repetitive and noncreative work, game developers have long wanted to
generate this content automatically, a process known as procedural content generation.
Research in this field must balance three requirements:
-
Speed: content must be generated in real-time using current computer hardware.
-
Quality: content must be compelling, interesting, and varied, containing
both elements of structure and randomness.
-
Controllability: content must be created according to parameters that can
be set by designers in a natural and appealing way.
Students
-
Dhanyu Amarasinghe (PhD thesis), interactive fire. The aim of this
research is to have fire interact with objects in the world, that is,
fire should spread and objects should be consumed.
-
Josh Taylor (MS thesis), level generation, including room layout and content clutter.
-
Jonathan Doran (PhD, pre-thesis), procedural terrain generation.
The object of this project is to generate interesting, controllable heightmaps fast.
-
Mary Yingst (MS, pre-thesis), Faisal Rabbani (MS, pre-thesis), William Verhoeven (undergraduate).
Waves in water. The object of this project is to combine work on wave propagation
in deep and shallow water with dynamics of breaking waves to build a surfing game.
PS3 Cell Processor Array
The Sony PlayStation 3 also provides a large amount of computing power that can
be turned to various purposes. We have an array of 20 PS3s that we
are configuring into a supercomputer using Unix. (Pictures here.)
Students
-
Faisal Rabbani (MS, pre-thesis)
-
Thomas Bryan (undergraduate)
-
Jason Beck (undergraduate)
-
Douglas Cobb (undergraduate)
-
Vincent Faires (undergraduate)
-
Adam McEndree (undergraduate)
-
David Truty (undergraduate)
Cybersickness
Students
-
Vincent Liguori (PhD, pre-thesis)
The Peaceful Queens Problem
How many ways can n queens be placed on an n by n
chessboard in such a way that no queen threatens any other? The answer
is known up to n=25. We plan to use heterogenous computing array
to compute this number for larger values of n by using multi-core
computers, shaders, and PS3s. Some preliminary results for n=24
are available here.
Algorithm Explorer
Algorithm Explorer brings high-quality rendering to algorithm animation
using C++ and DirectX.
For more details, see the
Algorithm Explorer webpage.
Students
-
Erik Carson (MS thesis)
Miscellaneous
|
These are students affiliated with LARC who are either between projects,
or are just beginning their academic experience.
|
|
Students
-
Will Garner (MS, pre-thesis)
-
Tim McMahan (MS, pre-thesis)
-
Mary Yingst (MS, pre-thesis)
|
