The Ivory Tower

"One type of enquiry we get is where a question is already solved, and you can just point them to some research, or consult for them, or send over some software," explains Colton. "That's not really what we're interested in - we want problems that don't have obvious solutions, something where we fundamentally don't know the answer so we have to do lots of experimentation, lots of old-style research."

"Whenever I give talks to games people," he continues, "They say things like, 'But that's really difficult, I can't imagine a solution to that.' I say, 'Yes! Exactly! That's why it's research! That's exactly what we're here for.' Most problems within games companies can be solved with enough manpower - we want to talk about fundamental problems, when we don't even know how to phrase the question properly yet, let alone what the answer might be. If games companies can think about things in that way and then come to us, the potential is immense."

Imperial is confident that the research it's undertaking now with games companies such as Rebellion - and which it hopes to undertake in future with other firms ("We could see a bit more of the bigger players," Kelly muses, observing that their contact thus far has largely been with small- to medium-sized British developers) will help to shape the future of video games. Work such as that undertaken by PhD student Robin Baumgarten, who created the dynamically adapting Pac-Man game mentioned earlier, has obvious, near-term practical applications.

"You buy the game and someone else buys the game, and within a week, you're playing two different games because it's adapted to your playing style," says Colton. "But developers don't know how best to do that, so it's a great research project."

Yet for all that, the university's researchers seem unanimous on one point, perhaps surprisingly. While they'd love to engage more with the games business, the industry also needs to look to itself and start recognising the value of fundamental research - rather than constantly fixing its eyes on the next deadline.

"They absolutely have to develop a culture of research and development, like other big industries do - like the pharmaceutical industry does, or the telecoms industry does," says Colton. We discuss the Google model, where the company gives its employees 20 per cent of their time to work on their own projects. This approach has created major products for Google, including the near-ubiquitous Gmail.

"That would be amazing," Colton says. "They need to see past the next game - it's damaging, how much everything is focused on that next launch."

But he admits that sadly, that's a pipe-dream for most of today's financially constrained developers. It could well be that the future of gaming is not being brewed up in a gleaming development office somewhere on a sterile industrial park in California. Instead, it could be coming to life in a bedsit flat, the work of a PhD student who is busily doing research that the games industry can't afford, can't spare the time for and perhaps doesn't even know that it needs.