Life After GaME

Did you come to this year’s Games and Media Event at Imperial College? If you did, thank you so much! I hope you had a great time. I can honestly say that this year’s event was the best we’ve organised yet. Over the course of six or seven years the event has slowly grown in scope, brought in a wider spectrum of attendees, and is now a driving force for getting students interested (and into) jobs in the games industry. Thank you again if you came, and particularly if you were one of our excellent speakers!

The day before GaME I wrote up and submitted a paper to AIIDE, which means I’m relatively deadline free for a month or two. It’s a nice feeling! I get this about once every six months at the moment, and it’s a good time to reflect on how the PhD is going, as well as a nice justification for playing with side projects. One of the purposes of this blog is to keep people in the know about what I’m working on, so here’s a quick list of the things I’m pottering between these days.

1. New Game Libraries
Having worked for a reasonable length of time on platform games, I imagine the remainder of my PhD will focus on this area and attempting to innovate within it. That said, I’m going to return to working on APACE (the point-and-click library implemented on top of Slick2D). The main push behind this is the hope that I might be able to dabble in this genre one day, particularly since the work of other researchers like Clara Fernandez-Vara is so incredibly fruitful and exciting. I’d particularly love to start looking at techniques for good, linear narrative generation. Point-and-clickers seems like a good platform to work on.

I’ve also gone back to a library that I didn’t bother mentioning before, unnamed, designed to produce SNES-era JRPGs. It’s so basic it’s not worth mentioning almost, but again this is a genre that seems very amenable to automated design systems, and offers a chance to investigate notions like narrative, the design of branching game flows, and the integration of system-balancing techniques into full game design systems. System balance is quite interesting to me – I’ve started trying to design RPG systems lately just to see how hard it is, and to me the task is incredibly complex. There’s research out there applying computer science techniques to it, so it’d be great to integrate these into a larger system.

2. More Creativity
I gave a talk at GaME about the latest work I’ve been doing with ANGELINA. While it’s very nice to look at, I’m sure a few in the audience were skeptical. ANGELINA still isn’t very intelligent, and has only just begun to move towards creative autonomy. In particular, ANGELINA needs more flexibility in the kind of input/starting points it can work from, and a lot┬ámore capability in the game design stakes. We need to start producing games that don’t all play out identically. This will probably necessitate conceptual databases like ConceptNet, something which I’m super-excited about.

First steps here will be to go back to the mechanical level of game design and give ANGELINA some new systems for planning game flow. I’ve put it off for months, but finally ANGELINA will be able to calculate where you have to go in a given game, in what order, and what is reachable from all of those areas. That information is crucial in pacing, puzzle design and so on. That will be a heavy implementation task, but worth it. If only London had beaches with free wireless.

3. Game Data Analysis Project
This isn’t actually my project, it’s a fellow researcher’s, and I don’t want to describe it as it’s his baby and he’ll announce it as and when he wants to. But I said I’d help out on a project aiming to classify players into different activity groups based on their behaviour. We’re going to be working on a really interesting game, running some public tests, and hopefully getting lots of exciting and interesting results. I’ll write up about this if he’s happy for me to, and let you know how that gets on. You can also join in and let us stalk you in digital worlds! Hooray!

So that’s my summer in three bullet points. I hope you’ll continue to keep in touch through the blog and on Twitter. More games by ANGELINA will be up soon! If you’re in Dublin next week for the International Conference on Computational Creativity, don’t miss out the demo session! I’ll be there creeping everyone out with ANGELINA.

Leave a Reply