I submitted my dissertation last week, exhausting my coursework commitments at long last. I won’t be graduating until next summer due to Brunel’s timetabling for such events, but I’ve now started actively looking for work. Entry-level game designers positions are hard to come by, but I won’t bore you with that just yet. Instead, I’ll bore you with a brief overview of my dissertation. Entitled “Gaming in the Round: The Overlooked Potential of Player-Controlled Perspective”, it examines how and why perspective controls are given to the player, and the effects on the play experience.

Pretty much all games these days let players control their view of the game environment – most of the dissertation focuses on first- and third-person games like Quake or Mass Effect, but even Catan lets you zoom and rotate the game board – but I think, with very few exceptions, these perspective controls are implemented in the same way, and for the same reasons, as they have been since the mid-90′s. Back then, a whole load of hardware developments – 3Dfx cards, the PlayStation, Saturn, and N64 – radically accelerated 3D graphics rendering in a short space of time, and suddenly every game was being adapted to take advantage of this.

Early 3D games could be pretty hit-or-miss. I remember playing Jumping Flash! on a rented PlayStation back when I was in middle school, and really enjoyed bouncing around the large, open spaces, but as a game that tries to combine platform sensibilities with FPS action it pretty much fails on both counts. At the same time, I thought Quake 2 was unrelentingly awesome, and spent days making my own maps and skins to trade with my friends on floppy disk. But then, to cut a long story short, Mario 64 came out and blew all other games out of the water.

The camera in Mario 64 works really well for two main reasons. Firstly, it was positioned much higher than in most other 3D platformers at the time, so you were usually looking down on Mario and his surrounding area instead of trying to look over his shoulder; Secondly, you could zoom and rotate the camera around him by tapping the handy C buttons (and switch to a fixed chase-cam using R, but that’s not really important), which meant that if you couldn’t immediately see what you wanted to look at, you could flip the camera round in a fraction of a second. And basically, this second principle has become standard practice in pretty much all modern 3D games, without much of a second thought.

Instead of just compensating for the inherent problems of displaying 3D spatial information on a 2D screen, I think this kind of perspective control can be used to really add to the game experience. In the dissertation, I look at the way audience-controlled perspective has been used in other art forms throughout history – through things like statues, theatre and landscape gardening – and how particular games have used it in interesting ways (although not always successfully, as Night Trap demonstrates). My main conclusion is that, if perspective control is to have any meaning, you have to give players meaningful choices to make in how they use it – they need interesting things to look at, but also interesting things to not look at. Much of my theory is drawn from Janet Murray’s ideas about ‘multi-stage narratives’ in Hamlet on the Holodeck, but with some other bits about non-linear narrative taken from Espen Aarseth’s Cybertext.

To back this up, I came up with a little game design that puts this theory into practice. Ice Station Lemur is a 5-minute ‘cyberdrama’ in which players watch a group of scientists and engineers engage in a tightly-knitted, EastEnders-quality theatre. It has little value as a game design in itself, but I think it demonstrates a few important principles quite well. The main lesson I’ve tried to make through all this is that games have a unique potential for not just explorable environments, but explorable situations. Instead of just herding players through topologically linear narratives, designers should present player with noticeably different routes to follow.

Come to think of it, I’m basically just advocating Metal Slug 3‘s multicursal level design.

My dissertation can be downloaded here and the accompanying design doc can be found here.