Games are for fun by Mr Jack

Well I've been working away for a couple of weeks now, and it occurs to me I've not actually explained anything about the game we're making yet. It's a simple 2D physics puzzler, in which you attempt to get across the screen by swinging off objects. There will be two game modes, one consisting of a series of independent puzzles of roughly increasing difficulty, and the other a ongoing challenge in which you attempt to travel as far as possible across linked screens containing a variety of objects and challenges. So far I've been working on the latter of those two modes, but 95% of the code is common to both anyway.

The game has very simple controls, you just touch the object you want to swing off, hold to keep swinging and release to let go. This was initially the only controls I added, and it was enough to make the game both fun and challenging. But on it's first public outing I showed it to a friend who made the obvious observation I'd completely missed: why couldn't you swing higher by tilting the iPod? Doh! Why didn't I think of that? A few minutes of coding later and "tilt swing" was born. Rocking the iPod back and forth makes your character swing higher in a way that feels instantly natural and opens up new areas of game play to be explored.

If you look at the videos we've posted so far, you'll doubtless notice the lack of polish and graphics in the game. This is a feature, not a bug. The more design goes into a game, and the more time spent on level designs and art, the more fixed it becomes because changes mean throwing more and more stuff away. At the moment, it's easy to put new game mechanics - like tilt swing - in, and easy to explore new game designs and modes (today I've been experimenting with levels in portrait rather than landscape orientation). This kind of tweaking and exploration is vital to producing games that are really good fun to play. Whilst the ability to envision how a game will play while its only a doodle on a bit of paper is one of the skills that distinguishes the best game designers; there's no substitute for actually trying stuff out.

Many years ago, when I was still at my first games programming job we were working on a racing game (no, I'm not saying which one it was) in which the tracks were all modelled in Maya (for those of you that don't know, Maya is a 3D art package popular in the games industry). Whole worlds were built around these tracks that had never been anything more than sketchs, polygons were placed, tweaked and textured; whilst over on the other side of the room one of my fellow programmers was busy putting together the car physics, complete with a modelled clutch and drive train. Then, many months later, the tracks were nearing ready and the cars could, for the first time, be dropped onto track.

They couldn't get up the hills with their 70 degree inclines.

But by this point, the tracks had literally hundreds of man-hours invested in them and making changes would have meant weeks of artist time; time we didn't have. So, instead, the car physics got bent and botched until they could get up the hills and, eventually, after a fashion you could race round the tracks. But because they hadn't been - and couldn't be - tweaked and tuned to the way the AI and the physics worked the tracks were too wide, it was too easy to pass and the corners weren't much fun.

That's not a mistake we plan on making, so much of this week will be spent tweaking, trying and exploring designs while the game's still fresh and maleable and before the art has been locked down. By the end of the week we should have the game - sans graphics and polish - playing much how it will when it's finished.