Society project next year

[Jonathan G]

It was discussed at the meeting that we may have a 'society project' at the beginning of next year, to help new members get used to game design and for everyone to get to know each other. It could take place over a weekend, at the learning grid if nowhere else. Again we have the problem of computer availability, and this isn't really something that can be done remotely.

Anyway, we need to think of some simple ideas for things to make within this period. I suggest having a game engine already developed, or using a really simple language (DB), so people can get stuck in straight away and see results. Also it might be useful to have a small media library in case we don't get many artists/modellers/sound engineers.

As for game design, we could do something like wario ware with mini games which have simple game logic - actions either cause win or loss, and timeout causes loss. Or we could do something like the mario party minigames, which are a bit more involved and are usually multiplayer. Something modular would be good too, but I can't think of any ideas at the moment.

We'd need a good project leader too, someone who could plan exactly what needs doing while still allowing for creativity, and who can help people learn what they need to do, and delegate stuff without being too bossy. We don't want to scare people off

[richardhp]

couldn't agree more. probably best to have the basic engine already done, ready to flesh out with art-work and game-play. could this also be done in conjunction with a series of lectures about game-design/programming tutorials?

i like the idea of mini-games, which gives alot of scope for creativity without trying to do too much. also could just re-make an old classic like tetris or asteroids, nice and simple.

[Ali]

I think that's a good idea, and I wouldn't mind being project leader. I know DarkBasic now too, which would probably be a good language to use.

A bunch of short, simple games a la Wario Ware sounds perfect. It will probably be worth putting together a simple basic outline program to begin with (i.e. with a menu system for selecting the games, and some of the basics started). Putting together a bunch of useful 2D graphics is something I could do too.

I'm not sure remaking another game like tetris would be quite as fun, and wouldn't allow as many wacky ideas.

[richardhp]

yeah fair enough. should we come up with some prototype ideas for the mini-games, or leave that to the start of term? we want to make sure the games are not too complex so it might be worth coming up with a few ideas before-hand.

[Ali]

It might also be worth having leeway for people to come up with minigame ideas at the event itself. Having a bank of ideas we've come up with ourselves beforehand would be a very good idea, of course. If we had a flipchart or something for people to brainstorm on that might work quite well.

[richardhp]

i was thinking it might be hard to get people coding in one day since there is so much to learn. maybe we could make the main focus of the event about content and game ideas/design rather than actually coding a game. we could have stuff like 3d modelling and sound effects production, level editing etc, and get them to put it into a simple game engine, then maybe we could have them code really simple things.

[richardhp]

have we decided what we are coding this game in, and who's doing it?

[Jonathan G]

Because we're doing this on University computers, it's unlikely we'll be able to use XNA or DarkBasic. C++ with SDL and maybe OpenGL could be a possibility, but might confuse beginners. I've read a tutorial or two on PyGame and from what I've read it's quite slow. It would mean us learning a new language over the summer in addition to creating the framework and teaching it to people next year, however people from FreeStyleGames might be able to come in and give us some info. Java is another possibility, we can install Eclipse temporarily on ITS machines and some people in the society already have Java experience. It's a bit more user friendly than C++. There have been some Java games written already, could we adapt that code into some sort of framework?

People would write a single .java file containing a class which inherits from a base class with init(), update(time) and draw(time) methods, similar to XNA, and there'd be some code there already. There'd be simple methods to load and position 2D images - using or wrapping Java 2D - and load and play sounds and music. We might want to provide a simplified way of accessing mouse and keyboard input. People could then make helper classes and use OO principles, or just do things procedurally within the update() method. The complexity of the gameplay could be changed too from simple minigames to space invaders or a side scrolling shooter or something like a boss stage in Wario Ware.

The created class would be loaded from within a game framework, if a Wario Ware style game then a random class would be instantiated and have the interface methods called with parameters to specify the difficulty of the round, and the time remaining. Menus and stuff would be included in this framework (unless someone wanted to make them separately), and there'd be a simplified framework for people to test their minigames in, allowing them to run their game over and over with different starting parameters, and output debug stuff to the console.

Thoughts? Who would want to have a bash at this?

[Alan]

My Java games have a codebase that could be suitable for this. I haven't settled down on an IO interface that I'm completely satisfied with, but I think everything else is there.

I wouldn't mind making the game-loading framework either. I'd probably want to work out some kind of specification before I began though.

What's happening with the GameStrip project? It seems to have fairly similar aims, so if Bryan has started making a framework for that, it would probably make sense to use that.

[richardhp]

java sounds good, cs students have to learn it anyway and it is easier than c++.

[Bryan]

I'll make a post about GameStrip later today. There may be some crossover in the code, but if Alan has something that's tried and tested it may make sense to use that as a starting point.

Are there plans in place for how this event will work beyond the coding side of things?

[Jonathan G]

We were thinking of having the event over the weekend, over 4-5 hours each day. ITS won't let us book computer rooms so we'll have to find some in the Learning Grid - hopefully it won't be too busy at the beginning of term. We'll encourage team work, probably between programmers, artists and sound people, or multiple programmers (on the same machine, perhaps?) so it's a bit of a social activity too. To lower the amount of commitment needed, people will be able to drop in and out. This means that learning will have to be in the form of tutorials and examples so people can come along whenever and learn at their own pace. The tutorials should be short, 15 minutes would be ideal but it'll change for different levels of expertise. As well as practical tutorials (code/art) there should be some stuff on game design too, especially the philosophy of the overall game we're making. Using the Wario Ware example, the philosophy of each minigame is to be based on a single verb and be completed in a matter of seconds. The challenge lies in deciphering what the game requires, followed by reactions or dexterity. However it would be good if we could come up with another idea that uses a similar framework, I don't want to rip off Wario Ware too much.

[Dan]

I don't want to rip off Wario Ware too much

I wouldn't worry about that, noone else making Wii titles seems to be. Zing!

[Alan]

Does anyone have any suggestions for how the framework should deal with menus? I've just been creating them with Swing for my games, but a simpler solution would be better, I think.

By the way, if we intend to look at Python, there's a week-long competition using PyGame at the start of September: http://www.pyweek.org/5/

[NickF]

Didn't we plan to do this a year or two ago?

The upshot is that's there's some code in the legacy folder in SVN (called MiniGames) you might find useful.

I think the main bit was some dynamic class loading code. That was when the plan was to have an app which could load minigames in a plugin type fashion - Although this won't be so useful if you're going web-based (I'm pretty sure applets don't let you do this). Still, it should help you out with instantiating a random game class.