Slashdot Mirror
Game Design Classes?
Scott 'dolo' Leonard asks: "A while back, you might remember, I taught about 200 people in an IRC level design course, where students would complete a number of assignments for grading and grow with the course to meet their personal goals. Right now I am in the process of working up a proposal for my local college to get a game design course approved for Fall 2002. This course would hammer out the fundamentals of designing a game, what works and what doesn't. Before I begin, I was hoping that the community might have some input as to what they would want in a course of this nature."
"Basically, this would be an IRC class where students would walk away with something tangible regarding game design, and approaches therein. Obviously we would focus on the aspects of games today, and different types of games, but we would also have to look at a regiment of building smart games that impact the bottom line of a company positively.
Send all input to sleonard@planetquake.com and I will post any and all comments to Dteam, where applicable. Please contact me if you have any comments."
Umm... doesn't this kinda defeat the purpose of an Ask Slashdot? You know, where we all get to do the communal comment thing and learn from one another and, if we've all been really good, make s'mores at the end of the night?
As so: I'd honestly love to see a course that deals with that elusive first hurdle -- getting startup funding. Where to start looking, how much proof-of-concept is needed, etc. Seems like we've all got ideas for that one perfect game (just like everyone claims to have that one great novel in them) but no concept of how much soul-selling is involved in even thinking of getting it to market.
That is to say, a course less about coding and more about coinage. (A sad reality, but a true one.)
I spent 10 minutes writing about this then Mozilla quit on me, so I'll keep it short out of frustration.
Teach them to use cross-platform toolkits (SDL, Crystalspace) and APIs (OpenGL, OpenAL) whenever possible. Companies don't put out Mac games because they're too lazy, or they don't want to spend money having someone else port it (and in-house developers are too busy fixing known release bugs). If you can tell your employer 'the game is ready to go gold, and we can ship Windows, Mac, and Linux versions immediately', they'll know that, even if they don't release right away, it's easy to do sooner rather than later, and if they DO release them at the same time, people will love you. I refuse to buy The Sims because they don't care about Mac users (we're two expansion packs behind so far, and I don't know if we'll ever get them), but if a company shows that they actually care about Mac users, rather than 'hey, these apple things, can they play games too?', I'll look more favourably upon them. Maybe I won't get this game, but I'll consider it, or the next one, more seriously.
Teach them to code well, not fast. Putting out a good game eventually is better than putting out a crappy game now. That being said, don't be Daikatana (putting out a crappy game eventually isn't the way to go). Teach them that bugs at release and patches a few months down the road are going to translate into people buying it from the bargain bin in six months, after fifteen patches, and that bad code, even if they're told 'write bad code, but make it fast', may be the boss' fault at release, but it's the programmers' fault when the company has to cut jobs. If you need another month to test, tell them that the game is unstable and corrupts important files (well, that's getting borderline). Besides, if nothing else, 'finishing' a game, having it lambasted with bad reviews, and then having to spend the next year patching bugs under pressure is not as fulfilling as being done when you're done.
Other points about games that are worth mentioning. If there's any way to make a game expandable and provide something different even over time (Half-Life mods, Escape Velocity plugins, Warcraft Maps, UT online play, Fallout open-ended play, Everquest expansions), it will improve your bottom line substantially. Half-Life and assorted add-ons are still selling for $30-$50 (depending on the store and the valuation of your country's particular dollar on the currency exchanges). This is a design decision, of course, but if they've got input, they should suggest it. Multiplayer is a quick-and-dirty solution, but look into other options. Even today, I'm scouring the city for Fallout 2, because it's such a fun game, and my friends still LAN party with UT. Not only will it increase sales at launch-time, money will keep trickling in long after the final patch goes live and the code is archived to a CVS server in the basement.
This is just the ideas I can think of right now. I'm sure other people can add more, but this is what I like.
--Dan
One of the most informative sites I have been to that teaches like you did is https://www.gameinstitute.com/gi/. You could perhaps show your college this and get some ideas from it.
Hope this helps.
-Vic
See, when I think of game design, I think of two seperate things:
1) Graphics and concept design
This is totally out of my realm. I imagine there are all sorts of software packages to learn and various gaming "styles" out there to learn.
2) Gaming engine design
This is what I would consider the actual programming. For this, I have to imagine that the biggest requirement is a strong physics background. Granted, there are many APIs out there for game design, but to my knowledge, they don't offer much world physics which is the key in gaming design.
I would recommend developing a physics for game designers lecture (I actually saw a book on this at B&N recently). Another area would be artifical intellegence. The only thing I really worked with here was QuakeC but that was a long time ago. I would imagine that a good lecture here on the limitations of gaming AI would be of great use.
int func(int a);
func((b += 3, b));