Standard Techniques

Started by
1 comment, last by jbadams 19 years ago
I just want to get a thread going where people can post and discuss methods of doing things (anything) in games. What are some techniques that you believe are "as good as it gets" or are so nearly so that there's very little to gain by trying to figure out a better or different way to do them. You can make the context as specialized or as general as you want, and if you can provide a link to a good tutorial or demo, even better. For example, to partition a three dimensional space to aid culling or event notification, an oct-tree is fairly standard. Certain flavors of game loops are also standard fare. Finally, do you think there would be any desire in the community for a standardized library of game techniques that requires your game objects to support certain interfaces and in return provides certain standard functionality? (Think middle-ware, but on a more general scale). Maybe this already exists, but I'm not aware of it.
Advertisement
I don't think it's practical because everybody has a slightly different approach to all these situations. I can't really think of any particular technique that I've applied universally in every game that it might work in, until you get down to basic programming practices (eg. delegation, encapsulation).

Also, the standard interfaces might work ok in Java or Python where they can be easily bolted on, but not so well in C++.
This is too general, and as Kylotan pointed out, few things are completely universal. Sorry, but I don't think people will be willing to ramble off general techniques when it's not in response to some specific task.

Quote:Original post by Penance
Finally, do you think there would be any desire in the community for a standardized library of game techniques that requires your game objects to support certain interfaces and in return provides certain standard functionality? (Think middle-ware, but on a more general scale). Maybe this already exists, but I'm not aware of it.


Sounds like using an engine, doesn't it? Generalise an engine too much and it won't be good enough at specific tasks. Ergo, we don't generalise to that degree, and therefore have a range of different engines suitable for various different designs.

- Jason Astle-Adams

This topic is closed to new replies.

Advertisement