lua-users home
lua-l archive

[Date Prev][Date Next][Thread Prev][Thread Next] [Date Index] [Thread Index]


On 22 Oct 2012, at 10:26, spir wrote:

Just to say I'm not a games developer, my opinion is only based on coding a few agents for optimization simulations - so probably isn't worth that much.

> Would you explain why? (the two points you say)

*) I dislike the approach of returning the decided move from a function simply because, I find in a lot of instances (I code some AI's for use in simple simulations - I'm not a games programmer) you tend to be in the middle of an algorithm, 4/5 functions deep in the call stack at the point you decide what move to make. To code a search algorithm where you have to return and be re-called is quite cumbersome in my opinion.

**)  When using an approach like this, the API can be styled in such a way to lead you to think the AI has control of the program rather than being a repeatedly called procedure (even though it is). I find this is just a far more natural way to think. Also as in (*) you avoid the issue of having to return from a few functions deep in the callstack.

Thanks,
Kev