Weekly update

I really hoped that I will do more during this week and especially weekend but I didn't. In fact I just did some little refactoring (related to projectiles), created specialised movement module for projectiles and added library references.

That's it. I hope that the next week will be better :)

Weekly update

Not on Sunday evening but on Monday morning.

Ok, rating stuff is almost complete. It just lacks some interface in AIBasic, but it is not only thing that is missing there.

I started to work on projectiles. They are stripped down objects with better memory management. I could have normal objects (actors, items) being pooled but it really doesn't make much sense as life-time of each is much longer than projectiles (which should be active in most cases for a second or something).

I also added forcing velocity via animation which should give nice looking attack animations. Fun fact: I wrote most of it while waiting for my wife in McD's (she was upstair in language school) and there was small girl, who was very interested in what am I playing and then disappointed seeing black and white screen (on my netbook, after I installed Ubuntu, after few weeks I realised that I didn't set up vim for vala ;) ) and then she started to be interested again what is so interesting in that black and white stuff and why I type so much.

Weekly update

This week I did some more work on ratings and added AI Collective. A way to group teams, factions, etc. And to rate other characters (or items or anything else).

During next week I plan to finish working on rating stuff and do some work on code or actions and projectiles.

Weekly update

I had even less time than the week before, but still managed to add two important things. Tagging and rating. Idea behind tags is very simple and doesn't require much explanation. They can be used to describe any object (any library object at the moment). Each tag is accompanied with a number - relevance. This means that for some objects it might be more important that they are (for example) evil than for the others. If AI sees something that is evil, scary and a ghost, and is scared by it (because it killed its ally), it will learn to be affraid of something that is evil, scary and a ghost. So when it sees something similar, it will run away instead of approaching. How such reasoning can be done? With ratings component. It is used to interpret tags and to rate individual characters. During game it may modify and learn what is scary, what is good and what is evil.

It may sound as an overkill for such a game as Dragontorc, but as with other elements, I'd rather create something more extensible which I could use in future for other projects and which other people could use to modify and extend Dragontorc or to create something completely different. And it still should help working on Dragontorc's content.