On Sun, 24 Jun 2012 10:59:45 AM Michael Vehrs wrote: >>[Danger paths] > I image the danger value to be so important to the AI that it might be > added to the PlayerExploredTile itself.
That would work, however given that most of the threats are mobile, caching across turns is not going to help much. I was expecting to just build/destroy the map for each AI player every turn. > True. Do you know any generic search libraries that might be helpful? No. However by now I have mangled every line of our A* implementation at least once, so I am comfortable with it and confident in its correctness, so I am not in a rush to rework the underlying algorithmic part of it. Getting back to the Great River problem, what I think should be done there is for the AI missions to remember path finding failures across turns, and degrade gracefully into a plan B. The ones I have reworked recently are very linear. They could be easily re-coded as explicit state machines, without much loss of code clarity and probably some gain in debugging ability[1]. Mission.java could then include support for recording multiple tries to change a state, with something like: void changeState(changePredicate, newState, failureLimit, failState); But first, enhance map searching, and fix TransportMission. Cheers, Mike Pope [1] The AI missions are now displayed on AI units in debug mode. We could probably squeeze in a string for the state.
signature.asc
Description: This is a digitally signed message part.
------------------------------------------------------------------------------ Live Security Virtual Conference Exclusive live event will cover all the ways today's security and threat landscape has changed and how IT managers can respond. Discussions will include endpoint security, mobile security and the latest in malware threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________ Freecol-developers mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/freecol-developers
