What about a community Battle Plan contest?

[KiwiNoob]

It's always easier to spot a problem than to fix it. What about giving the community a Battle Plan challenge? Provide an API and let people write their own mini Battle Plan AI?

It would be a blast because you could pitch the different entries against the others to see which one was best and stream the matches.

Even if you couldn't use the entries verbatim it might procure some interesting approaches, algorithms or ideas.

Worst case - all the entries suck and then you can laugh and remind us of all the times you said it wasn't as easy as it looked

I think I remember you guys doing a similar sort of community thing with path finding a while back.

 

[Keralia]

Sounds fun, no clue how PI would do it though

 

[Johuotar]

I thought this would be about making the prettiest battleplan and sharing pictures of it.

 

[Alex_brunius]

It's always easier to spot a problem than to fix it. What about giving the community a Battle Plan challenge? Provide an API and let people write their own mini Battle Plan AI?

There are already alot of variables for changing battleplan behavior exposed and available for modding in the defines.lue file, so most sensible approach would probably be for Paradox to provide a way to connect a different set of battleplanner AI defines to a specific nation only and then have it fight the base game battleplanner AI in a set scenario and see which side wins.


The problem though is that it would be impossible for such a simple contest to cover even 10% of every situation the battleplanner AI needs to cover. Everything from odd terrain, forts, invasion defense, riverlines, multiple fronts, multiple enemies, reacting to enemy air or naval superiority and other many many other things...

So even if you could find a set of variables that outperform the base game battleplanner AI in a simplified scenario, odds are it would fail miserably elsewhere under more rare / specific circumstances.

 

[KiwiNoob]

That's why I said that it could still procure some useful approaches, algorithms or ideas even if it's not 100% of the solution.