Areas of Influence

[Wiglaf]

Does the AI at all perceive troop movements as potential threats?

no.

[mkorin]

Originally posted by Kuciwalker


I'm not asking them to program it themselves. I'm asking them to come up with the mathematical solution to the problem, which has thus far eluded us.

Kuciwalker, if you're speaking for Faraxis and they are willing to actually fix this problem. (Armies walking right by each other or fortified positions) I'd take a shot at coming up with a mathematical definition and psuedo code. Post or send me your definition of ZOC and requirements and I bet I or the think thank on Apolyton can come up with something that fits.

Mike

[Kuciwalker]

What? I'm just and random person on the Internet.

[mkorin]

Originally posted by Kuciwalker
What? I'm just and random person on the Internet.

You said 'we' haven't been able to figure it out. I hoped Firaxis might actually monitor these boards and you were it. Either way, I don't believe the subject is impossible, just tricky. and depending upon the requirements should be doable. Most likely by taking a expected risk value of going through the ZOC or around it. Say for example one AI has a rifleman blocking a key access point. The math already exists for % of success for any combat. If say X% of your SOD is acceptable casualty, it would be relatively easy to decide around or through. Depending upon how far around you go, the X% gets larger. You could also have potshots from a fortified position if you go by it. Then if I go by the position, I will lose X hit points or x% of my SOD. You could also have a % chance of 1 army sneaking by or being destroyed. It would all depend on you design document.

Mike

[Kuciwalker]

All the Firaxis people have tags next to their name. By "we" I meant "people."

[Brael]

Originally posted by Kuciwalker
1) The AI will still be pretty terrible at setting up ZOC to its advantage.

2) Otherwise, I think that could work.

It would involve treating a ZOC the same as a terrain improvement essentially so yes, the AI would suck since it barely knows how to build things better than cottages already. I imagine it would get a bit complicated to teach the AI where to create a ZOC effectively.

[realpolitic]

Civ3s ZOC idea made it more realistic. Of course enemy troops would get shot at when they passed by.

[ahenobarb]

Originally posted by Kuciwalker
What? I'm just a random person on the Internet.

Never said with more emphasis or pride!


I'm nobody, just some "random encounter"!

[dsteve3]

It would be nice if the military aspect of Civ was a little more realistic - zones of control do represent protecting supply lines, actions by exploitative units that have standing orders to attack flanks, advantages of armies that have more sophisticated command structures and lines of communication, etc.

The issue is an adjuct of the whole 'unit vs army' dilemma. No empire ever built individual units in isolation of an overall military philosophy. Most times, the leaders had to use what military infrastructure was provided by them by the society. If you were to change the overall structure, the upgrading of units was only a small part of the whole - especially for gunpowder.

Wouldn't it be cool to develope whole armies and military philosophies, rather than micro-manage each dinky unit? Its also alot more realistic. I hate having to hunt down each and every archer unit hidden in my cities during the industrial age. As an empire-ruler, I should be dealing with the whole philosophy and infrastructure; my ministers should be updating the units.

That really should have been developed better for Civ IV. Hopefully we'll see a more dynamic system for Civ V.

Zones of control issues vasilate through history much like the missile vs melee dicotomy. Once one system gains universal acceptance, a new system is developed by the 'up and comers' to over-throw the established order.