Thanks to all for your posts to date. I am quite excited that so many are straightforward, supportive "let's do it" type replies.
Equally other replies questioning, or suggesting alternatives to, my original post are very thought provoking ...
Twill wrote:Or you could get really cutting edge, set up a rules based system for deciding if they should attack (for an AI coding example, see ... <snip> ... Or there are other systems this could be based on other than territory names or randomly. For example, it could be based on who holds the most territories, who has killed the most natives (revenge is sweet), path of least resistance, path of most resistance, closest target (map makers would have to take this into account in their designs), only attack when provoked (i.e. attack only if attacked first), attack if relative player armies in country being attacked is less than 10% or above 75% (easy pickings vs desperation), set by the map maker ("motives"), randomized motive but announced.
Just to reiterate my position my suggestion is not about AI players. It's an interesting topic, but not for this thread. Perhaps AS players? Artificially Stupid? That describes my zombies pretty well.
[Continuing the off topic train of thought - I like the idea that the site provides a programmable AI engine that could, perhaps, be 'programmed' using XML or something. Each of us would be able to write our own AI code and paste it into the engine. There could then be AI tournaments etc with the points going to the programmer ... Like I said - interesting, but off topic
]
Twill wrote:I think alphabetical territories is a mistake because in some maps (because this feature hasn't been designed for) the computer would always go one way. Always.
Yes. They would always behave in an utterly predictable manner. (And this will apply equally to maps designed before zombies and those designed specifically with them in mind.)
But that doesn't mean you could predict the behaviour with any certainty until it was too late. I'm probably being a bit over the top here, but I think a sub-chaotic behaviour would result. There would be a kind of "butterfly effect" possible in that players could radically change the route of the zombies by attacking tactically.
Twill wrote:For example, in midkemia, Qwan starts with 6 neutrals, it would always attack earennial, which would suck to start there, kind of like alcatraz in LA.
I agree that this option would not suit some maps. And certainly some of the newer maps with pre-defined large numbers of neutrals may not be appropriate.
Remember that neutral zombies is intended as an option not a default setting. It should be switchable just like Fog of War. Some maps will suit NZ's (Neutral Zombies) and others won't - we all have our preferences for certain game options on certain maps and this intended to be another option.
Twill wrote:it's just too simple of a system and will get boring after a while, there needs to be uniqueness if not randomness. if you have a classic map, and neutrals end up in kamchatka, they will always attack Alaska, Alberta, Northwest tarritories, greenland, iceland, great britain, n. europe, s. europe, egypt, congo etc etc etc. no matter what the opposition looks like - it's like a one player suiciding in a 3 person stand off - someone always gains and someone always loses, and nobody likes being on the losing side and nobody can control it.
And this is where the sub-chaotic behaviour kicks in I think ...
Firstly I agree that if the set up is on Classic as you describe and this position is at the start of the game. If you ignore all other factors, the NZ's will progress along the route you describe.
But those other factors are (i) the players. And we're not going to be ignored!! And (ii) the dice. I think these will affect the zombies route too ...
Let's assume that you, Twill, are a player whom it would suit perfectly for the NZ's to pursue the route described. Fine. You'll do nothing to stop them and will try to make the best of your advantage. However I really don't want them to turn up in Europe and so I'll attack them ... [At this point in my post I began to describe a hypothetical game to illustrate potentially alternate outcomes, but it gets too long to make good reading - see my closing comment.]
Twill wrote:I like the idea, it needs a lot of work before we consider implementing it though
I
think it needs less work than you think.
But you've got me thinking enough that I'm going to go and do some play testing on Classic map (since I have the board game) and report back to the thread ...
Cicero