Sury for sure.
Second place, I'll go for Brennus.
Second place, I'll go for Brennus.
|
March Madness: Civ4 AI Edition
|
 (March 16th, 2014, 11:53)novice Wrote: Isn't the map much more important than any other factor for deciding the AI's performance? Have you considered running this on mirrorland maps? Toruslands could be viable choice too, but I think running map script that at least tries to maintain semblance of balance is good idea. I’m betting on Ramesses to take the group!
Oh that's right, you were using an observer civ? You need somewhere to hide it, then.
I initially assumed you were planning on running an all AI game and watch the game from the debug console or something. I hear that's possible, never tried it myself.
I have to run.
Will be interesting to compare results with these settings (NTT, no vassals and agg ai) to the civfanatics AI tournament. I think these games will be quite long due to slower tech and also the fact that military victories are harder to get without vassal system. Maybe culture victories will be more common due to the slower tech (harder to get space)? The culture candidate would have to be either strong enough to defend or just lucky though. Agg ai also makes world a bit harder place for a cultural builder. Teching capacity is obviosly more important with NTT. Maybe someone like Ragnar could do well with these settings, fin trait providing economic strenght and high unit prob+agg military strenght. Then again imp AIs seem to do usually pretty well.
With these leaders I expect the first game to have a lot of fireworks  . Sury is one of the strongest ais due to good expansion, a strong candidate for win. Ramses has good uu but too low unit prob to use it effectively, I doubt that he's going to do well with this kind of company. Toku definitely benefits from NTT but still I doubt his ability to win since he has no economic traits. Also being isolationist means usually more enemies. I think one of Peters streghts is his lack of loyalty, he can plot at pleased or firendly not sure which one. Anyway he won't let his feelings get in the way of conquest. Also his unitprob is good. I don't have any strong opinion on Napoleon or Brennus, high unitprob is of course good. Usually Brennus doesn't tech so well though. So my bet is for Sury, second place for Peter. Seeing the map and positioning would likely change this.I don't think the uniques will make much difference here, certainly less than with human players. Rome could be an expection since praets have a long period of usefullness. Maybe ubs could be more important since the ai tends to build so much infra.
Ramesses will likely be dogpiled due to the warmonger bonus (warmongers like warmongers). It's likely that the more agressive AIs will actually love one another for a long time. Not entirely sure, since I haven't read SP stuff in a long time, but I think it works something along those lines.
Alright, Game One is in the books. Time to see how everyone's predictions worked out. Link to the game writeup is here, results are spoilered below if you'd rather cut right to the chase. http://www.garath.net/Sullla/Civ4/survivor1.html
These maps are random, by the way. I actually didn't want to do a perfectly balanced mirror setup. This is intended more as a fun storytelling experiment than a controlled laboratory test. I don't know how you could do a perfectly fair setup unless the whole system was a series of duel, done over a large number of repetitions, and that's not the goal here. For me at least, the randomness of starting positions and local resources is a good chunk of the fun. 
|
![[Image: 172650_slide.jpg?v=1]](http://i.crackedcdn.com/phpimages/photoshop/6/5/0/172650_slide.jpg?v=1)
