Does anyone know where I can find a real killer AI for the first Armada? 1. Please, no "Google it" suggestions, I've tried that. 2. As far as I know, there are only two AIs on this site, both of which I have tried, so unless there are more than two, please don't suggest any of these.
Yes. I'm sure you remember this forum: http://forums.filefront.com/armada-i-ii-discussion/349589-previous-problem-solved-now-i-have-another-problem.html But nobody answered my last question, so my AI just doesn't do what I want it to do yet. If you check out my last question ("I tried both ways, but it still doesn't work, so I've provided my work so that someone can look at it and see if they know what's wrong"), I'm stuck as to how to proceed. Since nobody is answering/nobody knows what's wrong, I'm hoping that there's another great AI out there.
To the best of my knowledge there is no advanced AI for stock A1 or A2 publicly available at this time.
Try lowering building time & costs of AI. Otherwise it's easy to beat 7 AIs in a team.
If this is about the 50 ship thing I have learned a little more while testing some thing and give you a couple more options to try. The first thing is that the AI needs the time to make the ships. Lower the AI build handy cap will help their as it can make ships faster. amother thing is to increase the amount of time devoted to a rest cycle as rhis if I'm understanding the AIPs right is the time between AI attacks.
Sounds like that's what it is! I'll try it, but if I'm still having issues, I'll let you guys know. Also: What are the variable names (such as int escort_force) for the AI build handy cap and the rest cycle?
Also, what variable is used to reduce/increase costs for AI buildings/ships?
I belive this is the first answer // RELAXATION OF THREATS // Iteration for relaxing grid borders int relaxation_cycles = 1; // Relaxation bleedover double relaxation_coefficient = 0.5; As for the second I'm not sure in the files but if you have the 1.1 patch I'm preety sure that can be done from the IA setup screen.
It didn't work. Thanks for the help anyways!