AI issues I've noticed playing on Extreme difficulty.
1. The AI's are usually broke. What are they spending their BC's on? They need to better prioritize what they spend on so that they have money available to have trade and research treaties with other races. Without ongoing trade it's hard to reach that level of partnership where you form an alliance with another species. Are they buying out ships? Are they buying out production? Are they building too many spies? I don't know I can't see but whatever it is they are doing is not working very well for them. Spending priorities need to be fixed so that they can have more ongoing treaties.
2. There should be different levels of trade treaties requiring different amounts of BC's to initiate. As it is out of 7 discovered races only 1 can have an ongoing trade treaty with me. They are usually sitting below 300 BC's. For an AI that is supposed to have bonuses over me something is not right. What are they spending so much money on and why are they always broke? That could be a large part of the problem. If they had lots of extra money they could push for more aggressive strategies. If they spent less they could initiate more trade and research which would allow them more money to spend on whatever it is they are already spending on. There needs to be something in place telling them that they shouldn't spend so much that they close out their opportunity for trade. Some limit built in. Besides that the developers should better guide their spending choices and habits to where they are increasing their income in the end and not diminishing it as the game goes on. You should be making more money not less as you expand.
3. A clue to the spending issue. If you set your colony to autobuild you see the AI make some senseless decisions. For example I have chosen unwaivering as one of my species traits. If I allow the AI to autobuild it wants to build a government building to increase morale. This is pointless and a waste of upkeep credits in the future. If the AI is making these kinds of decisions then perhaps that is why they are always broke. Not every building should be built on every colony if you want to have a healthy income. If you go and build every single building on every colony you will go broke. Some colonies should be production colonies with no research buildings. Poor planets should be set as research colonies and built accordingly. Make the build times reachable with production buildings and then set to build research buildings. For production colonies go all for production and mass population but no research. This would wiser and more money efficient. If the AI followed these basic rules for certain planet types then it would have more money and be more of a challenge.
4. Priorities for colonization. The Alkari had just settled a large terran or swamp with poor rating in a system. I had just began scouting this area because I was finally able to get past the unstable warp points. I saw 2 Ultra Rich planets in that same system. I quickly produced colony ships to colonize them. The Alkari had access to these planets long before I did. It was on their side of the unstable warp point. They should have already been colonized. Why pick the poor planet over the barren ultra rich? Yes its easier to farm but with terraforming that medium barren ultra rich will become much more valuable. Even short term it becomes quite productive with just a few structures in place. AI colonization priorities and aggressiveness needs to be tweaked.
I came along and colonized those 2 planets and the Alkari just warned me not to colonize near them but I did anyway. I then set to work building a military post on the warp point blocking access to their planet if we were ever to be hostile. So it was a clear aggressive move on my part which the Alkari should have gotten more hostile over. It would have been easier at that point in the game for them to take those colonies and protect them than it was for me. However I had enough time to move my fleet and build star gates so their window of opportunity was gone.
They should be faster and smarter when given such opportunities. They need to consider the long term and not just the short term. I realize programming such AI is a difficult task. Instead of trying to make them think of all the long term possibilities they need to default behavior to take ultra rich over poor despite the poor planet having a better environment. In the end those ultra rich will be more valuable.
5. Another example of AI lacking the needed aggression to compete. I had just invaded the Trilarians whom I was at war with. It was a move out of boredom because I had not seen combat all game except for pirates. I got to their homeworld system and destroyed the military base protecting the warp point. The Trilarian fleet should probably have stayed on their homeworld system but instead attacked me and brought their entire entourage of transports and colony ships as well (very dumb move). After I decimated their combat ships I then took out all their non combat ships. So now they just handed me victory when they could have just played more defensive and protected their homeworld. A move that would have been victory for them since I could handle their fleet away from their static defenses but not with them. I knew that because I saved the game and then attacked their home world. I realized I would take some losses though I could win with major losses. However if they had their whole fleet which they threw away I wouldn't be able to win. So moving on from that. The Trilarian were now without a fleet and their homeworld was open aside from static defenses. I then saw the Bulrathi coming with a massive fleet looking as though they would finish the job if I wasn't going to. So I ran off with my wounded ships not wanting to fight the Bulrathi when I couldn't even take the planet without transports. I was going to let them have it and just pit the 2 races against each other. However the Bulrathi didn't take the planet. They had a huge massive fleet and they took out the military post leading to the system but didn't continue on to attack the planet itself and take it. Judging by their fleet size and the lack of any defending Trilarian fleet they could have easily done it.
6. The combat AI which I discussed already in a previous thread. The ability for the AI to control the movement without controlling the attack targets. That way they can move the ships and stay out within a certain range while I control who the targets for attack are.
Conclusion: This leads me to conclude that the AI is lacking in several areas. The game has a potential to be great if these AI issues are dealt with. If not it will just be a passing amusement and not a lasting one as MOO 1 and 2 were. Everything else, good graphics, good voice acting, tactical beautiful combat, 1 more turn syndrome, refined and polished interface, simple but complex at the same time is there. It just needs that extra work on the AI to take it to the next level. It has the potential to be the best 4x game if this potential is seized upon.
Edited by Jevokan, 12 April 2017 - 12:54 PM.