In general, loving the game, but I think the AI can do better
I've been playing Intrigue + all DLCs and I'm having a lot of fun with it.
There's a lot to like about this game and I now prefer it over Galciv2, and I loved that game! I can tell that a lot of thought has gone into every mechanic currently in the game. I can try and read the devs mind (I'm not just a gamer, I'm also a programmer) on why things were done -- I think I totally get it and I it all works together great for the most part. Most of the time I'm very impressed. Great job. And I totally get that perfect balancing can be tricky, to say the least.
How I Play the Game
I pretty much play exclusively as the Terrans, on a Large and Loose map, with all options on their default values, against whatever number of AI civs the game recommends that map for (total of 8 players, including myself, for Large). The AIs are all set to Normal intelligence (no cheating). The AIs are always Drengin, Arcean, Torian, Yor, Altarian, Thalan, and the Krynn.
I don't use a custom race because I know there's no way to balance that as long as the default races have to play by the same point buy system rules.
I never save scum. I'd rather concede defeat and start over than save scum. The only time I'll reload an auto save is if the game crashes (happened exactly once), or if I misclicked and sent a critical ship entirely in the wrong direction (happened exactly twice -- freighters and colony ships use too similar icons).
If I play optimally (not skipping any turns), I basically win every game (in my mind) by somewhere between turns 20 and 30 (when I finish my side of the initial colony rush), and I'm not complaining about that fact because there's still so much more to do to actually win... and who knows -- maybe I'll be surprised once or twice along the way? However, I think the AI can do better better than that without cheating, and hopefully this feedback can help improve them because I'm not using a strategy that requires any long decision process.
I try to win the game as quickly as possible -- I don't drag out the game, and I don't roleplay my civ in a sandbox. I go for a Diplomatic Victory through killing everyone that would take too long to ally with. I may culture flip some specific planets early game, but generally don't spend too much time thinking about Influence.
Early game I get nothing but Administrators... maybe for the first 40 turns or so, sometimes more. Workers are never worth it because they only affect social production which isn't that big of a deal, and the researchers (while being good) are not as good as more colonies, constructors, and/or exploration ships. Later on I usually get one or two celebrities, then I usually get nothing but leaders until I can run near 100% tax rate with 100% approval on all colonies. After that, it's whatever I feel like because it doesn't really matter -- but never workers.
I always go pragmatic for the constructors and shipyard decay bonus, then switch the benevolent mostly -- but I recognize that malevolent is very strong, and it's a valid choice on par with benevolent and better than pragmatic. It's just that pragmatic is very front loaded with the later offerings being a little too weak, and those two things are very useful for my playstyle (I don't worry about the AI attacking me). I guess might be considered the closest I come to roleplaying in my game, but not really.
I'd be happy to make my saved games available for download upon request.
Initial Thoughts on the Economy
Tourism dominates: Relative to other ways to gain credits passively (trade routes and each planet's economy through other buildings), tourism gives too much benefit for how little effort you need to put into it. In general, I'm fine with Tourism being powerful, or even the most powerful because something has to be. The main problem I see with it is that everything else isn't really viable at all, so there's really no choice to make here. I like choices, and the harder the better.
I don't see a good reason to build anything other than +tourism buildings, with some +influence (if I can fit it, but they're not necessary), and maybe some +approval buildings (if I can fit it -- there are other good ways to improve approval without using buildings). THE tech to get to even begin to get my economy out of the red (which is a priority, )is the one that gives Port of Call (and before that, the galactic wonder that also gives +tourism). If I can't fit the +influence or +approval buildings on a planet, it's really no big deal -- as long as I use that one lonely slot for Port of Call I'll be fine.
I never build a Central Bank. The +10 would only be useful early game (but not really since it takes up a tile space for something more useful), but it's optimal to run at as close to 0% tax for as long as possible so it doesn't matter and once I can build a port of call, it's no longer worth the slot even with a high tax rate.
Two obvious problems with Tourism:
1) One slot does it all -- I don't even need 3 connecting planetary tiles or any tile bonuses to make one tile with Port of Call work just fine. With pretty much every other building in the game, I'd want at least 3 minimum (in a triangle shape) or at least a nice tile bonus to normally even consider building something.
2) I don't even need to connect my influence bubbles for this to work just fine, it just makes it work even better (I admit to not actually testing if this is a real effect, but I imagine it is).
Running the Economy before having any Income
Tourism is so powerful that it's optimal to just run the economy at close to 0% tax rate before you can get it rolling for the benefits that high approval brings -- most notable being able to counteract the negative approval from having more than 4 planets. Before tourism there isn't that much of an economic effect difference between having a 0% tax rate or 25% or higher. This is part of the tall vs wide problem, where right now the game favors going wide (many colonies) by a large margin -- which would be fine (other than there being only one viable choice) if the AI could keep up with the player during the colony rush.
Floating the economy before tourism kicks in involves using trade (diplomacy, which Terrans are good at with +diplomacy), and Anomalies (which Terrans are also good at with +1 movement to ships). I totally understand that some races are going to be worse at this but better at other things and Terrans maybe even excel at this -- that's fine. However, they all could do better.
I typically build around 3 explorers at some point before turn 30, but they're not a focus for scouting because... I find the bazaar and buy the mercenary scout as quickly as possible every game -- I believe it's overpowered and should cost a resource, like 1 prometheon or something.
How the AI Could do Better
1) Use the Bazaar: None of the AIs seem to try to leverage the mercenary ships from the Bazaar for an early game advantage -- namely the huge sensor range scout and/or the very fast exploration because neither require any resources and they're both very cheap in credits. The AIs should prioritize these things and be able to use them effectively because they give a HUGE advantage -- more than rush buying colony ships or anything else. Right now they are always available to me when I find the Bazaar.
2) Trading (diplomacy): Lower the cap on the amount of credits the AI is willing to hand over in any one trade. Early game the cap is 1000 credits per trade, which I think doesn't work well for two reasons:
_a) It's just too high because I would never trade 1000 credits for anything early game -- I would maybe be willing to trade up to 216 credits (but that would be rare) to sweeten a deal, but that's it. Maybe Terran diplomacy bonuses has something to do with this limit, and that's great if it does, but even if it does, it's still too high.
_b) Furthermore (and this is more important), I can keep doing more trades with them one after another, which would circumvent that limit no matter what it is -- instead that cap shouldn't be per trade, but instead per time period -- that would ensure I don't completely strip them of their money. This problem is magnified by...
_c) It's more lucrative to do multiple individual trades than one single trade. This "problem" existed in Galciv2, and it's still here. It's mostly a quality of life issue (it's more tedious to do multiple trades for each item), but it becomes a powerful strategy when using trades to drain AI credits and the trade cap for credits is reset for each trade. A tedious strategy shouldn't be more rewarding, especially when it's counter-intuitive and seems to abuse the AI.
The easiest way to fix this would be to allow only one trade per civ per time period (15 turns or whatever). It seems like this concept was intended, but it can be circumvented with successive trades without leaving the diplomacy screen (use the arrows up top to switch between civs). But I'd prefer it to work the way it does, except the credit cap being per time period for quality of life.
On another note, the trade offers that the AI asks for are so bad for me, that I never go for them and at this point I just stopped reading them. Particularly because they want everything I have, including my credits. I don't know exactly where I'm going with this point, except that it's something to be looked at because I'm not engaging with it so it seems like a missed opportunity for... something. Maybe tone them down, but still make them somewhat bad trades on purpose (the AI trying to drain ME of my credits), but give a serious enough diplomatic hit if we refuse (and display that fact to the player beforehand). Maybe there already is a diplomatic hit, but I can't tell.
3) Planetary quality should affect colonization choices: The AI should take note of, but skip over poor planets (at least initially). For example, they should never colonize the crappy planet next to their homeworld, unless:
_a) They ran out planets to colonize and have an extra colonizing ship already built; or
_b) They hit the population cap on their homeworld and there's nothing more important to built (I would build freighters if I have free trade slots or military ships if I'm not in the #1 slot militarily before considering building a colonizer for it)
4) Colonizer ship population: They should fill colonizers with the the minimum amount of population allowed, and be willing to run their homeworld down to 1 population in order to make more colonizers (and have a backup strategy for what to do when that happens and there are still more planets to colonize). Two should not be the default, it should be one (the minimum).
My backup plans for when my homeworld pop is running too low is to either: a) use population from other worlds I've colonized, usually by building another shipyard for them and have them create colonizers (possibly rush built), or _b) use the benevolent free colonizer both increase my homeworld population back to 5 AND provide a free colonizer with 1 pop (this is a very powerful strategy). I will run the pop on my homeworld down to 1 without giving it a second thought if that means getting a colonizer out to a good world (class 9 or higher) that much faster.
Colonization simply must continue by whatever means necessary until there is nothing left or I hit 8 or 9 colonized planets, which is double the cap for the first government type. I suspect the AI has trouble going over the soft cap limit successfully. Maybe the approval hit for going over the government cap should be doubled, tripled, or quadrupled... that would slow me down. Or maybe more exponential would be better (1 or 2 over is doable, but after that I'm screwed approval-wise? Right now, having 8/9 colonies when government cap is 4 is managed just fine simply by lowering tax rate to 0% without approval tech, and since I won't have tourism buildings built at this point there's really no significant downside doing so at this early stage. The AI does not seem to use this strategy -- dropping tax rate in favor of approval to go over the limit and float their economy through other means. Maybe it does use it, but doesn't try hard enough to get more planets -- if so it needs to try harder. I colonize planets near other civs last because I don't want to take the diplo hit for close borders -- I'd rather some other civ grab it. But I am able to colonize some pretty sweet worlds right next to their homeworld every game. How did they possibly miss it? Did they run out of administration? I'll happily grab administrators for my first 4+ citizens because the others aren't as important as getting more colony ships/constructors/explorers, and they should realize that.
5) Government choice. It's optimal to switch to new governments as often as possible for the powerful free ships they provide which you get to keep forever, but the AI doesn't seem to use that strategy -- I see them taking crap governments like Media Assimilation or something else which doesn't provide free ships, which is something I would never choose simply because it doesn't provide free ships. So, why is the AI taking them? Maybe the free ships are too powerful, or maybe some of the other governments don't provide enough of a benefit. Note that I've already virtually won the game before the first chance I have to switch government types, so this is relatively minor unless the AI is improved in other areas.
6) The AI doesn't build their worlds well. I'm not talking about making use of tile bonuses because those can be ignored/destroyed and still do well. In fact, maybe ignoring/destroying them is exactly what they should be doing. The optimal strategy is: we want at least 5 connected tiles with a space elevator (rush build that on new colonies if I can, and I always can) in the center (in a double line, snake, half-wheel, or half-star -- whatever you want to call it) of the formation. I will destroy whatever I need to (arable land, resources, whatever) and ignore tile bonuses to make this happen. If there's more tiles, then double down on factories or make a research/factory hybrid planet but only if I can have a flat bonus in the center of the formation (Computer Core, Galactic Mainframe, etc). You can float early game research soley on getting new colonies, even if they have no research buildings. No matter what, I throw down a port of call somewhere (the adjacency bonuses for/to it don't really matter much because no bonus will increase the tourism amount gained).
Maybe part of the problem is that factories are too important for production in general, but I don't know how that can be tweaked without making them useless. Maybe disable our ability to destroy Arable Land and Trade Resources land slots and tweak those in some way (more +wealth)? Just a thought, because that would slow me down and being able to do that it might be too hard of a decision for the AI to make. But that would remove player choice, which would not make me happy because choice is fun in a strategy game.
If the AI just cannot figure out when to destroy tiles to make room for something more important, then I'd recommend crowd-sourcing the solutions for how to build planets through the galciv3 website (or maybe named something else to confuse content filters at workplaces) and using that information to create a database lookup that AI can use to plan planets and/or groupings of tiles. Doing that would also generate some buzz for the game (free advertising is always good). If SD is interested in outsourcing that, we can talk about it.
It's hard for me to comment much on the mid/late game. I do use other strategies after the initial colony rush that I don't see the AI using, but because the AI decisions in the early game ripple forward through later turns so much it's hard to say what (if anything) they're doing wrong that's new. By turn 30, the game is already decided because I have so much of a head start (#1 in military, research, and production). The hole is too deep for them to dig themselves out of by that point without them all banding together and attacking me (and successfully taking worlds) by turn 40.
I hope that helps!
tldr; I love the game, and I identified things that I think could make it better!