The Art of Wu

At the end of my last game, I decided that I was going to try my hand at Always War next, the variant where you declare war on every civ that you meet and never sign peace with them. I played as the Aztecs for their "culture through killing" unique trait, King difficulty, on a Small Pangaea map. I also turned off the city states to see what effect that would have. I intended to write up a report on that game, however there was ultimately little to say. This picture covers just about everything:

I used my culture for the policies in the Honor tree, which very quickly resulted in some obscenely promoted units. I ran over everything in sight, and ultimately it wasn't a very entertaining game. I guess that King difficulty wasn't high enough; the AIs were completely feeble. I only lost a single unit the entire game, a trireme which was slowly worn down and killed over the course of some 30 turns because naval units can't heal outside of your borders in Civ5 without a second-tier promotion. Eventually, even with the really slow production, I found myself with nothing to build in my cities. Can't build more units because of crippling unit support costs. Can't add more buildings for the same reason, and most of them weren't worth it anyway. Can't train settlers for more cities because of the tight happiness cap. I had my core cities endlessly training scouts and disbanding them for the 10 gold refund cost (five times more effective than building Wealth, by the way). Meanwhile, because of the lack of city states and my razing of all the AI cities, the map turned into a giant swath of wilderness, with barbarian camps spawning endlessly. I wanted to settle in that wilderness, but couldn't, and even the one road I did have going across the map was killing me in maintenance. Not that I needed any more units! I had about eight highly promoted super-units eating their way across the map, which the AI couldn't do a darn thing about. It literally did not matter what I built in my cities - the units I already had would eventually win the game.

In a word, it was... boring. This was nothing like the Always War games I played back in the day with Civ3/Civ4, where the map would be full of cities and you had to struggle to create enough production to slowly push back the enemy armies. I realized that I was going about this the wrong way; all of that wilderness and subsequent barbarian activity was driving me crazy. Inspired by some posts on CivFanatics, I abandoned this clearly-won game and decided to take a completely new approach to expansion. (That's now three games of Civ5 so far, and none of them played out to conclusion. Not a good sign, sadly.) This time I would play as China, one of the game's strongest civs, and manage the economic side of the game differently. Time to explore The Art of Wu.

Here we go again on Immortal difficulty, with all of the default game options. Although normally I play the first map that comes up, I had a start that was all jungle with bananas for food, and that looked really rough, so I generated a second start and went with this one. This map had a strong starting position: on a river (a very big deal in Civ5) along with two of the best happiness resources, gems and silver. The two best happiness resources in this game are gold and silver, as they always appear on hills (the best tiles), they have very nice tile yields at 3 shields + 3 gold per turn, and you can further improve them by building a mint for additional cash. I had tons of silver resources here, and a hilly region with rivers to expand into. Second city would go south down the river, the third city west on the coast by the fish, and I would be in business. Even better, there was a Maritime city state nearby to befriend, excellent!

Now the downsides. There were no other happiness resources anywhere in this region aside from those silvers and gems. Going to be difficult to expand past the first couple of cities. Secondly, this was the only city state present. Thank goodness it happened to be Maritime, but there would be no second Maritime state or Cultured state to exploit. Finally, notice the mountain range off to the east. I'm trapped here until I can discover additional technology to embark my units! I would eventually get a scout past that mountain barrier, and this continent extended quite a ways further east, but for the moment I was alone.

Beijing built scout, worker, then settler in what appears to be my default starting build order. I'm still not at all convinced that going for Stonehenge is worth it - I'd rather have faster expansion than the extra policies. I was able to keep the barbs down with just my two units, until eventually I built an archer for help patrolling the region. I deliberately didn't patrol the southern area next to Helsinki, knowing that a barb camp would spawn down there, and then I could gain free influence from the city state for clearing it. Sure enough:

I cleared the camp and gifted back the worker to Helsinki, which gave me 30 free influence. Then you can see that I've just topped 500 gold on the bar at the top of the screen, so that meant another 75 influence for instant Allied status, which would extend for the remainder of the game. Keeping one city state as an ally is a trivial task in this game... I used what I had available here.

You might also notice that I still haven't picked a social policy yet. I was saving them for a later era, as I wanted to experiment with skipping the early trees and concentrate on Patronage and Order later on. However, the cost for each policy still keeps increasing as you add more cities, and thus I probably wasted at least one policy by saving them up for so long. Don't get me wrong, it was nice to enter the Medieval era and instantly grab two policies in Patronage, but my gut feeling in retrospect is that this was ultimately wasteful. In my next game, I'm going to go back and experiment more with the Liberty tree, which I still believe is the strongest of the early three when not planning to engage in early warfare.

Eventually my scout was able to slip past the blockading mountains and found some civilizations off to the east: Germany, France, and England. Meanwhile, I was about to plant my fourth city, and the first of a new type:

This is Nanjing. The land here looks pretty terrible, given the lack of food. However, Nanjing practically forms the model case for the new type of city that I planned to employ in this game: the filler city. Let's look at this from the standpoint of happiness, the limiting factor on growth in Civ5. There's a finite amount of happiness out there on each map, probably something like ~50 happiness to be derived from natural wonders and luxuries. Everything beyond that you have to provide yourself, through buildings and social policies and whatnot. Each point of population costs one unhappiness, and each city costs two points of unhappiness - with some exceptions (India, some social policies/wonders, etc.) Now Civ5 has some other fundamental aspects of its design:

* Cities grow incredibly slowly as they increase in size. It is virtually impossible to get much beyond size 13 or so until the Industrial Age unlocks hospitals/medical labs, which of course cost more money and more production.
* More advanced buildings do exactly the same thing as the basic buildings, only they cost more gold each turn to support and take much longer to build.
* The center tile of each city can produce ridiculous yields, by leveraging magical maritime food and the Order social policies. It's not uncommon to be getting 10 food + 8 production on the center tile!

The only real cost to founding more cities in Civ5 is the increased expense of social policies. In every other way, more cities are always better: they provide more production, more gold, more research, and so on. Thus, if you can manage the happiness costs, you pretty much always want more cities. And it is in fact possible to manage the happiness cost: the secret is to limit your own population. Have a few "normal cities", then cap all of your other cities and do not let them grow past a certain size. I've found that size to be 4, where the total unhappiness cost of each city will be 6 overall (4 + 2, and notice this is exactly the same for India, only reversed). With Liberty's Meritocracy (+1 happy per city connected to capital) and Order's Planned Economy (-50% unhappy per city) or the Forbidden Palace wonder, each city costs only its own population in unhappiness. Cap its growth at size 4, build a colosseum, and every city is happy-neutral. Now you can spam them endlessly across the map, and every city simply adds more production, more research, more gold...

Nanjing would be a perfect example of one of these filler cities. It will only get four buildings ever: monument (cheap and for minimal border expansion), library (for specialist use if desired), market (free gold at no expense), and colosseum (pay for its own happiness). Nanjing is an even better city because it can also add a mint for that silver resource, and even more income. With one Maritime city state ally, the center tile makes 4 food/turn. I need another 4 food/turn to support a size 4 city (8 total food) and I can get that from farming the grassland river tile post-Civil Service. With its food taken care of, Nanjing is free to work three trading post hills, probably the best overall tile in the game right now at 0/2/2 yield. Or, if I want science, I can turn two of its citizens into Scientist specialists, and instantly produce 12 beakers/turn from this little mudball of a city. Some civs are particularly well suited for this: China (due to the Paper Maker replacement for the library), Egypt (replace Monument with Burial Tomb), France (free culture), and Greece/Siam (city state benefits) come to mind. Nanjing will nearly support itself on happiness, turn a profit on gold, kick in extra research, and give me pretty good production too. This is my response to the empty map of my past game, to spam these filler cities all over the map. Infinite City Sprawl (ICS) is back again, unfortunately. Ironically, the designers of Civ5 set out to make a game which would favor small empires, and have ended up with a game where bigger = better in nearly every way possible! More on this later, you'll see what I mean in time.

Getting back to the narrative, I met the AIs over on the eastern half of the continent, which proved to be significantly larger than my sealed off corner in the west. I had excess silver resources on hand, so of course I started selling them away for free cash. As I wrote in the previous Immortal game report, you might as well sell your luxuries at all times for whatever gold you can get for them. The AI completely ignores happiness in this game - even on King, they always seem to have more happiness than they need. Sell, sell, sell and don't worry about the consequences. You can also sell your strategic resources for cash, and even sell one-sided Open Borders for more money. It's very easy to raise gold in Civ5, which is part of why the city states feel so unsatisfying to me. Sell your stuff, get money, and then you instantly have free city state allies. There's... not a lot of depth to this system.

This was the terrain on either side of that blocking mountain range. I quickly realized that controlling this area would be key, if I didn't want to be stuck in my little starting area for the rest of the game. If I could plant a city right by that barbarian camp, it would be pretty well protected from attack and would give me a secure base of operations over there. Plus, there was gold and cotton over on the other side of the water - I had to have them! I prepared a settler to sail across this little sea, and also some extra units to defend my claim over there.

I was somewhat surprised to find that these AIs were already at war, with Germany and France teaming up to attack England. Poor Liz was already on her way out of this game:

I sent her a silver resource for her remaining gold, and England was very shortly thereafter finished. Goodbye Liz, thanks for playing. Of course, this wasn't good news for me, since I wanted a bunch of evenly-balanced AI empires to exploit, not a runaway civ. It was very clear that Napoleon was already turning into the runaway of this game:

We're currently on Turn 90, still not even to 500 BC yet. Napoleon has founded four of his own cities, plus he's absorbed most of the English empire. He's also conquered the city state of Copenhangen (no more Maritime food there!), and has thus ended up with a grand total of nine cities thus far. Obviously I can't match that myself... Remember, the AI ignores happiness for all practical intents and purposes in this game. Although Civ5 is supposed to favor small empires, it's the reverse that's true when it comes to the AI, as size = power without fail. Remember in Civ4, how a small AI empire could be weak militarily but strong in terms of technology? Not so in this game. The biggest AI empires will always be the most technologically advanced, and have the largest military, and have the most gold per turn to spend. As a result, a handful of warring AI personalities are by far the most dangerous; I keep seeing the same AIs come out on top over and over again. Napoleon (France), Alex (Greece), Nobunaga (Japan), and Hiawatha (Iroquois) seem to be the main culprits. The warlike AIs will take out their neighbors, growing bigger and bigger in the process, and then they will keep that snowball rolling until they control their whole continent. The more aggressive the AI, the better they perform. I do not think this is what the designers were going for!

So I have a runaway France to deal with in this game, and that's going to change some of my planning going forward. It's not a question of "if" I'll end up fighting with Napoleon, it's a question of "when". Have to make sure I'm in a better position to fight than I was in that last Immortal game. In the meantime, I got my settler past the little sea and staked my claim to a toehold on the eastern part of the continent:

This seemed like a good spot for a city. I purchased two tiles immediately for Xian, the hill and the cotton tiles to the east, as I wanted to lock down that border with France. What I wasn't expecting was that the AI would come calling me to complain about founding a city near their borders. Both Napoleon and Bismarck dialed me up literally the next turn and protested about this city! Seriously, huh? Xian was four tiles away from Troyes, and a full six tiles away from Berlin! That's a normal distance for a city, not a particularly aggressive plant. Does the AI expect you never to settle anywhere even remotely close to them in this game? When they will march a settler halfway across the map to settle next to YOUR cities?!

Within the next few turns, Bismarck and Napoleon both canceled their "Pacts of Cooperation" with me. Oh God. I can see where this is going. Not this again...

In the span of five turns, Bismarck went from my good friend to declaring war on me. Because I settled a city six tiles away from him. Yeah. Uh huh. Sorry folks, but this is really absurd. I've played quite a few games of Civ5 by now, and it seems to be completely impossible to become a friend or ally with any of the AI civs. Eventually they are going to declare war on you, because virtually everything seems to irritate them, and those maluses override any benefits from trading or Open Borders or whatever. That's not to say that diplomacy is completely meaningless (as you'll see in one second), but it does mean that diplomacy is extremely limited. The only real uses for diplomacy are extracting the maximum cash concessions from trades, and timing your wars to avoid being dogpiled. You can never be friends with anyone in this game. They are all your enemies. Get what you can from the AI with resource trades, and be prepared to fight everyone at all times.

This isn't good, but I've caught a huge break here nonetheless. It was Bismarck that declared war on me, not Napoleon. Because France was the runaway AI civ, it was trivial for me to get him to jump into the war, against the much weaker German army. I paid 150g and my surplus four horses (completely useless to France, who had dozens of horses already!) and he entered the war on my side. Now much of the German military will be locked up fighting against France, rather than coming after me.

Here's the situation at Xian. I have a warrior and two archers, plus the city's own natural defenses. That wasn't much, so hopefully old Boney will be able to keep Bismarck occupied for the time being. I was hoping to reach Machinery tech for China's unique unit Cho-ko-nus before fighting, apparently not that fortunate. The good news was that Xian happened to be founded on iron, something I didn't even know when I first grabbed the location! No iron anywhere on my starting subcontinent. None. Now I could upgrade my warrior to a swordsman, and start building a catapult back at home. With those extra units on hand, I was safe for the time being.

Here's where I stood 100 turns into the game:

This situation was pretty comparable to what I've found in my other games. Five cities usually appears to be the limit after 100 turns, unless the player has been very lucky in acquiring extra happiness resources. I had three here, my starting gems + silver along with whales from Helsinki. I could not have supported much more population than this. Research and especially gold were both in excellent shape for this point in time, thanks in part to China's awesome unique building. The Paper Marker is a library which also adds four free gold per turn. (Although it still costs 1gpt in maintenance; why they didn't just make it free and give +3 gold is beyond me!) That unique building along with China's excellent civ ability (stronger Great Generals that appear more often) makes Wu one of the best leaders in the game.

I was pleasantly surprised by my performance in the Demographics as well. I found myself a surprising second in many of the most important categories, and that with only a single city state on my side. Of course, there were two major problems here. One was my poor rating in military, which I was now belatedly starting to fix. I hadn't needed units up until this point in time, which was why I had let that one slide. Secondly and more importantly, notice which civiliation was dominating the Demographics rankings. France was by far the #1 civ, way out in front in every single Demographics category that matters. Not a good sign...

As far as the war goes, I was doing surprisingly well with just my handful of units. The AI remains abysmally stupid when it comes to combat, and thus my little group of two archers + one swordsman had no problem churning through double their number of German units without loss. The AI will literally walk its archers right up next to your melee units to be killed. Granted, the AI has never been good at combat in the Civilization games, however the new tile restrictions certainly highlight that stupidity to a much greater extent. I also discovered that Bismarck had apparently lost some of his iron to Napoleon, as the combat effectiveness of his swords dropped dramatically. 50% combat penalty? Ouch! Seems appropriate though. Better make sure not to lose your extra strategic resources.

I killed about five or six German units, and then tried to advance on Berlin. I was nearly able to capture the city out from under Napoleon's nose, but fell just a bit short. Should have moved forward a couple of turns earlier. Once that city fell, I went ahead and signed peace with Bismarck:

The AI gave me gigantic concessions for peace, as it continues to do right now. I don't think this was appropriate given how I had killed only a handful of units. Whatever though, guess I'd better take it! You can really get some insane rewards from the AI right now by beating on them like a pinata, and waiting for the candy to drop out. It's possible to get cities in the peace deal, and then immediately turn around and SELL those cities to other AIs, for more gold and more resources as rewards. I don't play the game that way, but it's very possible to pull of exploits like that right now. I've had a number of readers ask me questions like, "why don't your reports sound more positive about Civ5?" Well folks, it's due to things like this. I've already played this game: it was called Civ3, and over the course of a year-long patching process, Civ3 removed these diplomatic exploits. Someone playing this game for the first time isn't going to see this as an issue, but I've already been down this road before, and I don't have the same kind of patience. I simply don't understand how the same exploits that were noted and corrected in past games can all be present in Civ5 again. Was Firaxis simply not aware of its own past games (?)

Oh, and another thing on that same note: due to a bug, there is currently no research overflow in Civ5. Any beakers you produce beyond that needed to discover a tech simply get wasted, vanished into the ether. What this means is that you must tediously micromanage your research for every new tech, or else suffer large amounts of waste every few turns. It is Civ3 all over again, and it's driving me crazy while playing the game. Please, please fix this bug ASAP, Firaxis!!!

Anyway, now I have peace with Germany. However, that doesn't mean that my problems have been solved. All I've done thus far is to buy myself a temporary meat shield, by feeding poor Bismarck to Napoleon's giant armies. That will last for perhaps 20 turns or so, and then Germany will be gone and I'll be facing the enormous French empire all alone. And I knew very well this was going to happen - the moment that I signed that military alliance with Napoleon, Bismarck was doomed. Fortunately, those 20 or so turns should be enough to be decisive. Notice that I've now finished researching Machinery tech, and I have been churning out some cheap archers for upgrading purposes. Only 110g to upgrade archers into Cho-ko-nus, a laughably small amount. It costs something like 250 gold just to cash-rush an archer! These units are absolutely amazing, slightly less effective crossbows (strength 10 instead of strength 12) which can fire twice every turn. Along with two swordsman to serve as melee blocking units, and one catapult for some siege, I'm prepared to make my stand here. I must declare war on Napoleon right now, while half his armies are off fighting Bismarck, and make as much progress as I can. This was my one window of opportunity to make a move: Fortune Favors the Bold.

Napoleon's initial response was underwhelming, and I saw few units after starting the war. Look at how much damage my Cho-ko-nu was doing here, with an unpromoted unit attacking a French spearman on rough terrain. The text might claim that Cho-ko-nus are only strength 10, but that's misleading because you'll almost always have a Great General fighting with your units, and the Chinese generals provide a massive 45% combat bonus. One of those badboy Cho-ko-nus with a Great General becomes strength 14.5, the same rating as a normal catapult, only the Chinese unique unit fires twice per turn and doesn't have to set up before firing. This is an insanely good unit, and I chewed right through the nearby French units with them.

All those hills made it very difficult for France to attack me at Xian. I held in the south and pushed forward slowly in the east; upon finding Berlin empty, I went ahead and captured it:

Now I had a relatively easy chokepoint to defend in the north, as there were lots of marshes near Berlin which made movement difficult. Enemy units walking up to the city would be shot down with highly promoted Cho-ko-nus and catapults. Defending to the south was somewhat tougher, although I was dug in along that ridge of hills and had plenty of killing fire on hand. I've already played this game enough to know how the AI operates, as it will attack right away with everything that it has. Wait and kill the AI units, and then counterattack and take its cities once all the units are dead. I did this against Japan in my American game, and I would try to follow the same basic strategy here. A little tougher in this game, given how much of a runaway France was, but still doable. Hopefully.

So much for poor Bismarck. I didn't expect much from him, but I expected more than I got. He lost his five cities in a real hurry. Now it was down to just me and Napoleon, with a lot of blue dots on that minimap.

With the German corpse left to the vultures, I began seeing French units in much larger numbers. They attacked from the same two directions I expected, a flow of units down from the north near Berlin and another steady stream coming up from the south near Marseilles. Initially the units were Ancient age trash that was easily killable, however soon I began seeing more advanced French units. Lots and lots of pikes, plus increasingly longswords instead of swords. There was no question that Napoleon was ahead of me technologically, as his innumerable cities churned out tons of science for him at discounted costs. Fortunately Napoleon didn't send any Mounted units at me, which I don't understand. Horses would have been the best way to reach my Cho-ko-nus over terrain obstacles. Worked out for me. I suffered basically no losses and killed many, many, many French units.

That went on for a while, until I noticed this:

Check out which civ was now *LAST* in the military rankings! Napoleon was thoroughly gassed, and that was my cue to go on the offensive. He no longer had a field army, and would only be throwing random units at me as they were produced in his cities.

When I advanced on Marseilles, I found... a French cannon?! Whoa. That's bad news. See how I'm researching Physics for trebuchets? I am not close to having my own cannons, let's just put it that way! Napoleon cash-rushed this unit in the city, along with the city walls. He had a monstrous treasury (2000+ gold) and a ridiculous income over 200 gold per turn. I would find this pattern repeated many times in other cities, nearly always a cannon waiting on defense inside. That's a good use of the AI's money, even if it is somewhat formulaic and predictable. I had plenty of siege units on hand, my weakness was a lack of melee units to do the dirty work of actually capturing the city! My one swordsman just barely managed to take Marseilles with one or two hit points left. I realized that I needed a few more melee units, and made sure in the future that I wouldn't face this situation again.

I temporarily puppeted Marseilles, just to control the cultural borders while pushing forward. I had no intentions of keeping the city though, and would raze it later on. Troyes, the nearby French city on the west coast, would simply be razed down upon capture. This game is really silly in that regard; the -5 unhappiness penalty for occupied cities is very severe, and those courthouses take forever to build, can't be cash-rushed, and cost an appalling 5gpt in maintenance. Puppet states endlessly waste your income on buildings that you don't want, like barracks and armories in cities that will never produce military units! Argh. So your only real option is to burn down everything, and replace all of the AI cities with your own. (Of course, the AI also gets mad at you for razing cities, heh.) I've never seen another Civilization game like this, where you literally raze every single city upon capture, except the capitals which the game doesn't allow you to raze. Is this really what the developers intended? It sure isn't particularly fun to play...

Napoleon offered me gold for peace, but that wasn't good enough. I wanted his cities, and I wanted to cripple him right here and now. I could not leave an AI civ that large unattended, or France would simply outtech me and crush me. Here at Paris, Napoleon once again rushed a cannon, and due to some mismanagement of my units, he managed to kill both of my melee swordsmen with it. I had to wait for another replacement unit to come in, wasting several valuable turns, and that cannon was killing my units all the while, with Paris sitting there at one hit point left. Sigh. Well, I was able to get a horseman down there eventually and finally capture the city. With his capital gone, Napoleon was willing to sign peace and give me five cities in the peace treaty. I still had no idea which ones he was giving me, but there was one clear option: burn them all down.

Civilization 5: promoting mass genocide since 2010! I mean, it's not like I could keep these cities, not with that -65 unhappiness. Not going to puppet them and leave my fate in the hands of the idiot AI, not going to build a half-dozen courthouses. That left me with few options, and so I burned them all down. I just wanted to deny these locations to Napoleon, which I did successfully. Now France only had seven cities, and Napoleon's back was effectively broken. Interestingly, I checked all these cities to see if there were any worthwhile buildings inside, and there was exactly one colosseum in Hamburg. Not a single building in any of the other cities, including Troyes and Marseilles. I think Paris had one building inside. Let me just reiterate this point: Napoleon was the dominant AI in this game, by spamming military endlessly, and never building any infrastructure in any of his cities. Does that sound right to you?

Continued on the next page, where I go into more details on the bizarre economics of Civ5.