BuShiDo - the Way of Exploit - Tackling 1 vs. 7 Deity Standard Pangaea

[maltz]

M. Intermission - I

Here is how Japan's world looks like in the Medieval Era. Not much extra exploration from the Ancient-Classic Era map.

Spoiler :

056. But I know where my next target is! Now I have successfully secured the eastern end of the Pangaea. The workers are now trying to connect Kyoto to Osaka, and I will launch probably the second most important campaign of the game there (the most important is of course, the first campaign!) I was allied to Copenhagen as I completed its "Kill 3 Songhai units" mission.

Spoiler :

057. Remember I mentioned that a barbarian archer joined the party (the Siege of Gao)? I definitely know what they are here for - they are here to sabotage Gao's resource. I would be glad to watch the smoke if the city was still under Askia's rule. I would be... actually even more glad to see the smoke when the city is under my rule.

The reason is that (the current, patch .221) Civ5 treats a diplomatic deal as a single entity. For example, if I sell 4 resources, including to soon-to-be-lost Marble to Isabella for 1200 gold, guess what would happen on the next turn? The Marble is lost, and the entire deal is therefore canceled. I get the Whale, the Gold, and the Spices back. And of course I get to keep that 1200 gold. I just regret that I was still running a -15/turn deficit. Or I can even get a little more out of it.

(Wait, I forgot to include my strategic resources, too! Doh!)

Spoiler :

058. This is the outcome on the next turn - all four resouces come back at once! And I keep the 1200 gold!

So, potentially barbarians are really welcomed guests to your territory when the to-be-pillaged resource is actually available for sale. I heard that people could pillage their own resources in the past, but patching has fixed it - only part of it. I heard that now, people can still build a Fort on a resource and improve it back after the deal is cancelled. I have never tried it out, though. I find that an exploit is only fun if I discover it myself.

Spoiler :

059. Now I have brought down one of the AI Civs, they certainly like me less. Throughout the game, my resources will mean less and less to them. But fortunately my economic power will grow more and more powerful to sustain my needs. I guess we have to thank the AIs for parenting us when we are weak and needy, and train us to be independent gradually.

Spoiler :

060. Now I am as rich as a Deity AI! I have a long shopping list!

During the first war and a few following turns, the capital finished my planned military buildings: Barracks, Armory, and Heroic Epic (Osaka's first building was Barracks). So I can immediately buy 2 Samurai to make them instant level-3 Open Terrain specialists with 15% combat bonus! They will be beasts! It is actually 10 gold cheaper to buy them directly than buying Swordsman and then upgrade them. And of course I don't want to do it anyway for reasons described before (picture 033) - but I should be able to upgrade the unit before promoting them, therefore delaying the promotion to the next turn.

Spoiler :

061. Time to activate the Patronage tree! Also, after Gao has finished revolting and its citizens return to work, my income returned to positive (albeit a very low positive). The worst is over!

Spoiler :

062. Very soon I used up 6 units of Iron - 6 Samurai. I can't last very long into mid-game without ranged support. So I defintely want to get some Catapults out. There was no more nice 6-unit Iron to settle on, so I turned my attention to City States. Luckily, Lhasa has it, and it is also a Cultural CS!

Spoiler :

063. A perfect match of my needs - with the exception of no extra luxurious resource for me (only a frequently-pillaged Gold mine). I wasn't reading the numbers carefully - I thought 1000 gold would give 160 influence points - I should have spent 500 gold first and wait for my next Patronage policy to renew the alliance with enhanced influence increase.

Spoiler :

064. Now I have 6 more Iron, so the shopping continues. Ordered two Catapults, obviously.

Spoiler :

065. So here goes the army, Four Samurai and two Catapults, (trying) to conquer Egypt! I have 2 super Open Terrain specialists and 2 not-so-super Open Terrain specialists. I didn't bring bring any Rough Terrain specialist, as I did not see any rough terrain inside Egypt's border (for what I could see, anyway). Later I truly regreted of not bringing in a rough-terrain specialist. Two Samurai stays at home to take care of barbarians, which were quite active at the time.

I only assigned two workers to construct the road between Osaka and the Capital. The rest of them are assigned to convert Farms to Trading Posts in ex-Songhai cities.

Spoiler :

066. The peace treay expired on Turn 84, but I needed a few more turns to travel. The leaders of the pack soon spotted an Egyptian settler escorted by a weakling Warrior. From the way the border is drawn, I can safely say that the settler must be at least 3 tiles away from any existing Egyptian city by the end of this turn. Since I know the AI loves to spawn its cities as close together as possible, I have exactly one turn - this turn - to get rid off the settler before it transforms into an annoying city.

So what should I do? Declaring war! Not before I got another 1000 gold from the opposing team, though.

 

[maltz]

N. Grinding down the Second Target

In a regular Deity game AIs, usually get their Knights and Longsowrdsman between turn 90 and 100. And since the AIs here advances even faster, I didn't expect this second war to be easy.

Spoiler :

067. War is declared and the Settler is no more! Previously I already noticed a 6-unit Iron sitting in Egypt's territory. That improvement definitely has to go!

Spoiler :

068. I know I am going to need more ranged support for this war. So it is time to use the latest AI donation to get 2 units of Capapults and send them to the front line.

Spoiler :

069. Got enough happiness to start the first Golden Age, and the first Great Engineer is born. I am still quite far away from Education, though. So the Engineer has to sleep first.

Spoiler :

070. Somehow, Egypt think recapturing the Worker (used to be Settler) is a top prioty - so the AI sent out a Chariot Archer to accomplish that task, essentially kissing that Chariot Archer goodbye. I can kill the Chariot Archer easily and even recapture the worker back, allowing it to be recaptured by another Chariot Archer on the next turn...

Egypt is quite rich at this point, and it might be able to afford to buy one Chariot Archer per turn all the way to the Future Era. I can't afford to play the game, though. I want to take some city! And since I am not very powerful yet, I decided to attack the weaker city of Memphis (where the Iron mine is located) first.

Spoiler :

071. The AI team still holds a large technology lead over me (But I again have the largest army! Yes!). I am still curious about how teamed AI's technology advance compares to a regular game. So here goes more comparisons:

Spoiler :

This is the custom-map game that I mentioned with otherwise identical settings. The technology leader got to 27% of the tech tree on Turn 82, and according to this picture above, 41% of the tech tree on Turn 112. If I interpolate the progress (not a very good idea, but good enough for... getting an idea), at Turn 90, the tech leader should have a progress of 33%. Quite a bit worse than the AI team's 38% performance here.

Spoiler :

This is from the OCC standard Pangaea game. 32% at Turn 86 for the tech leader (Darius seems to be pretty good in terms of technology in mid-game, eh?) The next data point I have is 135 turn for 51%. So if I interpolite it, I get 33.5%. Very close to the last example! So the AI team outperforms the tech leader here as well. One more game to compare:

Spoiler :

This comes from an OCC standard Earth map with standard settings. The tech leader is 31% at... Turn 90. No amatuer approximation required.

Teamed 38 vs. non-teamed (33%, 33.5%, 31%) - the verdict is out! The AI team techs faster than an un-teamed technology leader in a regular game... at Turn 90 and before! I will do still more comparisons later.

 

[Aaroc]

The AI should be hard-teching faster on teams than they would on their own, but I'm also pretty certain that they can't sign RAs with eachother, so that should slow them down somewhat. Keep an eye out for RA popups. I could be wrong, but I'm fairly certain when I attempted a team game, I was unable to sign an RA with my partner.

Now, if you had set he AI up to be on 2 3-civ teams, then they would probably be blowing you away.

ETA: .. er, more than they are now, anyway.

ETA2: Also, I wish the demographics screen would remove conquered civs from the stats >_<

 

[maltz]

Spoiler :

072. Back to war! A second CS fell victim to the AI gang. The conquered City State is Hanoi. At the time, Geneva was conquered already. I am glad I didn't follow that up with 500 gold!

As I approached Memphis, I saw very little defenders. I was quite glad for a moment, as I think Egypt either has a very little army, or has sent them away to fight a City State or something. This was Turn 90.

Spoiler :

074. And guess what happened during the end turn! A very coodinated showing during the end turn - the map was filled with the AI team's units! Gotta give some credits to the coders for arranging surprises.

But I am not too afraid of the horde. From a previous OCC domination game, I learned that the defenders does not actively attack if they know they can't kill me in one turn. (A side effect from "improved AI" - they really want to kill your units, and won't do anything if they can't achieve it.) If my open-terrain specialists stay in the open, they are virtually indestructable for these lowly units of Pikeman, Companion Cavalry, plus Iron-deficient Catapult and Long-swordsman. So that huge horde actually ignored me!

Spoiler :

075. And this allows my Samurai to advance quickly to the siege positions with minimal loss of hit points - cutting through Pikeman like heated knife on butter: 36.9 vs. 9.

Spoiler :

076. Got second Patronage policy. One more to Scholasticism!

Spoiler :

077. I was amused to see Alexander's Companion Cavalry ignored me and ran to the right. For a moment I thought they are going after Osaka, which might be vulnerable to three of them. But then I realized that they after the City States, which feature nothing but Pikeman. I am speechless...

And my Samurai and Catapults are ready to assault Memphis, with all Egyptian units watching nearby. As long as I don't leave any unit any chance of dying by attacks by the time I press End Turn, they will be left alone.

Spoiler :

078. Last hit on Memphis! That was easy!

Spoiler :

079. Discovered a Natural Wonder. Not within Memphis' border, unfortunately.

Spoiler :

080. I immediately went to Ramesses II to get a peace treaty for 10 turns. Why don't I press on and finish Egypt? It is a lot more difficult to say than done at this point, because Thebes (Egypt's capital) has 30+ city defense and a large number of forest tiles to its north. And each tile has an unit hiding in it. Since I have no rough terrain specialists, I am limited to Open Terrain, and there are only two Open Terrain tiles, and only two Catapult tiles to attack Thebes on the east side. There is no way I can take down the city without reinforcements, which will certainly be ready after 10 turns.

So I chose to wait for 10 turns. Every peace-war cycle generates lots of profit!

Spoiler :

081. I just busted a Barbarian camp for Kuala Lumpur to turn them Friendly, so I may as well push it to Alliance with 500 more gold for their Incense. Also, I will get to Scholasticism a little faster this way.

Spoiler :

082. Egypt really hates me now - only 136 gold for the resource. But I will take that 136 gold over nothing on any day.

Spoiler :

083. As I felt the pain of having no Rough Terrain specialist, I am definitely going to rush one now. I have 10 turns to walk them to the front line. Also bought the 5th Catapult. Not showing in the pictures is that I paid 500 gold to CS Singapore after busting a barbarian camp for them during the ceasefire.

Spoiler :

084. I did bribe Napoleon to declare war on Edinburgh later, that's why they also gave me the mission to kill N units and became friendly to me once I killed enough Egyptian units. So here comes the free unit! A useless Chariot Archer. I gave it right back. But thanks to the free unit, I accidently completed another City State mission about discovering Greece. There are some visible border to the east. That places Greece west of Egypt.

Edinburgh was under attack! Its City defense should be enough to keep it alive for a while. But eventually the AI team will overwhelm it.

Spoiler :

085. Education is researched, and therefore I was able to accelerate the Porcelain Tower with the Great Engineer. I will save that Great Scientist to rush Rifling along with Kyoto's own G-Sci, which wouldn't come for another 15-20 turns.

Spoiler :

086. My favorite survey - every time I see those happiness figures, I am again full of motivation to abuse the cr*p out of AIs.

Spoiler :

087. I plan to build both the Natioanl Epic and a Garden in the capital, so I can accelerate my next Great Scientist. I should be able to get one on Turn 117 or so.

Spoiler :

088. I am still trailing by 11% of the tech tree - the difference seems to be quite constant. Peace treaty expires on Turn 104, and I am very ready to wave goodbye to Ramesses II! Let the war resume!

 

[maltz]

Now, if you had set he AI up to be on 2 3-civ teams

Interesting idea. I will try it someday. But I guess I will again use the old "bribe A to war B" technique to break all RAs, and to give myself emptied cities to capture.

 

[maltz]

Spoiler :

089. Thebes - 20 population and 34 defense on Turn 104! Full of juice!

At the time I already had my second Great General, so I split my army into two. On the east, I have two Open Terrain specialist Samurai responsible for pushing forward, followed by two Catapults, which is protected by one Open Terrain specialist in the back. There are quite a few Egyptian units wondering around back there.

On the north, I have one Rough terrain specialist (the latest addition) Samurai, accompanying two Rough Terrain specializing Catapults. On the west, I have another Open Terrain Samurai, a Catapult, and my Scout-promoted Archer just to reveal some more maps if possible.

Spoiler :

090. The Egyptian army is revealed! The Knight is out! And I am going to pillage that horse... nope, these are cows. It was a little late and I mistakenly pillaged a cow pasture (and started wondering why the knights didn't get that -50% penalty - the horse can't come from City States). The real horse ranch is to the west.

Spoiler :

091. Of course, I collected another round of donations from the ever-happy and rich AIs. First, I rushed an Aquaduct in the capital. The size of this city is now a laughing stock in the world. And since more than 50% of my research comes from it, I better make it as populus as possible.

Spoiler :

093. I was only one turn away from Scholasticism, so I donated 500 gold to Ragusa to make it an ally. Now I have 5 of the 7 City States on the east side of the Pangaea allying with me. Military CS can wait.

Spoiler :

094. Here it is - another keystone of the 1 vs. multiple strategy - Scholasticism! With 5 City States each contributing about 9 flasks per turn, I enjoy a 44% increased research rate at the time of policy activation. This is huge! (Less than 10 turns ago, I was on only +55 per turn - without University, research specialists, and Scholasticism.)

Spoiler :

095. There aren't so many Egyptian defenders left in the capital. During the peace time, Ramesses had a bright idea - he wanted to attack one of the CS to the north, and split half of his army away. That split army is now belong slaughtered by one of my Open Terrain beasts in the back.

My units advance at a good pace, killing the defenders according to their specialized terrain. I want to get rid of that horse ranch to the west - this time I am certain where it is!

Spoiler :

096. Egypt has Musketman! Good for her. That means I have to work fast! Luckily, that Musketman came into the firing range of 3 Catapult. So there is no more Musketman in 20 seconds.

The Knight who used to guard the horse ranch have made a smart retreat... to protect a worker in the back.

And the barbarians at home have subdued a bit, so I sent one of the two staying Samurai's to the frontline to help. That decision was made about 10 turns ago. It took quite a long time for them to walk here.

Spoiler :

097. The catapults aren't doing too much to Thebes, while Ramesses keeps buying units out of Thebes. Fortunately, I have completed Physics, so I can instantly promote three Catapults to Trebuchets and send them into the siege party!

Spoiler :

098. Somehow, I thought making a screenshot of tech tree and discuss my next war in the middle of a heated ongoing war was an interesting idea for this section. So... I am going to coplete Gunpowder on Turn 116, during which I will receive a Great Scientist from the University of Kyoto. Then, together with the Great Scientist from Procelain Tower, I will rush Rifling.

That saves me 19 turns. Not a lot, but might be enough to buy me the time to severely damage my next target - don't know whom yet.

Spoiler :

099. Double-checked the G-Sci progress. Yep, Turn 116 is the magic number!

Spoiler :

100. Thebes' defenders were gradually killed off and I am almost ready for total assault - I have to move a dying Samurai out of harm. I didn't realize that I can switch place of two units across another unit (the Catapult in the picture above). Nice feature. I know the Samurai's hold part-time job as jumping Ninjas.

Spoiler :

101. And here the final assault begins! Thebes' hit point was reduced to lower-red in one turn, and I happen to have a Samurai left to act - if he attacks the city, he may take it, or die from trying. Do I take the chance? Of course!

Spoiler :

102. And here goes the end of Egypt! Two 7-turn wars for two large cities.

Spoiler :

103. Songhai's capital Gao contains two wonders - the Great Wall and the Hanging Garden. Egypt's capital Thebes also contains two wonders - the Colossus and the Great Lighthouse. Not so useful.

Spoiler :

104. My tech gap has been slightly narrowed. That's nice! Now let me do another comparison...

Spoiler :

I showed this one before, but now the Turn number mathces better. The teamed AI is only doing a little better than the unteamed AI leader.

Spoiler :

This one comes from a 2 vs. 4 game on a small Earth map, with standard settings. With four more turns on it, the four-teamed AIs actually have the same percentage of technology researched! But shouldn't 7-teamed AIs do a lot better than 4-teamed AIs? I am really puzzled. More comparisons to come later, I guess.

Spoiler :

105. There is no new luxurious resource in Thebes, so my happiness level took a straight free fall to red. It is OK - war is over and I am again well-funded. Purchased Colosseum in Kyoto!

Spoiler :

106. Rifling is coming soon. And it seems obvious where I should go next - there is a horde of Chinese/Greek Pikeman to the south. If I don't finish them off before they are promoted to Rifleman, I will be in some serious trouble. So China is the next target! (Greece should more to the west of Egypt based on Edinburgh's map...?)

And Washington has popped an annoying little city just north of me. On the bright side though, the City States are going to take a good stab at it when they get their Rifleman upgrades.

 

[Tlc2011]

 

[bryanw1995]

Great thread, keep up the good work!

 

[Rubin]

A great and, uhm, terrifying read! Excellent display of strategic stream-of-thought with a predefined clear goal, but also a worrying display of the diplomacy and team mechanics...

Unlike most of the strategy "guides" I've browsed through, this one resembles my own Japan gameplay strategy the most, and I've enjoyed reading every bit of detail! The samurai is awfully powerful, and peace-time is all about moving siege units into position. The execution here is smooth and beautiful, maltz!

Twisting the mechanics of city states into your favor is clever, but the grand prize goes to the 1200 gold marble deal

As you note, seeing the happiness bonus of the deity AI motivates you to abuse the AI--and I agree; but I also sense an indication that fixing these "exploits" would improve your gaming experience (for future games; not this one). The way you've thrashed the AI so far clearly shows that the "AI team" is clueless, and that you've got the upper hand

I hope Firaxis devs read your thread! It's pure elegance how you manipulate the game

 

[baldfalk]

Regardless how this game turns out you deserve massive props.

 

[Rudimentos]

I'm not quite sure, but how many RAs can a teamed AI do? My guess is that they can't do even one because they are in the same team. That would explain their slow research rate although their beakers are combined.

 

[Architect]

This is an excellent read thank you for doing it.

I find it very hard to play CiV in the way that you do because it feels like cheating! However, I love that you are doing it because hopefully they can make some changes to the way the AI and game works to remove these obvious trade exploits.

The lump-sum trade exploit you are doing has to be changed. I would suggest:

1. Only allow per-turn deals with new relations and the AI should only accept lump sum deals if you have demonstrated friendly relations for generations and your diplomatic rating should be hit much harder if you take advantage of them.
2. Deals could suspend during war and resume upon peace. While the deal was suspended you would not be able to use this resource to trade, build with, or for your own happiness.
3. Allow lump-sum trades but once you broke one deal with an AI it would never trade with you like that again. I know I would never trust someone again who did that to me.

 

[blac]

Wow, just wow. I never thought someone would be able to take on 7 deity level AIs on a team!!

 

[Morningcalm]

Nice presentation and explanations! Good strategy too, and I especially appreciate your patience having to deal with shuffling samurai around every turn (Civ V is especially annoying in its micromanagement of troops thanks to IUPT).

 

[vanjito]

You're probably going to out-tech the AI's since they cannot sign RA's and your /turn is pretty high.

Maybe annex the largest city (that has University, Mountain if any for Observatory) and hiring a couple of Sci's?

 

[Omega124]

Yawn. Boring.

Huamn vs. 21 Deity Team. That is interesting.

Moderator Action: Warned for trolling. If you don't have anything constructive to add to a thread you are welcome to not post in it.
Please read the forum rules: http://forums.civfanatics.com/showthread.php?t=422889

 

[_hero_]

Great game so far, Maltz. I think part of the explanation for the early boost in tech that the AIs get when teamed is that when one AI discovers a tech from a hut all of the team gets that tech. You figure in a normal game a single player is going to get 1-2 techs if they explore well. With 7 AIs on a team, that's more like 10 techs. And ultimately the reason that it is slowing down is because of a lack of RAs. I imagine if your game continues on long enough, the 7 AI team will actually fall a decent way behind a regular game.

Actually I know they will since you will be taking them out.

I wonder if teamed AIs get their tech costs reduced every time one of the allies gets eliminated? Interesting experiment.

 

[Itstoearly]

Wait, I thought that since they were all on a team, they share techs?

 

[Aaroc]

Wait, I thought that since they were all on a team, they share techs?

They do share techs. So if one of them pops a tech from a ruin, they all get the tech. They cannot, however, sign RAs with eachother because they are all on the same time. It's is still unclear as to how they gain their technology. That is, at what rate they gain science. I don't think it's a combination of all of their science together, because they would likely be teching much much faster than they are. My guess is that it goes by either the rate at which the civ with the highest science rate would tech on their own, or some sort of average between all remaining team members.

 

[Rittmeyer]

very impressive, really top stuff. only sad part is that it reinforces how much iron affects the balance in the game.

games have be entirely different when there is no acessible iron. playing normal deity I`m already winning most games when I can longswords rush, but I get into trouble when I have no iron at all. I play small maps, so it aggravates the problem.