SGOTM 18 - The Shawshank Redemption

Another point is that Genghis can't hurt Mansa, while Mansa being at war may just mean REXing and Missionary Spam, without competing for Wonders.

If we leave Mansa alone, he may start on The Great Library at any time and if we don't bribe him into a war now and if he does so, we'll have to self-tech Literature if we want to try for The Great Library (Mansa already knows Literature). So, that's a good reason for bribing Mansa into a war.

But this is all speculation Dhoom.
Mansa might go for WW even if we place him in a War. What makes you think he wouldn't? I remember you said the same about Missionary spam, but the AI does what he will. I've never code dived, so I'm not sure what of your statements are strong and what are weak.
If a good reason is based on a weak assumption it becomes a weak reason, no? So, I've no idea how to judge.

The second point, about GK sending his stack at us could also be weak:
1. is there a reason to suspect Mansa being at War with him but not threatening him with units, will stop GK from sending his stack if his mind is to send a stack at us?
2. does GK have a stack of Chariots? Can't we establish that? And is it so that AI can send stacks of Chariots and Archers? I've no recollection of ever seeing that. I don't remember having been attacked in a War by an AI with no metal. My feeling is that GK would mostly play defensively.

In any case, I do agree that we need to have a much clearer plan of whom we want to War and when. A Worker from GK is not worth absolutely any risk at this point in the game. It's not like we are hurting for those 60 silly hammers.

During a war, an AI increases its chances of building units. That claim is a fact that is based on several people having code-dived. Apparently, Missionaries are included in this category (units), or at least the "Missionary Spam" Civ-wide mission isn't excluded from occurring during wartime.


Rarely do you see an AI interrupt a World Wonder build during peacetime. Once a Wonder gets started, it usually stays in the build queue.

The opposite is often true in wartime. Often you will see an AI switch from building a Wonder partway through completing it... and then it seems to randomly decide when to build said Wonder again... to the point that it doesn't necessarily "remember" that it had started on the Wonder in City X and can easily start to build the Wonder in City Y and then switch to building an Archer (or in this case a Skirm) partway through building the Wonder in City Y. You'll sometimes see that an AI has more Gold for trade after having completed a Wonder when it had been doing a lot of warring for this very reason (the AI earned itself some Failure Gold by accident).

Being at war messes with an AI's ability to start and complete World Wonders.

We have a good 10 to 15 turn window where we won't be able to get Literature in trade.

Let's say that we get Aesthetics for Metal Casting but don't bribe Mansa into a war. Well, he's not powerful enough to start a war on his own, while no other AI is likely to declare war on him (Dr. Evil would probably be the only potential candidate, but Dr. Evil won't plot a new war while he's at war with us).

So, anytime during that time, Mansa can start on The Great Library, meaning that we won't get Literature in trade from him. Sure, we could THEN bribe him into a war for Horseback Riding, but we'd still have to self-tech Literature, as we wouldn't have a tech to trade to him for Literature.

If he's involved in a war, he has a lower chance of building Wonders.

If he's involved in a war and he starts to build a Wonder, he has a much greater chance of abandoning that Wonder mid-build... all that we need is a 1-turn window where he stops building The Great Library after we have learned Horseback Riding and we can get Literature from him in trade.


It's also obvious that an AI REXes best when it has more Military Units and doesn't dump as many Hammers into Wonders. Getting an AI into a war, which code-diving has proved increases an AI's odds of building a unit, therefore does help an AI to REX better.


On turns where Genghis (or Joao) is at war with both of us, there's nothing to say that Genghis (or Joao) will send any less units our way than they would have otherwise done... it may or may not happen.

HOWEVER, on the turns where we take a Cease Fire in between Worker steals, we CAN count on Genghis' (or Joao's) stack to head toward Mansa. It won't sit around camping near our empire. That's something which we can count on.

And, in between Worker steals, we have been Cease Firing, so such a period of time will exist.

What's not clear is, once an AI starts sending its stack toward Mansa, and we redeclare war a couple of turns later, whether said stack will keep going for Mansa or will come back toward us... but minimally, we'll have bought us a couple of turns with the stack walking in the wrong direction... so, we'll have double that amount of time from which we'll be safe from the stack (if it walks away from us for 2 turns, it will need to spend 2 turns walking back toward us if the stack chooses to beeline us at that time).


I have no clue how to tell which units Genghis has. All I know is that if I look at the F5 -> STRATEGIC ADVANTAGES screen, Gandhi, Genghis, Joao, and Shaka all have "Chariots" listed as "Their Units -> Possible," while Mansa does not have Chariots listed but Mansa does have Skirmishers listed under "Their Units -> Known."

After we Cease Fire with Genghis, we can ask him repeatedly until he brags about a particular type of unit, but until that time, I don't know of a good way to tell which units he can build, and I don't have any clue of how to figure out which units he actually has.


An AI will not Declare War unless it can build 2 "non-Archer" types of units. However, it CAN send in stacks of whatever units it has available if declared upon, be they Archers, Chariots, a mix thereof, or even Warriors at lower difficulty levels.


Assuming that the F5 -> STRATEGIC ADVANTAGES screen gives us a complete-enough picture, Genghis doesn't have Copper anywhere else. If that's true, the only thing keeping him away from getting Spearmen is us continually stealing Workers that attempt to hook up his Copper Resource.

So, attacking him is not just about getting free Workers, but it's protecting our future investment in Horse Archers and War Elephants. It also means Genghis not building a stack of Axemen, Chariots, and Spearmen to come after us.

You are doing it again, aren't you? Going after all plans and doing everything at once!

War with Joao, while stealing from GK and cutting off his Copper (only copper, not all metal!), while beating Mansa to TGL after trading Lit from him.

^That's NOT going to work. Let's make a realistic plan.

I don't fear a back-door attack from GK myself. You have an archer in Cap that can plug that hole in ~8 turns. I doubt he'll have a SOD down the railroad by then.

Regarding dragging Mansa into a war with Joao, it's exactly as Dhoom said. If Mansa is at war with Joao, some of Joao's units will leave his cities and go north rather than east.... The faster we take out Joao, the faster we get Swinger's Pad. The more units Joao sends north, the faster we'll take out Joao. This does not even assume that Mansa has a SOD of his own to cause problems. Don't forget that Lisbon is a production power house and can likely spit out a unit every other turn...

EDIT: Missed the four posts above which were on a new page...

@ RH

I don't think we're trying to do everything 100%, but we have our primary objective which we should be 100% focused on (capture Swinger's Pad) and secondary objectives (keep GK weak, take out Joao on the way to the Pad, GLib [is this a primary objective], etc.).

In my mind, when Mansa beelined Literature the odds of us getting the GLib went way down unless we're willing to self-tech Literature, which I'm not willing to do personally.

So, I think we need to keep the capture of Swinger's Pad at the front of our minds for every decision we make. Everything else should be considered secondary. If you agree with this, then you understand my reason for wanting to drag Mansa into a war with Joao.

BTW, I think Dhoom and I are 100% in agreement that Mansa needs to be dragged into a war. If anyone feels that Mansa shouldn't be in a war, that person needs to convince one of us (or others that have been quiet lately) of that. What seems to be up for debate is who the war target should be: Joao or GK.

I talked to GK 20+ times every turn and he never bragged about what units he had. Maybe GK doesn't brag about his units. Load up the last save when we weren't at war with GK and see if you can get him to brag. When I was playing, Gandhi was bragging about archers and Mansa was bragging about axes.

Since dragging Mansa into a war is so cheap (another turn of 20-30% science), I think it's a no brainer to do so... it's just a matter of who. For me, Joao stands in our way of Swinger's Pad. I agree with RH that we could just forget about GK now if we want (or after one more worker steal) but we CANNOT forget about Joao.

Keep in mind, if we just leave GK alone and let him hook up his metal but leave him 100% out of the tech trading game, he's going to be a pushover once we have the Swinger's Pad. He's got awesome land, so I'd love to take him out, but we can do so with elepults + an axe or two stack defender. Even if GK has metal he'll still be an easy target.

My vote, CF GK now, steal one more worker (which we could use), CF again in 5 turns time, and then just leave GK alone. Having stolen 3 workers from him and leaving him out of the tech trading game should set him back enough that we can just roll over him later.

GK is NOT a primary objective right now...

^I agree with that. GK is just a distraction. If there was risk in stealing Workers, I wouldn't steal any. But since there isn't any risk...

The risk with bribing Mansa on Joao is that they could make peace before Joao assembles a stack.

It would be better to do it when Joao is ready to War. (i.e. his stack exists and can move to Mansa)

What's going to be more important in the war on Joao?

We should get a Cease Fire with Joao after a couple of turns' time.

After that, we have 2 Archers and possibly a support Spearman who can help with stealing another Worker and then we can get another Cease Fire.

We're not yet at the stage where we're ready to send in units to Joao, since we're going to be building Forges next, probably revolting to Judaism for a turn, and then slow-building Military Units while we regrow from whipping the Forges.

So, we're not going to have enough units to go Pillaging Joao's Copper for a good many turns.

With Mansa going for Currency, he may not go for Math for a while, so we might give him Math for war with Joao and Construction for Currency.

We certainly won't get Currency with Math... the tech-trade is too lop-sided to try chasing after partial Research into Currency.

We should be able to pick up Iron Working from Gandhi at some point in the game, so no need to use Math to get it from Mansa.


One very real fear is that Mansa will make Peace with Joao when we're ready to move our units into Joao's territory... when that time is precisely when we will want the two of them to be at war with each other. Let's look at our precedent:
Turn 30, 2800 BC: Mansa Musa has declared war on Genghis Khan!
Turn 45, 2200 BC: Mansa Musa has made peace with Genghis Khan!

If it's going to take us roughly 10 to 15 turns to build enough of a Copper-Pillaging Party and get it to the front lines at Joao, then Mansa is going to leave us fighting Joao all by our lonesome for a 10-turn period that may very well coincide with exactly when we want Mansa to be at war with Joao.

OK, then I'm with RH. I say we not put Mansa into a war at all because I see him warring with GK as a complete distraction. Mansa seems to be in missionary spam mode outside a war, so there is no need to drag him into a war to "convince" him to spam missionaries. Let's just trade MC for Aesthetics and get some extra "Fair Trade" out of the deal.

The bad thing about this is that in 15 turns when we want Mansa to declare on Joao, what are we going to use to get him to declare? We want HBR for Literature and Construction for Currency. So there will be no way to get Mansa onto Joao in 15 turns time. That is why I prefer to sick Mansa on Joao right now... EDIT: while it's cheap (i.e. one turn of 20% research).

Or we could wait a few turns for Mansa to learn Currency and make the trade of MC for Aesthetics + gold. We don't need Aesthetics right now. There is a risk that Mansa will start an Aesthetics wonder, but even if he does, I don't recall Mansa ever refusing to trade a tech, do you?

I just ran a test and Mansa "had his reasons" for not trading Literature once he had started the GLib in one of his cities. So there is a risk of not getting Aesthetics for some time should Mansa start on one of those wonders. There is a very good chance that he'll start on the GLib before we can get Literature too. Although, Mansa needs a library before he can build the GLib.

@ Dhoom Can you look at Mansa's cities in the real game to see if they have a library? I can look at the latest uploaded save (which I will do now) but it may not be the latest.

It would also be nice to figure out how to figure out what "Sabotage Building" tells us. I'll look at that too.

T55, 1880 BC: EP Values with Mansa:
Mansa spent 138 EPs on us/We spent 236 EPs on Mansa

Djenne (Mansa's capital)
Sabotage Building 402
Sabotage Production 136
Steal Treasury 466
Poison Water 160
Forment Unhappiness 160
Support City Revolt 871
Steal Technology 194
Perform Counterespionage 134

Timbuktu
Sabotage Building 235
Sabotage Production 157
Steal Treasury 259
Poison Water 157
Forment Unhappiness 157
Support City Revolt 851
Steal Technology 189
Perform Counterespionage 131

Kumbi Saleh does not have any Buildings.

Mansa is running Organized Religion.


I have no idea what a Library Building looks like, so I'll play around in a test game to figure that out then look in the real game.

I library has 3 humps with pillars out in front. I don't see one in any of Mansa's cities (EDIT: in the last turn you uploaded). I do see a monument and rax in Timbuktu and a granary, rax and ToA in Djenne. So I think we have some time on the GLib as Mansa would have to complete a Library (= 69H for an AI). So we can track Mansa's production each turn. If he completes something > 69H, then getting Literature is at risk as he could start the GLib at any time. My gut says that if we "beeline" HBR, we can get it and get Literature from Mansa before he completes a library and starts on the GLib.

Mansa currently has 17H into a build in his capital, which is where I would expect him to build his first library.

T55, 1880 BC
Pictures of Mansa's first 2 Cities.


EDIT: So, from my newbie Building-gazing skills, it appears that Mansa does not have a Library anywhere.

Note that I am wantonly pillaging older uploaded files in OSS SGOTM threads, so if anyone wants any of those, they should go download them before they all disappear.

Dang, is appears the "Sabotage Building" is to destroy the cheapest building in the city, not the most expensive. I put a library in Timbuktu in a test game and the value was 1004. I then also added a monument and the value dropped to 334.

So we just need to keep an eye on the Sabotage Production values and/or look at the city each turn trying to spot a library in any of Mansa's cities.

If you hit the "Toggle Bare Map" button above the mini map, all units are removed and the dark areas turn light. It's also good to turn off tile yields so that the food, hammers and commerce don't block your view.

The granary is a building with a round top. There is one in Djenne but not Timbuktu. The one in Djenne is right in the middle of the city.

The monument is a statue of a guy, which can be seen in Timbuktu.

The barracks has a green roof and also has a tall tower. Both cities have one.

If you go to the F12 Civolopedia, you can browse through the buildings to see what each one looks like on the map.

OK, so where do we stand on the following issues:

1. Drag Mansa into a war or not? If so, with whom?

2. Trade MC for Aesthetics now, wait one turn at ~30% research for a possible war bribe, or wait until Mansa has Currency and try to get some gold?

3. Keep stealing workers from GK or stop now?

My votes are:

1. Drag Mansa into a war with Joao. It will start as a "fake" war since neither has a standing army right now, but Mansa will eventually send pillaging skirmishers plus maybe axes, which will cause some trouble for Joao and Joao will likely send a few units Mansa's way at some point.

2. Wait one turn at 30% research and trade MC <-> Aethetics + war on Joao.

3. Steal at least one more worker by CF this turn and hope to get another one next turn or soon thereafter on the copper. After one more worker, we can reassess where we are, how safe we feel at Corn and the back door, etc, and either continue to steal or just leave GK alone to settle all of the spots we'll eventually take from him.

Alright, I'll concede this argument and agree with declaring war on Joao.

I, personally, think that it is more important to get Mansa into a war now than not at all, if we want Literature. Given our luck with Mansa going for Literature so soon, we'd be fools to pass up on getting Literature. The Great Library is one of the key Wonders that we all agreed we wanted to get. Plus, an early Heroic Epic and early National Epic when you have Marble and are Industrious are just too big of prizes to pass up on.

Cease Firing with Genghis this turn is better than waiting to do so next turn.

Mansa's Jewish Missionary will probably never make it our way, but I suppose that there are costs to every decision. There's always a small chance that he'll find another way across the map to get to us other than through Joao's lands.


So, the plan will be:
T55
Cease Fire with Genghis
Move our Warriors at Genghis into Worker-capturing position (one 1W of the Copper and one 1S of the Copper)
Adjust our Science Slider with Aesthetics selected (I have pictures from the test game to compare against for values that will and will not work when Gandhi also knows Aesthetics)

T56
If Mansa is Annoyed toward us, manually revolt to Judaism
Trade Aesthetics + War with Joao from Mansa <-> Metal Casting
Research Horseback Riding (0% Science for a bit to save up Gold)
Be prepared to Declare War on Genghis at any time
Move our Spearman westward and our Archer to Gold City

It looks like we'll need a 40% Science Rate in the real game, otherwise we'll fall short of the needed Flasks (I have the needed Flasks down to a range of roughly 3 Flasks difference between what's enough and not enough and am using pictures of the Tech Progress bar at the top middle of the main game screen to compare the total amounts, since it's hard to know how many overflow Flasks you actually get).

That difference is probably because in the test game, we still had an extra Foreign Trade Route in 1 City for several turns.


Are you still okay with the plan by using a 40% Science Rate for 1 turn?