SGOTM 03 - The Real Ms. Beyond

[Tatran]

Lurker's comment :

IIf we see an archer and a settler on an AI galley, they won't attack, but will found a city. If we see 2 archers, we are going to get attacked.

IIRC, you can't see units anymore on boats after the latest patch in Warlords.
And that's a nasty change, but you're playing the Vanilla version.

Crazy leaders, like Montezuma, can declare very early. Pillaging is something
you don't want if you're going for an early wonder.

A screenshot from a backdoor diplomatic win. 62% was required.

 

[Iainuki]

Something fishy is going on here. I'm trying to figure out what the odds are that the AI attacks us early.

I whipped up a sample game where I put Mao and Alex on a continent right next to me (Peter). Just like in our game, they are the only civs on their island, and I'm well within range of theirs. I ran the game out to 700BC by building a single warrior and then building the pyramids (at a slow rate; some AI built them in 895BC!). Not only did both AIs eventually go to +1 peace (Cautious), but neither one developed sailing by then.

So, I went back to 3100BC and gave them both all three early strategic resources (horses, copper, iron) in their fat cross. And I gave them Sailing. Same thing. No worsening of relations, no "enough on hands".

It's possible that I was so close to them that they got the +1 long-time peace diplo bonus before they planned a war, but....

Anyway, I'm just having a tough time interpreting what that "enough on hands" means for us.

Did you make the AIs aggressive? I think aggressive AIs have a hidden diplomatic penalty that makes them hate the player more.

The gap between "We have too much on my hands" and the actual declaration can be considerable. I've seen gaps as long as one hundred years in the post-1500 era, so 20 turns, but that was on normal speed. I don't know how being very early in the game or being on epic might affect that. I've also seen it turn around: one game where I was Washington and was facing Caesar, he had "too much on his hands" so I started a military build-up, and he apparently reconsidered because he never attacked me. (Of course, I used the troops I'd built to kill him after he decided he no longer had too much on his hands: no point in wasting the production.) I've never seen an AI declare so early, but there's a first time for everything, and I'd rather not take the chance. The other thing is I've never seen an AI declare war until it decided it had mostly run out of places to expand to; but we don't know if that has happened yet.

I notice that Mao and Alex both have Bronze Working. Mao has had it long enough so that he probably could have mined any copper if he had it. (Though, sometimes the AI is really slow about hooking up resources, so this is not a certainty: he could show up with axes next turn.) Alex didn't switch that long ago, so who knows if he has copper? Will the computer attack with archers if it has them? I don't know how strict its scripts are about defensive vs. offensive units. Did anyone notice when Mao and Alex got trade routes with us? That would tell us when they got Sailing, so how long they've had to build galleys.

The more I think about it, the more worried I get. I think our next tech has to be Archery. This is going to be a huge pain, but I think that until we can afford to build real defenders, we're going to need to keep checking Alex and Mao's diplomacy screens, preferably every turn, and see if they start threatening us with better units (chariots, horse archers, axes, or phalanxes). A fully-fortified warrior has slightly less than 50/50 odds against an attacking archer because of first strike, and slightly better odds given +20% or +40% culture defenses. I suppose we could build walls in St. Pete if we come under attack. My other big fear is pillaging, but I don't know what we can do about that: our nets are vulnerable, but even if we had Sailing and the production to build galleys, the coastal defense bonus isn't large enough so that there wouldn't be a substantial risk of losing galleys to bad luck.

It would be better if we could keep them from declaring at all, but I don't know how to do that without spending too much production. They have a lot more power than we do, so we'd need a lot of archers to frighten them.

 

[Kodii]

I have an intuition that tells me neither Alex nor Mao has any plan to attack us. Perhaps they are preparing to declare war on somebody we haven't met yet. Remember, there are 17 AIs. There is a lot mentioned, so I need to look through it so I can update the summary post.

EDIT: I updated the summary post, but I have left out war considerations until we have a firm decision on that.

 

[Compromise]

Thanks Tatran, always good to have lurker insights. Feel free to make any comments on our game at any time!

I just finished some tests with worldbuilder. I'll give details later because I'm rushed, but here's what I believe "have enough on our hands" means:

First, I think it's set by a random number results based somehow on the civ's aggressiveness and the relative power ratio between the human and that AI. (This is speculation.)

Next, 11 turns after the AI first "has enough", if the AI has mustered enough strength, it attacks you. Of course, the AI usually has to spend a *lot* of time mustering "enough strength."

I did this by setting up a game like this one with aggressive AI settings (and all the other settings) and me (Peter) alone just offshore of an island with Mao and Alex on it.

I just ran turns with 1 warrior and built the pyramid (which fell to Rome in 1300BC on an island barely big enough to put his second city on.

Eventually, I noticed that Alex "had enough". I tracked the date of the change to between 2050BC (not mad) and 2020BC ("had enough"). I let the game run its course until he declared war with 2 galleys, 2 chariots and 2 axes in 715BC after a 5-turn sail from his capital where he'd been building up troops.

I reloaded to the turn before he declared, filled my city with 5 tanks and the surrounding coast with 4 battleships (all within his view). He still declared!. This means (to me) that once the AI decides to declare, there is *nothing* you can do to stop him. He will attack you when he can.

I wiped out his forces before they landed to see what Mao would do. He too had "had enough" between Alex's first change and Alex's attack. I kept the troops. Eventually, Mao too declared war, but it was with a passing galley that took out seafood nets. I found a city with some galleys in it--presumably building up an assault force--but his galley got to me first.

I then went back and did several forms of adding troops to the AI city, coast, waters, etc. The earliest I could get him to send out an attack force was 11 turns.

So, I think the order is:

1) Random number generator result; causes change in build script in a major coastal city

2) 10-11 turns later, if sufficient forces are present, launch an attack by sea

3) If empty galleys are near the human when the assault force launches, they declare an start ripping up nets.

4) The assault force will not attack a coastal galley which blocks the way. But the pillaging galleys will.

Must run, but more comments later.

Bottom line: Although I think it will take a while for the AI to muster enough troops to attack us, I think we're going to prioritize blue dot *a lot*. Maybe even take sailing after mysticism. And I don't know how we want to prioritize troops vs. pyramids.

Iainuki, I haven't had time to digest your post yet (just skimmed it) because I wanted to get this out, then I may have to go offline for a while.

 

[Compromise]

Some more info on my test game with the war declaration. At the start (4000BC), I made sure that both Mao and Alex had copper, iron and horses within their fat crosses.

I also kept my start island far enough offshore that we didn't meet until the same dates that we met them in SGOTM3. (I opened WB and placed a warrior nearby and "met" them at the same dates as our game.)

After meeting both civs at the appropriate times, I added grassland peninsula's so they could look with envy at the gloriously resourced (and WB-improved!) capital I'd created for myself.

And, as I said, my entire defense consisted of a single warrior. I wanted to give them every incentive to want to attack me as early as possible.

Oh, and I gave them Bronzeworking (if they didn't already have it, Alex got it about 500 years before the date in our game) on the turn before they "switched to slavery" in the log in our game. And I gave them Sailing around 2000BC too (neither had it by then).

So, I completely stacked the deck in the AIs favor. (I think I even put an extra 5 goody huts near each capital too, in the hopes that they'd pop some early techs and I wouldn't have to give them anything.)

There were several iterations through the same war cycle. In one, he sent an assault force of 2 galleys and 2 chariots (one galley was empty) despite having oodles of axes dropped in the same city.

On a side note, a very-cramped Cyrus had a settler/archer galley roaming around in 800BC.

What I didn't test was what effect AI resources have on the decision to attack. It may have been triggered (or even become an option) upon the discovery of copper when they got BronzeWorking.

Bottom line: We are going to get attacked by both Mao and Alex. The only question is when and by what.

We may want to seriously reconsider whether or not we build the pyramids.

 

[Kodii]

I still say go for the Pyramids. I still don't think they'll attack, but we cannot be sure of that. I say research Archery, and get the red dot to produce some archers in its build order somewhere. Either that, or we build archers while building the Pyramids, in the same fashion that we planned to build the blue dot settler.

PS: Don't forget SGOTM2

 

[Compromise]

Did you make the AIs aggressive? I think aggressive AIs have a hidden diplomatic penalty that makes them hate the player more.

I probably answered this in there somewhere, but in case it wasn't clear: yes.

Did anyone notice when Mao and Alex got trade routes with us? That would tell us when they got Sailing, so how long they've had to build galleys.

It was between the last two uploaded saves.

Note: Mao is much more powerful than Alex if you look at the demographics screens (there's one huge leap in production, equivalent to our quarry). Given that he got bronzeworking first, I am almost certain he will attack us first.

Another unrelated note: Mao has a great fondness for building the Great Lighthouse.... I saw it in 3 of my test games.

The more I think about it, the more worried I get. I think our next tech has to be Archery. ...My other big fear is pillaging, but I don't know what we can do about that: our nets are vulnerable, but even if we had Sailing and the production to build galleys, the coastal defense bonus isn't large enough so that there wouldn't be a substantial risk of losing galleys to bad luck.

The problem with archery is that it is a defensive tech. It seems to me that the AI will attack with chariots, axes and swords but not archers. So, we probably need offensive units to eliminate them.

It would be better if we could keep them from declaring at all, but I don't know how to do that without spending too much production. They have a lot more power than we do, so we'd need a lot of archers to frighten them.

My investigations have convinced me that we cannot dissuade them. The only way to stall their attack is to block the path of their assault force. (There seem to be two AI scripts for the galleys: attack (which pillages nets and attacks our galleys) and assault (which only transport/protect troops and do not attack naval vessels).

Given how early it is in the game, I expect at most the following (from each AI, but Mao much earlier): 1 galley to attack any galleys we have and to pillage our nets (note that our fish are safe because they're in the ocean!) and 2 galleys filled with 2 or 4 units which serves as the main assault force.

One option we have is to get sailing quickly and pump out galleys. I would feel safe expecting to lose 2 galleys to each AI galley. That means we'd need 6 galleys. At 75H each, that's a *lot* of production, more than the pyramids. And if they get through... We'd better have blue dot founded!

To be honest, I really don't expect the AI to come at us before 500BC or so. I seriously weighted the deck in its favor when I did my test game by giving them every strategic resource right in their capital.

My current suggestion is: blue-dot with a worker escort asap. Get horses and bronze online in no small part because that will let us know (via the trade screen) which one of those the AIs have. Also prioritize sailing because we need the resources in the capital long before we'll have roads built, and we can send out a couple galleys to see where the AI is building up its forces. When it has two galleys sitting in a city, at least one of which is full of troops, then an attack could be imminent.

Lots to discuss, I guess.

 

[LKendter]

My current suggestion is: blue-dot with a worker escort asap. Get horses and bronze online in no small part because that will let us know (via the trade screen) which one of those the AIs have.

Where do you propose to squeeze even another worker from? Build 3 total, or steal St. Pete's worker build lower the odds of the pyramids. Or does St. Pete build 2 workers?

 

[Compromise]

Where do you propose to squeeze even another worker from? Build 3 total, or steal St. Pete's worker build lower the odds of the pyramids. Or does St. Pete build 2 workers?

I propose using our first worker to help improve Moscow and get the pyramids. The second worker (now under construction in St Pete) can maybe help with the pyramids (if there's time after it finishes) or just build some roads until the settler can finish in Moscow. Just a suggestion.

 

[Compromise]

In case anyone is interested, I'm attaching a couple saves from that test game I ran. One is the 4000BC save. The other is after I added some land to make it easier for the AI to get to me (after dropping a couple warriors to make contact, then deleting the warriors.

 

[Iainuki]

My computer (not the one I play Civ on, the one I actually use) is broken right now, so I can't really post much.

To counter my previous post, I think we should not panic. Our current plan is fine for now, we should be able to do the things we need to before the computer attacks. We shouldn't give up on the Pyramids, we should use both workers to chop them, we should settle blue dot, and then deal with our position. I have lots more thoughts (and justifications for what I'm saying here) but they'll have to wait until I can get my other computer working again.

 

[Atlas]

Well they are definately coming, but may be going after someone we have not met yet. As yet there is not much to worry about besides pillaging since neither Mao or Alex have horses or copper yet. I think we should just chop the Pyramids pronto. Let the worker finish the pasture then hustle him back to Moscow and get to work felling trees. I agree with Compromise, they are probably coming for us, but not for a bit still though.

 

[grangerm]

I don't like the idea of archery next (or ever really). I don't think that we are in any danger of being wiped out yet. The AIs are across an ocean from us. If they were right next to us, I'd feel much less safe. I think our best defense is to settle blue dot and get our strategic resources connected in case we need to do an emergency defensive whip.

It would be very nice to know about which resources the AI have. I wonder if there is a way to tell what they have, but are unwilling to trade in the domestic adviser?

Have we discussed enough that I should play my turns tonight? BTW, how many turns are we playing this turnset?

 

[Compromise]

Does anyone know what the AI handicap is on Monarch level? I suspect that by assuming that half of the civ's production (given in the demographics screen) is in the capital or whichever city is building an invasion force, then we could estimate how long it will take for the AI to build the forces it needs to send out to get us.

I think we should not panic.

This is sage advice.

Today, I find myself not running around like chicken little. Even in the scenario where I gave Alex and Mao everything, they still didn't declare war until 700BC. The attack date in our game should be a bit later.

they are probably coming for us, but not for a bit still though.

I agree. We should have time to concentrate on setting up our cities and getting the pyramids. (Unless Rome or someone is isolated on a tiny island, but then I guess it might be easy--if anywhere near us--to take the pyramids later.)

Because the horses have popped up near blue dot, I think it's sensible to have a settler there as soon as possible after we get mysticism. An obelisk is 45H, which we can get in 15 turns by working the horse site. Then another 15 turns (it could even make its own workboat in that time if we want to use St Pete for something else) before the fish come in range.

Since we have great food in Moscow, and since poprushing is much more efficient making a settler than the pyramids, is anyone intrigued by the possibility of growing to size 6 and doing a 3-pop whip to get the settler out? It's slower to get to blue dot, but probably generates more hammers for the settler+pyramid project.

Almost crossposted with grangerm

I don't like the idea of archery either. This is probably one of those gambits that separates 1st place from 3rd or 4th.

Unless the AI has more than one source of something, I think the only way to tell if they have it is to have it yourself and see if you can offer it in trade.

I think we're probably getting to the point of 10 turns per turnset. With multiple cities and AI interactions, the number of turns into the future we can confidently predict the results of our actions is shrinking.

My understanding of the goals for the next turnset are:

-Switch research immediately to Mysticism.
-Finish the Moscow warrior and then start the pyramids. (You can decide how to jockey the sentries around. Their present locations are great, especially with another on the spot W-SW of the rice to illuminate that other island's hill again.)
-Finish the worker in St Pete and send him to chop pyramids near moscow.
-Next build in St Pete is obelisk (I think), then workboat (I think).
-In Moscow, we'll want to switch to (and whip) a settler at either size 4 (which will require some time building the settler) or size 6 (which requires 1 turn into the settler to avoid the 0H penalty). That decision point may be a natural stopping point for the next turnset since it may be at 9 or 10 turns.
-I'm not sure about pottery vs. sailing for the tech after mysticism. That might be another natural stopping point.

 

[Compromise]

One other thought about the upcoming wars: As much as I dislike the prospect of a two-front war, it would be nice if the wars ended with us in possession of both Beijing and Athens, wherever they are. Since we will be the declaree, there will be no diplomatic penalty after peace is declared. Early control of the AI capitals will probably marginalize them for the rest of the game, strengthen us, and have many turns for "motherland" and culture woes to abate.

 

[grangerm]

OK, last chance to comment. I'll be playing in about an hour and I'll post a report later tonight.

 

[Iainuki]

Because the horses have popped up near blue dot, I think it's sensible to have a settler there as soon as possible after we get mysticism. An obelisk is 45H, which we can get in 15 turns by working the horse site. Then another 15 turns (it could even make its own workboat in that time if we want to use St Pete for something else) before the fish come in range.

Here are my tentative build orders for St. Pete and blue dot:

St. Pete: finish worker (goes to Moscow to chop), build something while growing (a warrior?), whip an obelisk, build something while growing, whip a workboat, start a settler(?).

Blue dot: grow while working obelisk, whip obelisk, build worker.

Ideally, the fish would be hooked up by the time blue dot is founded, so it could use them to speed up getting itself working. I think we really built one too many warriors from Moscow, since St. Pete is going to have to build something while growing, and it doesn't really have anything else to do.

Since we have great food in Moscow, and since poprushing is much more efficient making a settler than the pyramids, is anyone intrigued by the possibility of growing to size 6 and doing a 3-pop whip to get the settler out? It's slower to get to blue dot, but probably generates more hammers for the settler+pyramid project.

If we want to get maximum productivity out of growing Moscow to size 6, we'll want to mine/chop two hills so it can work them while at sizes 4 and 5. Do we have enough time for this?

My understanding of the goals for the next turnset are:

-Switch research immediately to Mysticism.
-Finish the Moscow warrior and then start the pyramids. (You can decide how to jockey the sentries around. Their present locations are great, especially with another on the spot W-SW of the rice to illuminate that other island's hill again.)
-Finish the worker in St Pete and send him to chop pyramids near moscow.
-Next build in St Pete is obelisk (I think), then workboat (I think).
-In Moscow, we'll want to switch to (and whip) a settler at either size 4 (which will require some time building the settler) or size 6 (which requires 1 turn into the settler to avoid the 0H penalty). That decision point may be a natural stopping point for the next turnset since it may be at 9 or 10 turns.
-I'm not sure about pottery vs. sailing for the tech after mysticism. That might be another natural stopping point.

My suggested build orders are above. I think we should go Mysticism, then discuss. There are a number of important techs here.

 

[grangerm]

1930BC - 0)
Swapped to Mysticism, research to 100%
Moved warrior out of St. Pete cause we don't need MP there yet.
Swapped Moscow to Pyramids cause we've got all the fogbusters we need
Alex and Mao are both cautious to us with the +1 peace bonus.

1900 BC - 1)
Nada
IBT) Barb warrior moves out of the fog towards our scout. Good thing I moved St. Pete's MP that way.

1870 BC - 2)
Move scout back
MM Moscow to get it to grow this turn
IBT) Barb warrior moves SW

1840 BC - 3)
Decide to grow Moscow so assign 4th citizen to FP

1810 BC - 4)
Nada
IBT) Barb impales himself on our warrior. Warrior down to 0.3 health.

1780 BC - 5)
Pasture done, worker due in 8.

1750 BC - 6)
Moved worker back towards Moscow

1720 BC - 7)
All of the fog is busted now. We can see purple borders on the island to the west. Can't make contact though.

1690 BC - 8)
Moscow borders expand.
Mysticism in, set research tentatively to Pottery.
Worker starts mining plains hill.

------------------

Save

I stopped at 8 turns cause we've got a couple decisions to make.

First off, we have our next tech to choose.

Secondly, we need to decide whether to whip the settler at size 4 or 6. Moscow will grow to 5 next turn if nothing is changed. Or we can swap to the settler and whip in a couple turns. I prefer the 3-pop whip, so that we can save all our produced hammers for the Pyramids.

Third, there's a warrior that still has 9 hammers in it in Moscow if we want to complete it. I thought we had enough units to fogbust, so I swapped to the Pyramids immediately. It will start to decay in the next couple of turns.

I had a couple more thoughts too:

If we are really afraid of Mao and Alex (which I didn't get that feeling BTW), we can gift them fish and clams. It's not like we're using them, maybe that would help relations...

I think we really built one too many warriors from Moscow, since St. Pete is going to have to build something while growing, and it doesn't really have anything else to do.

I think St. Pete should build a second WB that can be used for scouting in the short term and snaring some seafood in the long term.

IMO, seeing the other civ filling up the continent to the west puts more emphasis on settling blue dot before we finish the Pyramids.

 

[Kodii]

Since we are unsure about whether or not Alex or Mao will attack, I think we should go Sailing first, then Pottery after that. If the purple border is Julius, then thats another rather aggressive AI beside us.

I say, start the settler at size 6 rather than 4.

EDIT: I would prefer that the worker started by chopping the grassland forests along the river, then chop the forests on the hills BEFORE mining it. That way we can get the shields (er.. hammers) in earlier.

 

[LKendter]

I think St. Pete should build a second WB that can be used for scouting in the short term and snaring some seafood in the long term.

I would like a worker for blue dot long before the second workboat. We want the copper hooked up ASAP, and I feel we need two workers chopping to snag the Pyramids. Blue dot will need an obelisk first to have any chance to be useful.


Thoughts for the next round:

1) Mao and Alexander aren't a threat yet. Neither can offer us copper. I would really be surprised if there are up to iron working already. We have valid trade routes, as we can ship them goods.

2) Now that we see more borders, I vote for the next settler ASAP. Hit size 5, and swap to settler. Whip ASAP. The smaller the pop, the faster we can grow back. We should get back to size 5 quickly to finish the Pyramids. In the mean time, we will have two workers chopping toward the Pyramids. It shouldn't take 11 turns to build the settler.

3) Even with closed borders China and Greece can get to our island. That won't be the answer to protect our city sites.

4) For the next tech, I have a request for the number crunchers. Do with hit size 5 faster with or without the granary? The only way pottery makes sense would be granary in 10 turns speeding our target of size 5. If it doesn't pay, then we may as well get sailing out of the road. We will need that to start producing troops. I don't want iron working until after sailing, as it could take to long to connect copper city to the network.