SGOTM 12 - One Short Straw

[babybluepants]

Some stuff from the Deity Team Game SG (not sure how relevant):

The max war plan can be 20 turns for normal.

else if (AI_getWarPlanStateCounter((TeamTypes)iI) > ((20 * iTimeModifier) / 100))
	{
		AI_setWarPlan(((TeamTypes)iI), NO_WARPLAN);
	}

iTimeModifier is 100 for normal [unless the AI has over 200(!) more power, it was supposed to be 2 but there is a bug in the code]

The AI doesn't wait for catapults, just enough troops for attack

if (bTargets)
{
	if ((countNumAIUnitsByArea(pArea, UNITAI_ATTACK) + countNumAIUnitsByArea(pArea, UNITAI_ATTACK_CITY) + countNumAIUnitsByArea(pArea, UNITAI_PILLAGE) + countNumAIUnitsByArea(pArea, UNITAI_ATTACK_AIR)) > (((AI_countMilitaryWeight(pArea) * iOffensiveThreshold) / 100) + 1))
	{
		return AREAAI_OFFENSIVE;
	}
}

Once the AREAAI_OFFENSIVE is returned the AI will start marching.
UNITAI_XXX are types that the AI can possible use to attack city.

The number of the units must be more than number of (cities+population) divided by 4 for total war. The AI totally doesn't care if will attack w/ warriors or catapults (or modern armor) once it has passed the initial power threshold. And the units fall in the UNITAI_XXX roles above, catapults promoted to barrage are actually not counted since they are most likely UNITAI_COLLATERAL.

War is a team decision so it calculates the units/cities/population of all the members.
-----
I'd be glad we can fit 5 archers/1axe/1spear and walls on the hill city by the time attacking. I'd dedicate delhi/bombay for units and perhaps start growing cottages in the copper city.

1st. AI is sort of honorable and never starts a total war in less than 10 turns (limited are started on the 6th turn, but that's not limited war)

2nd. (the reason) Monty and Ragnar (he had low power) dont have enough units w/ the roles described above, that's the only reason the AI doesn't start a total war (besides not path which cancel the war altogether)
Units can be 'counter' or 'reserve' (UNITAI_RESERVE are the most suicidal ones) and they are not counted (but participate in the stacks). Defense units (UNITAI_CITY_DEFENSE) are not considered either. AIs can have quite a bit of them but they are not counted towards the decision, for instance spears are likely to be counter type and they are not counted. Axes are often counter (or even reserve, esp. for unit spammers like Monty). Chariots also get often into counter division. Suicide (barrage) catapults don't get considered either.
So assuming they together have 11cities and probably like 40-45 population, let's say: 55/4 = 13, 13+1=14, they need 15units (together) designated for attack which they don't have yet.
---
Bottom line, many (or most) of the units produced during peace time are simply not "city attack", so they have to build the offense.

The Rook, try counting mostly units w/ CR promos and archers w/ combat. Also HAs w/ combat. I think they are the supposed city attackers. The rest are support which the AI doesn't consider viable preparation.
If you see some CR promoted units wandering (not stacked) it means that Monty/Ragnar are morons (which we know anyways) and will spend some time chasing barb. cities (which is good for us).

 

[Soirana]

Save uploaded. I gave him 1gpt, canceled pigs and canceled banana in succession - didn't work. Maybe next turn? Though I'm pretty sure Gandhi had it immediately, earlier in the set when I canceled his deal.

doesn't work if AI has negative gpt per turn to start with.

I am proud of fact this team spammed over 120 pages

 

[mdy]

Maybe we should sign a cease fire agreement with Mao and bypass Nanjing for now to focus on Churchill?

We could also draft an extra musket in Rheims/Bananas this turn to give us more troops available to pick off his stack. Even unpromoted muskets have good odds against axes in the open field.

 

[babybluepants]

The crash is np, btw. Had a short PM exchange with Alan.

 

[babybluepants]

Maybe we should sign a cease fire agreement with Mao and bypass Nanjing for now to focus on Churchill?

We could also draft an extra musket in Rheims/Bananas this turn to give us more troops available to pick off his stack. Even unpromoted muskets have good odds against axes in the open field.

Can we even talk to Mao? I forget now... Yeah, I'd CF if possible, if only for WW reasons. We'll see how the situation with Churchill develops when we're actually ready to attack Nanjing.

Drafting in Rheims and Bananas this turn sounds right. Rheims should get Walls right away, too.

 

[Mitchum]

How do you ask a civilization to target a specific city?

Talk to Churchill. Select "Why don't we discuss something else..." then select "Why don't you attack..." A list of possible cities will be available for selection.

@bbp, I noticed in the last upload a spy snoozing in Rheims. He's still there. Was that intentional?

So is the plan now to let Roosy capture Hangzhou so that we only need to settle Kamchatka 2S and Trojan City for the city-gift chain?

I agree that we should move our entire stack southward toward both Nanjing and Churchill. We should be able to deal with his stack when the time comes. We were going to have to do so anyway...

I'm not exactly sure how we're using our galleys, but I usually don't have them doubled and tripled up like this. Is this for protection from enemy boats or are we planning to use them to exend our galley chain further toward Roosy soon?

 

[LowtherCastle]

@LC: if u had joined OSS a few games earlier (well you wouldn't have won as many medals but I digress), you wouldn't fear this situation so badly ... once you have lost a city to monarch barbies in a SGOTM, you can accept everything. Very good mental training.

I'm well aware of the OSS history. Actually, you may not know that Murky Waters also lost a city to the barbs in SG10. Our stone city, no less...but after we had built the Kremlin, not just before...

I have no problem accepting everything, but I'm not sure you and I have the same definition of game over...

 

[babybluepants]

I'm not sure you and I have the same definition of game over...

No, you don't.

 

[Mitchum]

Looking closer at the save, I have a few more questions:

Moscow has access to more hammers now that St. Nick is in revolt.

What is the work boat in Rheims going to be used for?

Do we need a harbor in Bananas? Same think in Bahamas, although it is almost done already...

Did Sid assign that engineer in Bermuda or did you do it? Is seems better to work a coast tile.

Kamchatka could use an MP...

The workshop Orleans 1S at can be done in one turn if needed...

Does Shanghai need a theater?

Have we bought Ghandi's and/or Roosy's map lately?​

 

[babybluepants]

What is the work boat in Rheims going to be used for?

Mentioned it before playing - it's for spying on London and York.

Do we need a harbor in Bananas? Same think in Bahamas, although it is almost done already...

Perhaps not, but we can certainly use some extra commerce. The Bananas one is more of a placeholder, maybe.

Did Sid assign that engineer in Bermuda or did you do it? Is seems better to work a coast tile.

Agree. I was playing around with tiles there a bit - don't remember what it was left at, but I'd do a sweep before turn end. Bermuda's having some issues soon. I guess we can work coast instead of seafood later, if needed.

Kamchatka could use an MP...

Seemed like necessary evil. I finished the CH, since it was close to finished. Then I planned spy-settler. The chariot from Cuba was the first available. If we're too slow in ep to steal from Roos, maybe I'll delay the steal city for a chariot.

Does Shanghai need a theater?

It's a dirt cheap build, gives an extra , and 3cpt - I wanna work on popping borders twice for that extra fish. Mentioned in one version of PPP - no one commented.

Have we bought Ghandi's and/or Roosy's map lately?

Yes, this turn. Nada.

@bbp, I noticed in the last upload a spy snoozing in Rheims. He's still there. Was that intentional?

My bad. Don't even remember that spy...

I'm not exactly sure how we're using our galleys

It's a mess, is how.

 

[Mitchum]

@bbp

Thanks for the answers. There has been so much happening in our game and RL that I missed some of the points from your PPP. This is a complicated turnset with a lot to keep track of. I just hope that you take care of Churchill's stack before I take the helm...

 

[mdy]

Perhaps not, but we can certainly use some extra commerce. The Bananas one is more of a placeholder, maybe.

We could use the extra commerce, but right now an intelligence agency would be much more useful for us, the same applies in Shanghai, and anywhere else where we don't have an urgent need for anything else, and we can build it relatively quickly.

 

[bcool]

Once we steal the techs from Rosey we won't need too much more espionage other than the espionage we passively make from the stolen intel agencies and spy specialists.

Will we benefit from building intelligence agencies before we want to swing the slider to 100% espionage after teching Steel?

I think building wealth would be better than an intel agency. The wealth right now gets us Steel that much faster. Building wealth could also be directly converted into espionage (by allowing us to run the slider at 100%).

Basically we need the espionage before the extra espionage from an intel agency would really pay off.

 

[babybluepants]

^^ Fair enough. IA in Bahamas would be great. There was some talk about extra settlers from there after Taj, but now I don't really see the need to rush those so I went with a relatively cheap Harbor.

Edit: xpost w bcool - that was what I gathered from the PPP discussion. It wasn't clear that everyone wants to invest in IA.

 

[LowtherCastle]

As long as we're talking city MM/builds, here's my wish list:
Fish: work a merchant (+1 more at popgrowth) instead of the wkshp for 31hpt=> 149/150h in the market in 3t.
Siberia: MM for growth (+4 in 4t) and no more drafting here. Finish the Torjan Horse settler asap!!! The more culture we put into it the better.
Bananas: lighthouse next (it should give up the shared farm to GP Farm)
Bcool Island: work the spices tile, scrub the river tiles with 1 wkr each then wkshp asap. Forget steenkin' mines.
The Crabbes: chop the granary asap
Beijing: WORK 3 SPIES!!! ALWAYS, TOMORROW, INFINITY, FOREVER, YESTERDAY. - JOE MOMMA.
Guangzhou/Chengdu:
* Send two wkrs immediately to scrub CHengdu-1S (hopefully ROosy's culture doesn't spread next turn and they can finish--we hope Roosy will farm this tile)
* Pillage all cottages in this area for gold!!! Maybe use el Cid to pillage then heal, pillage, heal...or something.

We should pillage any cottages we don't want in the future. This is good gold for us.

Watch out for Mao's catapult near Guangzhou and his lb south of Shanghai!!!

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

I agree with bcool on IAs. Basically we only need them where we 1) can build them really fast and 2) plan to work lots of spies. That means Bermuda, since we're building the Kremlin there. If the city doesn't have the food to work the spies, there's no point in investing all those hammers in an IA, imo. BUt we should hire max spies everywhere we can except Pigs and GP Farm.

 

[LowtherCastle]

I've been studying klarius' posts on how the AI conducts wars and attacks cities from SG9 (deity OCC). Here are the ones I found. It's better if you follow the link, to get the context of his responses. Also, bear in mind that in SG9 we were looking at it from the opposite point of view--we were trying to get Kubla Khan to attack, so some of what klarius says is sort of the opposite of what we want, because we're trying to prevent an attack...:

Spoiler :

* ATTACK or ATTACK_CITY units will lead to a target city. COUNTER units stack with attack or attack city units.
* What promotions tell us about the types of units.
* Types of units, their functions and characteristics.
* Telling the AI to target a city and how RNG affects whether the AI targets a city.
* More about AI city targets and attack units
* Types of AREA stategies. Distance types of units will attack a city from.
* When AI SoDs will leave when on the defensive.
* Ways to distract an SoD from targeting a city
* When cats bombard, when they attack ("vanilla formula")

My conclusion is that the best chance we have of minimizing our danger is to:
1. Threaten Marseilles by having a unit within 2 tiles (and maybe Tours too), and
2. Perhaps be ready, if possible, to launch a counter-offensive on Marseilles.

The reason is that when threatened, if a city doesn't have enough defenders, it will use a SoD leader, if available, thus freezing that SoD, however big the SoD happens to be. What we see may be one or multiple SoDs.

I don't actually know that the coutner-offensive would draw an SoD back, but we could give it a shot. Furthermore, I don't know how many defenders Marseilles will want to have if we threaten or whether that depends on how many units we threaten with.

 

[babybluepants]

Do we wanna try working towards a 10t DoP?
If not, do we wanna DOW him instead?

 

[LowtherCastle]

More City MM:

Bermuda: MM for max growth. It will also gain soon. And don't draft here. :crazyeyes:
Orleans: Walls
Rheims: Walls

 

[LowtherCastle]

Do we wanna try working towards a 10t DoP?
If not, do we wanna DOW him instead?

How soon will we be ready to take out his SoD?

 

[RRRaskolnikov]

@LC: I see what you mean (I am maybe 10% of the competitor that you are )! I am really impressed anyway by what you brought to this team

Cheers (and I will spamm less, promis )