SGOTM 12 - One Short Straw

bcool · Nov 30, 2010

One more issue here: Mao was willing to talk after 5t last time. His standard wait time is 8t, then it's shortened as I described in an earlier post. The sooner we DoW, the sooner we'll be able to talk. Do we want to DoW this turn, just in case? It's possible we'll want to cease fire immediately at some point, to preserve our capture.

The issue with that is happiness. I suppose it isn't too bad right now (is it only +1 :mad:

in GP Farm?), so a DoW this turn wouldn't hurt too much, but I think we should do everything we can to mitigate those happiness issues.

-run 5 artists in siberia this turn to pop borders next turn, redirect two? workers to finish the wine fort next turn.
-settle on spices as LC suggested to get 2 crab and spices in 2 turns (needs road perhaps to connect to trade network?)
-commit enough workers to get dyes going by T+2 from this turn.
-end the war with DeG asap

I think we want to grow Rheims this turn (so we can draft from there when the culture allows).

babybluepants · Nov 30, 2010

We're at 2 :mad:

in GP Farm already and losing a citizen to it.

LowtherCastle said:
One more issue here: Mao was willing to talk after 5t last time. His standard wait time is 8t, then it's shortened as I described in an earlier post. The sooner we DoW, the sooner we'll be able to talk. Do we want to DoW this turn, just in case? It's possible we'll want to cease fire immediately at some point, to preserve our capture.

Where did you post that? Isn't this based on war success? I'd expect it as soon as we capture Beijing.

bcool · Dec 1, 2010

found this in the archive, posted by LC

When is the AI willing to talk?

I believe I've got this figured out now, and it's just a bit tricky. The formula for when he'll talk if we DoW him is:

RefuseDuration = RefuseToTalk * {20+(80*AIWarSuccess*2/HumanSuccess}/100

If he Dows us, the above equation is doubled.

Here's the code: int iRefuseDuration = (GC.getLeaderHeadInfo(getPersonalityType()).getRef useToTalkWarThreshold() * ((GET_TEAM(getTeam()).AI_isChosenWar(GET_PLAYER(eP layer).getTeam())) ? 2 : 1));

int iOurSuccess = 1 + GET_TEAM(getTeam()).AI_getWarSuccess(GET_PLAYER(eP layer).getTeam());
int iTheirSuccess = 1 + GET_TEAM(GET_PLAYER(ePlayer).getTeam()).AI_getWarS uccess(getTeam());
if (iTheirSuccess > iOurSuccess * 2)
{
iRefuseDuration *= 20 + ((80 * iOurSuccess * 2) / iTheirSuccess);
iRefuseDuration /= 100;
}
The tricky part is calculating what I called AIWarSuccess and HumanWarSuccess. (iOurSuccess for AI and iTheirSuccess for human in the code.) The formula for these two, respectively, is:

1 + WarSuccess

And there's also this:WAR_SUCCESS_DEFENDING: 3
WAR_SUCCESS_ATTACKING: 4
WAR_SUCCESS_UNIT_CAPTURING: 1 (Workers/Settlers)
WAR_SUCCESS_CITY_CAPTURING: 10but, guess what? It turns out that when you capture only one worker or settler on a tile, it counts as WAR_SUCCESS_ATTACKING: 4. If you happen to capture 2 workers together then it's WAR_SUCCESS_ATTACKING: 4 + WAR_SUCCESS_UNIT_CAPTURING: 1 = 5.

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

Okay, so that means for de Gaulle, with RefuseToTalk = 5t, if we lose no battle:
If we capture 1 worker, he'll talk in 2t.
If we capture 2 workers, each on a different tile, he'll talk the next turn!
If we capture 1 wkr and 1 unit, each on a different tile, he'll talk next turn!No matter what happens, he'll talk after 5 turns, as long as we DoW him.

If he DoWs us, the wait time doubles.

babybluepants · Dec 1, 2010

Ok, so if we were to, say, lose 5 trebs and kill 8 LB capturing Beijing, it would still be 7t. With 4 losses, it's 6t. Is that right? Or, is war success cumulative, including the previous war?
The way to reduce it is by blasting them with no losses - that's easily confirmed by my experience.

bcool · Dec 1, 2010

I assume the treb that are successful count as well?, so if we lost 5 treb but won with 3 and killed 8 LB that would be 44 from successful attacking + 10 from capturing the city

He would have 15 successes

So 20 + 80 * 15*2/54 = 5.16 or 6 turns as well

war success is reset.

LowtherCastle · Dec 1, 2010

Played to T175. Captured Beijing. Lost 2 trebs, 1 cat, and 1 mace at 99.4% odds (rigged RNG :rolleyes:

), but I can't complain since our first two trebs survived while redlining his lbs. :cool:

I have an announcement: Beijing is sweeeeeeeet. :king:

Spoiler :

Oh, and did I mention Beijing also kept its granary? The building list was too long to show it too. :cry:

T172
Cease Fire Mao
Trade Mao Thology for 140g
Trade Gandhi PP for Feudalism+130g+maps
Decide not to DoW Mao yet, cuz I don't want him occupying any of Shanghai-NW/W/SW.
Revolt to Nationhood
Upgrade two swords for 220g.

IT
Roosy makes a GSpy in Washington
Galley City builds a granary

T173
Cease fire de Gaulle
Forgot to MM last turn for Moscow treb(1t), but no problem because the trebs were ahead a turn in the galley queue
Wines connected for +1

Beijing has 6lbs + 1HA

IT
Mao and Gandhi now have Optics and they'll both trade it
Nanjing now has 3lbs, 1 archer, 1 HA
The Beijing HA is now south of Shanghai so I can't empty Shangahi or work any tiles nearby. Scramble to protect Orleans and Rheims from a sneak attack.

T174
Dyes connected
Settle The Crabbes
Spices connected
Peel pikeman off the SoD and guard workers with it.

IT
Mao's HA goes northwest of Shanghai, futilely trying to return to Beijing :mwaha:

T175
Settle Glace de Gaulle and gift it to dG. We now have -4 for You have traded with our WE with Gandhi, but we're still okay.
GP Farm spawns a GS.
First two trebs red-line Beijing lbs and survive. Capture Beijing, remove all posters and sculptures of Chairman Mao and replace them with pictures of a smartly uniformed Captain babybluepants.

Beijing seems to have defied all odds and kept every possible building.

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

Now what. guys? Can we actually capture two more cities in 3t?

Save uploaded.

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

Here is your Session Turn Log from 1120 AD to 1150 AD:

Spoiler :

bcool · Dec 1, 2010

whoot!

Looking at it I don't see how we could capture 2 more cities in a 4 turn window. We might be able to capture 1 and hope Rosey captures 2... seems highly optimistic.

Tough call on what to do next... Options I see

1) hope for the stars to align and go for Nanjing in 4 turns
2) hope for the stars and planets to align and go for Guangzhou in 4 turns
3) Take Guangzhou in a sane manner(more than 4 turns) and take Nanjing with reinforcements-- (draft a musket in Rheims maybe produce a macemen or 2 in Moscow), switch to communism with GA from Taj (maybe we can accelerate that 1 more turn?)
edit: Depending on the type of Great People produced in Moscow and Pigs we could consider triggering a 2nd GA with them. But I think we do not if we produce a Great Spy.

babybluepants · Dec 1, 2010

LC

LowtherCastle said:
Capture Beijing, remove all posters and sculptures of Chairman Mao and replace them with pictures of a smartly uniformed Captain babybluepants.

Excellent! I was wondering if you'd remember to do that. Such a complex TS and all...

bcool · Dec 1, 2010

We can finish liberalism-SM-biology in 2 turns (can't we?) going 100% science?

babybluepants · Dec 1, 2010

^^ Yeah, I think we should just finish it. We need both as pre-req for Communism anyway.

Churchill's got a stack going north now. I think we should commit the two axes in France, and maybe another unit or two to shadowing him in Lyons. Maybe we can steal this one.

Did you try any begging, LC?

LowtherCastle · Dec 1, 2010

bcool said:
We can finish liberalism-SM-biology in 2 turns (can't we?) going 100% science?

Well, I screwed up the timing of the monasteries a bit, so we can't finish SM this turn. If we want to put too many beakers into SciMeth, then we could bulb it next turn and complete LIb that turn too. Otherwise, in two turns.

babybluepants said:
Did you try any begging, LC?

Nope. Didn't want the negmod with Gandhi because I knew we were going to get -4 for gifting two cities to dG.

Prefer to keep our beg option with Churchill, in case he DoPs dG and then gets his hands full of us. Churchill could roll our cities if he wanted too right now. We can't take that chance. But we could sell him Paper.

bcool · Dec 1, 2010

Well, I screwed up the timing of the monasteries a bit, so we can't finish SM this turn. If we want to put too many beakers into SciMeth, then we could bulb it next turn and complete LIb that turn too. Otherwise, in two turns.

With a little finesse putting "too many beakers into SciMeth" might amount to only <50 beakers lost. I would assume that would be worth getting Biology 1 turn earlier.

By the way that is a great idea

bcool · Dec 1, 2010

We are going to take a relations hit with Rosey for gifting Mao a city aren't we? I was thinking for a moment we could get him to friendly.

babybluepants · Dec 1, 2010

We're not getting anyone to friendly this game.

I think we can waste some beakers here. Let's just go with what LC suggested and take Bio next turn.

Can we get Nanjing and Hangzhou on schedule? Did you check the garrison in Hangzhou?

babybluepants · Dec 1, 2010

@LC,
are you playing through a few more turns? I think you should. Hand off last turn of GA?

Off-Topic:
Plastic Ducks haven't posted in a couple of days, and their last few play sessions were oddly short and quick in succession... What do you guys think of that? Our graphs just look too similar for them to crush us with some totally different approach, right?

LowtherCastle · Dec 1, 2010

bcool said:
With a little finesse putting "too many beakers into SciMeth" might amount to only <50 beakers lost. I would assume that would be worth getting Biology 1 turn earlier.

Okay, someone check my math and population count, please.

SciMeth: 3744 :science:

Already: -1454 :science:

subtotal: 2290
GS bulb: -1911 (pop137= 1500 + (137*3) = 1500 + 411)
subtotal: 379
Prereq20: /1.2
Total: 316 base :science:

still needed

So I can put the slider at 326 base :science:

for a margin of error, plus we gain pop3 next turn which is +9 final :science:

.

Does that check out?

LowtherCastle · Dec 1, 2010

babybluepants said:
Can we get Nanjing and Hangzhou on schedule? Did you check the garrison in Hangzhou?

I have not seen the Hangzhou garrison yet. Needed the spy full-time around Shanghai.

babybluepants said:
@LC,
are you playing through a few more turns? I think you should. Hand off last turn of GA?

Off-Topic:
Plastic Ducks haven't posted in a couple of days, and their last few play sessions were oddly short and quick in succession... What do you guys think of that? Our graphs just look too similar for them to crush us with some totally different approach, right?

Yes, I think I should play up to the last turn of the GA, because that's gives us a clean starting point for further planning.

Speculation at this point tends to be wrong, but my thought today was that they went for a bigger SoD before attacking and then they attacked, which is where their Power Graph jags to the right, similar to what ours has done twice, but their's had a bigger jag so they probably hit a bunch of cities in quick succession, maybe even two fronts.

I've been thinking for a while that they were ahead of us. Now with our power and score graph following theirs, I'm not so sure. THe big question mark is where their research is at. When did they go to castes, how fast did they kick their research forward with that, what's their tech path, etc. Did they bulb GSes? I'm pretty sure I read somewhere that they don't like bulbing too much, but rather save GPs for GAs.

THen there's also the question of whether they got into tech stealing as early as we did. We may have had an advantage "gifting" Gandhi Galley City before he got past 3 cities. Once you're past that, then you have trouble gifting to anyone but dG, who's close enough to accept it within 9.5 tiles.

Then there's the question of how much worker stealing did they do, and whether they got Hagia (I doubt it, since they would have had to steal it).

There's also the question of whether they got the stone before dG, because that would have given them a very early Representation+castes and that would have powered their reserach past ours pretty rapidly.

Too many unknowns, but I don't think they're stomping us. I hope not, anyway.

bcool · Dec 1, 2010

Off-Topic:
Plastic Ducks haven't posted in a couple of days, and their last few play sessions were oddly short and quick in succession... What do you guys think of that? Our graphs just look too similar for them to crush us with some totally different approach, right?

IF they somehow got the Great Lighthouse that could have made a significant difference. I think we are really going to start flying with the tech steals, corporations, etc. But to launch pre-1600 seems quite amazing.

So I can put the slider at 326 base for a margin of error, plus we gain pop3 next turn which is +9 final .

Does that check out?

That looks right to me.
The reason you divided by 1.2 is because the displayed value of beakers doesn't include the 20% discount for having 1 prereq for SciMeth.

bcool · Dec 1, 2010

I really hope we can hit Churchill before he has Feudalism for too many turns. Do we want to think about just crushing Churchill via Nanjing?

babybluepants · Dec 1, 2010

bcool said:
IF they somehow got the Great Lighthouse that could have made a significant difference. I think we are really going to start flying with the tech steals, corporations, etc. But to launch pre-1600 seems quite amazing.

That looks right to me.
The reason you divided by 1.2 is because the displayed value of beakers doesn't include the 20% discount for having 1 prereq for SciMeth.

Well, I wouldn't be too surprised if they got both GLH and early stone for Pyramids. We were quite slow to expand initially, so stone is def. get-able.
Edit: except I remember I studied their early graphs very carefully, and concluded no GLH...

No, the 20% isn't displayed.

SGOTM 12 - One Short Straw

Deity

Deity

Deity

Deity

Deity

Deity

Attachments

Deity

Deity

Deity

Deity

Deity

Deity

Deity

Deity

Deity

Deity

Deity

Deity

Deity

Deity

Similar threads