SGOTM 15 - Kakumeika

[mabraham]

I kind of like putting a very early second city 1E or 1NE of the Deer (provided we settle Delhi on Stone):

• can be founded in 3160 BC and finish Stonehenge + TGW until 2000 BC if we go Poly->Masonry->Ag->Hunting
• strong land grab in probably contested area
• immediately connected via river for (likely) instant net gain in commerce and Hindu spread
• depending on real map size, overall city maintenance might start with just 2
• as mentioned forested Deer is very strong tile to work unimproved
• Hunting is useful due to the likely nearby appearance of Fur
• One of the tiles E or NE might be a PH, not sure
• nearby hills can give important early hammers, our Fast Workers can quickly mine them
• can help grow capital cottages later should we aim for these

Imho the 3rd city should go for Marble to build Oracle, but I would put this city 1W of Marble as river quarry gives good commerce.

So I vote for using the Warrior to move 1S in order to check out the Marble spot but then scout the Deer area first.
Afterwards either explore further NE and return to guard Settler or in case the Deer spot is crap (peak cluster, close AI) move west along the Stone river to find a better city spot.

Yeah that all looks pretty strong to me. Well spotted. I'm rather expecting we're in a pocket of green land surrounded by low-food highlands.

 

[mabraham]

The settler first decision might hinge on what our goals are (what victory condition we plan for).

Are there victory conditions that do (or do not) suit settler first? Any VC can have worker stealing or an early rush on the front of it, and that is likely precluded by settler first. Early warrior scouting is good for everything, especially with so many unknowns and fears out there. That's compromised by settler first.

By going settler-worker-SH, and warrior-TGW and settling along the river near the deer, we can have a second warrior, a worker and both wonders complete by around T41-2. If we test when barb units would invade, we can probably send both warriors scouting once the second city settles.

Is that what we want? What else could we do with a fast second?

 

[Walter_Wolf]

Goal oriented PPP
The goal for this turnset is to settle the capital on stone, build two scouting warriors till size 2, and reveal as much land possible.
I can't run the game right now, so can't do the actual testing but this is the variant I propose to the team:

1. Settle on stone.
2. Build 2 warriors till size 2.
3. Research Polytheism.
4. Scout the NE territory with the initial warrior (will draw the path later).
5. Put a „Real game“ sign (using Alt+S) near the capital.

Stopping points:
Run into an animal with unoptimal route of escape,
Meet an AI,
Loose the warrior.

Every turn:
Take screenshots of the demographics screen,
Look for AI culture in sattelite view,
Save game,
Note (using Alt+S) forest/jungle growth, animals appearing etc.

 

[Walter_Wolf]

The stone site is as good as I expected it - no bonuses yet, but there's plenty of food for short term use, three(?) Ghills for hammers and getting a library to run scientists, and grasslands for cottages later.

I'm still keen to settle on stone.

Upon further reflection I think you are right. It might be a nice cottaged bureaucratic capital in the end, and the initial hammer boost is great.

cheers

 

[mabraham]

I would suggest turning on the option for animating friendly and enemy moves. We want to see from which direction AI scouts come.

 

[Walter_Wolf]

I would suggest turning on the option for animating friendly and enemy moves. We want to see from which direction AI scouts come.

No problem.

 

[bcool]

Right now I feel a settler first opening has more advantages than a 2 warrior worker opening.

I think the wisdom of stealing workers from a buddhist coalition of AI with extra military units/strength is limited. There is a very small window we might be able to do it without being locked in a war with one of the them and then declared on later by their friends.

I think the value of several early scouting warriors is less than taking maximum advantage of our early access to stone.

I'm not convinced stonehenge is a wonder we need, but an early great wall is valuable I think. I can be convinced otherwise, but I would rather have our first great person to be a guaranteed great spy.

In a buddhist world we might find it difficult to trade with AI for technologies, so espionage might be more valuable than we initially thought.

Those are my initial thoughts, we might want to delay Walter_wolf's turnset until we can either prove that the settler first opening really is better or if a 2 warrior opening is better with testing.

 

[bcool]

Are there victory conditions that do (or do not) suit settler first? Any VC can have worker stealing or an early rush on the front of it, and that is likely precluded by settler first. Early warrior scouting is good for everything, especially with so many unknowns and fears out there. That's compromised by settler first.

Early warrior scouting might reveal more about the terrain and get a bit earlier AI contacts, but what more can we learn about the challenges this sgotm will have?

I think we know that we will be facing a Buddhist coalition of AI.

What victory condition is going to the be the easiest given this starting situation?

The warriors might figure out we are isolated (but I highly doubt we are).
I think the opportunities for worker stealing are limited or unwise until we get the AI not to like each other quite as much. (If we flip an AI to Hinduism and then declare on it steal its workers, etc, then we are much less likely to get dogpiled).

If we piss off the buddhist world early on, then we will be forced in situation where most of the AI hate us and we give up on tech trading, etc. Or at least make those things significantly more difficult.

2 cities early on gives us a lot of early production that might translate into more than just 2 early wonders. We probably can catch up with 2 warrior/ worker opening.

We really need to see where a 2 warrior/worker opening gets up by T50 compared to a settler first.

Anyways scouting 1S with the warrior so we can start testing in earnest to really get a better idea of what is best.

 

[Walter_Wolf]

I could:
a) move the warrior 1S and quit;
b) move the warrior 1S, settle the capitol (on stone) to see the whole BFC and quit.

I could do it tonight, so we can test in the next few days to see what's best if you agree.
Do we need a 24 hours delay to pick a or b?

EDIT: do we actually need any more moves before the tests? I think we probably have enough info to make the decision on the 1st build / path we are going to take.

 

[DanF5771]

Here is a new save with updated terrain of the stone area for tests.

I agree about the reduced usefulness of Warriors early compared to a standard game.

The interesting info in the new demo screen is that the AIs GNP hasn't changed. This means it is very likely that the AIs had already chosen their techs in 4000 BC (otherwise there would be an increase in GNP due to modifiers being in effect). I'm not sure, but Neil may have programmed an AI to beeline Theology.

Worker stealing might give us the decisive -1 "Declared War on Friend" from many close AIs, that lets us drop from CAUTIOUS to ANNOYED -> Worst Enemy, which might otherwise be prevented by +1 "Years of Peace" from T60 on (the usual time we get Writing for open borders).

 

[Manco Capac]

I think the opportunities for worker stealing are limited or unwise until we get the AI not to like each other quite as much. (If we flip an AI to Hinduism and then declare on it steal its workers, etc, then we are much less likely to get dogpiled).

Worker stealing might give us the decisive -1 "Declared War on Friend" from many close AIs, that lets us drop from CAUTIOUS to ANNOYED -> Worst Enemy, which might otherwise be prevented by +1 "Years of Peace" from T60 on (the usual time we get Writing for open borders).

Although it saddens me, I agree how unwise in long term it would be to steal a worker from someone else, unless there is a natural early hatred AI before the religion sharing brings to CAUTIOUS or even PLEASED other AI's towards him/her.

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

Although this might not totally fight fit to the present situation, Vicawoo wrote a thread about a settler first isn't that good. I have few time right now, so I let that thread until time I verify its content's usefulness. First thing that might deviate that thread from our situation is we can't get a food worker tech right away to improve and grow.
Still, testing will bring the final decision.

 

[Manco Capac]

3. Research Polytheism.

Better uncheck any tech choice as accumulating beakers trick via pre-requisite tech (Mysticism) was proven to be always benefitting. But we have to keep in mind at the end of T5, we have to get Poly on line or we lose the game.
Most probably, we will settle on stone (or the lost turn won't ever be recovered), thus four turns before committing tech choice.

 

[Manco Capac]

I'm not convinced stonehenge is a wonder we need, but an early great wall is valuable I think. I can be convinced otherwise, but I would rather have our first great person to be a guaranteed great spy.

Agreed totally. We won't need its free monument bonus at all.
Better dump 60 into another worker.
The GProphet points might be useful to compete with the AI who have the Mahabodhi and getting the first GProphet in the world (doubting he would have the pre-requisite for Theology), but I doubt Theology bulb will happen. Anyways, we have at max 100 turns to know what to do.

Only the location in the maps is interesting, but not much on a highlands mapscript.

 

[mabraham]

OK I did a settler-first test game on DanF's latest save.

T14 Poly->Hindu
T18 settler complete, moved to 1E of deer, started worker
T20 settled, started warrior (NB I edited the terrain so that this was flat land) Maintenance was just over 2 per turn, so costs us three. So we might like to settle 1N of deer for lower maintenance, or 2E of deer for more land, given the capital's border pops very soon. Will depend on the terrain.
T22 I was the WE of several AI
T23 Done masonry
T24 border pop
T25 Start worker in second city
T28 Done capital worker (temporise on some mines, lots of cutesy worker micro available), start TGW
T31 Done Ag, start corn farm
T36 Done Hunting
T37 Done worker in side city, move to deer, start Pyramids
T40 Start camp
T41 camp done
T44 Done Wheel, done TGW, start settler
T50 (AI builds SH)
T53 Done settler, done pottery
T61 (AI builds Oracle)
T63 (AI has Alpha) Done AH
T68 GSpy pops
T69 Pyramids. Next capital border pop still 20ish turns away. Writing still 4 turns away.

Note that we could have had the Pyramids a few turns ago if we'd had BW for chops, but that would have delayed granaries and libraries, we can't run specialists yet anyway, and the empire is not exactly food-rich for whipping. If we want PH->Oracle with marble, then we'll have to do that after Wheel. It's hard to see us being able to Oracle CoL (for example).

In retrospect, delaying TGW in favour of something useful (settler for quicker marble access?) is probably safe.

I'll do a comparison with some warriors and worker first tomorrow when I have some more time.

 

[mabraham]

I made some more time today. We can do much better with judicious use of Stonehenge fail gold, and getting marble for the Oracle.

As above, then
T20 settle side city, start worker
T28 start Stonehenge.
T32 side city worker done, temporise on Gmine and Gfarm, start Pyramids
T35 move worker to deer
T36 Done hunting, start camp, start wheel
T39 At size 3 switch capital to settler with one Gmine available
T41 Switch capital to second Gmine, resume side city farm work
T42 start FP farm, finish wheel about now, start PH
T44 start side Gmine
T47 settler done, start TGW
T49 Settle somewhere 1W or 1NW of marble
T50 Hope an AI completes SH about now. Finish PH about now, start Writing. Workers unite to improve and hook up marble and new city to trade network, then they add a Gmine to each of the capital and the second city
T54 TGW completes, start Oracle.
T61 Finish oracle -> CoL. Start libraries as suitable
T69 Finish Pyramids

Note that the above has built no warrior at all. We would need to test how likely barb units are to invade before T54.

 

[bcool]

Wouldn't the tech path of poly, agriculture, masonry be better?

Then we can build a warrior and might get less fail gold from Stonehenge but we get the corn farm online much faster. I'll try to play that out but I don't have much time this morning.

Also we need to compare a test game where we build 2 warriors first then worker and then settler at some point to compare the settler first strategy with others.

Settling on the marble denies us the trade routes, but does give us 1 more hammer per turn to play with. Something else perhaps to test.

 

[bcool]

tech poly, agriculture, masonry, wheel, priesthood

T14 poly
T20 settler 1E of deer, starts warrior
T26 warrior
T28 worker in capital immediately starts farming corn, starts stonehenge
T31 warrior and masonry finishes, start wheel (mistake?)
T32 farm finished
T33 FP farm started (mistake I think)
T38 Delhi 3 pop starts settler (stops stonehenge)
T45 priesthood (skipped hunting another mistake I think)
T47 Greatwall finishes in deer city
T48 settler settles 1NW of the Marble
T51 worker built in capital (Oracle built by AI)

worker priorities were also poor. And tech wise perhaps we can skip wheel if we settle on the marble and build the oracle in that city.

I think the Oracle is in serious danger of being built by the AI if DanF's test game is an indication. Perhaps we want to settle directly on the Marble instead of the Deer site for faster access to Marble? The lack of trade routes will hurt us.

Sorry I definitely can improve on this. This is just my first attempt. I have to go, but I will do more testing this afternoon.

 

[DanF5771]

Thanks Tachy for the link and thanks guys for trying out my test game.

Unfortunately I'm very limited in time as well at the moment, so just a few remarks:

Why not settle 1NE of Deer? From fog gazing it's imho ascertained that the tile 1E of Deer is an unforested GHill with river = nice tile to work before BW, no?

The early AI Oracle date should be taken with a grain of salt, AI has quite a head start techwise as I wanted to create a challenging worstbad-case scenario (they have all 1st tier techs, team AI has Meditation too). And I have to say I underestimated the AI-Wonder-Lust under AggAI. Feel free to change that.

 

[Walter_Wolf]

I tried the 2 warriors, worker, G wall, SH plan. Scouted almost everything, finished the mentined wonders, and got 120/500 hammers into the mids. The 2nd settler is half way done in capital.

Researched Polytheism, Agriculture, Masonry, BW, Wheel. Pottery just started.

It's not a suggested path, just a test of where we get in 50 turns with a 2 warrior worker opening. Here is a screenshot:

Spoiler :

Check the minimap. I made the whole circle. If the real game map resembles this one I guess we don't have to scout this aggressively in order to find the AI and evaluate the diplo situation. They will find us very fast which makes the warrior opening less valuable. Worker stealing being not advisable makes this opening even less appealing. I'd say worker or settler 1st is the real question.

 

[mabraham]

Thanks Tachy for the link and thanks guys for trying out my test game.

Unfortunately I'm very limited in time as well at the moment, so just a few remarks:

Why not settle 1NE of Deer? From fog gazing it's imho ascertained that the tile 1E of Deer is an unforested GHill with river = nice tile to work before BW, no?

Sure. No strong reason, except that I hadn't downloaded the real save to gaze the fog.

The early AI Oracle date should be taken with a grain of salt, AI has quite a head start techwise as I wanted to create a challenging worstbad-case scenario (they have all 1st tier techs, team AI has Meditation too). And I have to say I underestimated the AI-Wonder-Lust under AggAI. Feel free to change that.

Right, that's good to know.