SGOTM 07 - Murky Waters

[FiveAces]

Is there any drawbacks to deny a trade IBT? They won't demand any tech, so the trade would be repeatable? I guess there are factors, but are they significant?

Back online now. Yes - I've found they will offer you say BW for writ IBT, but if you deny and then propose that same trade during your turn they will not take it - probably b/c they but some beakers into writ that turn and it's not a "fair value" anymore.

EDIT: they will also do this with feudalism for CS. This one should always be denied. You can easily make the trade later for CS+ only 90g. Other than that, they won't be coming at us with "reasonable" IBT trade offers we need to consider - unless maybe for drama/lit, since I've not usually held these early and don't have much experience there.

EDIT2: I think an early (pre-alpha) question we need to answer is if we build GLib, how do we get poly? By trade or self research? That affects what we research after alpha unless it is available then. In my experience unless Mansa has it, you can't trade for it until Parth is finished. And Mansa doesn't start with Myst, so hindu will likely be founded before he starts poly (assume he tries for med first) so he won't get it quickly.

RANDOM THOUGHT: Is there any benefit to building a bunch of cho-ku-nu's?

 

[klarius]

Pyramids or not, both can lead to good results.
But pyramids delays alphabet by 15-20 turns.
Alphabet is the point where we actively can start to control diplomacy. Even if we cannot get shared war directly we can get people up to pleased by gifts to avoid wars.
Pyramids greatly increases the risk to be DoWed before alphabet and by that lose early.
I have a record now of about 20% losing early with pyramids versus 5% w/o.
So, I would go pyramids only with BW and the chance to build some strong military. That delays alphabet even more, which I don't like at all.

 

[FiveAces]

I don't think the risk/reward on the Pyramids is worth it either. I'd rather take a chance on GLib - I don't think we need it - we could build early GT instead and grow faster, not research Lit and get to comm in a decent amount of time - but the extra sci and GS points makes it a better use of hammers IMO.

 

[Erkon]

Here my updated map (with Beijing settled me thinks, and warrior not moved).

I got Alpha on T45 but could not trade for BW. All AIs were known at this time.

 

[LowtherCastle]

I think I agree. I expect some good team will go for Pyramids, and if all goes well for them and for us, they might beat our finish date. On the other hand, if we play really well, especially our end-game, we might be able to even overcome that. I don't think the number or strength of AI defenders will make a difference if we use our cats/cannons cleverly. My biggest fear is that if Railroading becomes a factor, the first to railroad wins, hands down.

That said, I'm in favor of the 'sure' route that we've carefully analyzed. I'm comfortable with Gnejs playing a turn or two...

 

[klarius]

B

EDIT2: I think an early (pre-alpha) question we need to answer is if we build GLib, how do we get poly? By trade or self research? That affects what we research after alpha unless it is available then. In my experience unless Mansa has it, you can't trade for it until Parth is finished. And Mansa doesn't start with Myst, so hindu will likely be founded before he starts poly (assume he tries for med first) so he won't get it quickly.

I never have problems to trade poly 2-5 turns after alphabet. And not from Mansa. But I'm giving out alpha aggressively so the AI start to trade also.
BTW Mansa doesn't help for the Parthenon problem. That's the one reason even Mansa will not trade. The problem with the others is usually that some build Parthenon and others don't trade because it's not spread wide enough (most will trade only if 3-4 have the tech).
I still have best results with globe immediately after GLib, so I always start drama in the meantime.

Latest result on my (LC modified and changed the horses to copper) map: 640AD PA with Hatty. Alphabet path. GLib then theater (some time with 10% culture) globe.
First shared war only at CoL. Hatty snuffed Monte then Alex, but after that I needed a war on the other continent for the final points, because everybody was Jewish.
No pacifism and only 3 scientists in total.
BTW this time I could have traded up to military tradition and divine right, if I would have been delayed by a GA instead of GS. That means some research could have come back by using the GA on communism, instead of the GS on sci method (+GLib and monasteries expiring later).

 

[Gnejs]

I never have problems to trade poly 2-5 turns after alphabet. And not from Mansa. But I'm giving out alpha aggressively so the AI start to trade also.
BTW Mansa doesn't help for the Parthenon problem. That's the one reason even Mansa will not trade. The problem with the others is usually that some build Parthenon and others don't trade because it's not spread wide enough (most will trade only if 3-4 have the tech).
I still have best results with globe immediately after GLib, so I always start drama in the meantime.

Latest result on my (LC modified and changed the horses to copper) map: 640AD PA with Hatty. Alphabet path. GLib then theater (some time with 10% culture) globe.
First shared war only at CoL. Hatty snuffed Monte then Alex, but after that I needed a war on the other continent for the final points, because everybody was Jewish.
No pacifism and only 3 scientists in total.
BTW this time I could have traded up to military tradition and divine right, if I would have been delayed by a GA instead of GS. That means some research could have come back by using the GA on communism, instead of the GS on sci method (+GLib and monasteries expiring later).

Could you post a save from 640 AD so we can compare the AI tech level vs. my Pyramids game?

 

[klarius]

Could you post a save from 640 AD so we can compare the AI tech level vs. my Pyramids game?

Here you go.

 

[FiveAces]

I never have problems to trade poly 2-5 turns after alphabet. And not from Mansa. But I'm giving out alpha aggressively so the AI start to trade also.

That's the answer - I'm holding alpha out of habit to slow them down unless can I get something good for it. We'll have Lit/drama to trade early in lieu of alpha for HR, calendar, currency (maybe). Let's follow your suggestion. GLib should mean we still get Liberalism 1st without difficulty.

BTW Mansa doesn't help for the Parthenon problem. That's the one reason even Mansa will not trade. The problem with the others is usually that some build Parthenon and others don't trade because it's not spread wide enough (most will trade only if 3-4 have the tech).

I think though that he will only hold poly for his "reasons" if he's actually building Parth, which with only 2 cities he (probably) won't do unless he has marble? Not that it really matters though.

I still have best results with globe immediately after GLib, so I always start drama in the meantime.

I support this as long as our initial trader will ensure we get poly early.

I'm ok w/ Gnejs playing a couple turns too. Maybe we'll meet our neighbors? Anybody want to wager a guess? I'm thinking we only have 2 - Liz and Ghandi.

Also, the post-settle screenshot is interesting. Nobody will settle to our immediate W, so no culture pressure there, and we've got a nice mountain blockade, only broken by a hill where we can fortify a defender (in a fort if we do capture a bunch of workers - maybe I'll actually build one!). Gyanthaar is definitely giving us a fighting chance if we're DOW'd.

Finally - just from what we see, the AI won't settle close to our initial borders, so we're not likely to get -1 for close borders AND the neighbors likely won't have sufficient border tiles to be bribed cheaply into war against us.

 

[klarius]

Finally - just from what we see, the AI won't settle close to our initial borders, so we're not likely to get -1 for close borders AND the neighbors likely won't have sufficient border tiles to be bribed cheaply into war against us.

Don't hold your breath for no near settlement. It's a crowded map. There can be cities SW as well as SE.
And with sailing or open borders maybe also W.
The AI has no problem to settle pretty close as long as there are resources to get in their fat cross. And they don't care if it's so that they will never win the culture battle.

 

[FiveAces]

Don't hold your breath for no near settlement. It's a crowded map. There can be cities SW as well as SE.

You're right. In fact, the fogged tile in the mountains has 3 food resources and silver. Another reason not to rush camping the deer - we might lose it for a while early on if a creative civ is below us.

Is there anything specific anyone would like me to experiment with in a test game tonight?

 

[FiveAces]

Originally Posted by klarius

The AI will never bomb a last city, so that's no option.
From the SDK it's a necessary condition that the team's plot culture is below 60%. And that's not the case when the previous owner is dead.

Let's say our home continent is ~5 tiles short of the dom limit. So we need 1 city overseas. Now say that we take one from an AI that has more than 1 city. If we immediately gift a GA to our partner in that city, will they bomb it? They'll have well less than 60% culture since the overseas AI won't be dead, and the city will immediately come out of resistance and expand borders, putting us (hopefully) over the dom limit.

 

[klarius]

Originally Posted by klarius


Let's say our home continent is ~5 tiles short of the dom limit. So we need 1 city overseas. Now say that we take one from an AI that has more than 1 city. If we immediately gift a GA to our partner in that city, will they bomb it? They'll have well less than 60% culture since the overseas AI won't be dead, and the city will immediately come out of resistance and expand borders, putting us (hopefully) over the dom limit.

Well, they should bomb. But that doesn't solve the problem how we get them to get a holy or wonder city in the first place (other cities they will raze overseas unless it is the last city of that civ).
And if we get to this position it might still be that no tiles at all are gained by the bomb, because the enemy culture is so high (note: wonder or holy city).
But, OK we could kill this civ then after. That's a lot of ifs. I rather say we wait first and hope for our continent being enough .

If we get to the point needing overseas holdings it might well be faster if we just raze cities on our own and hope for a settler from our ally.

 

[LowtherCastle]

You're right. In fact, the fogged tile in the mountains has 3 food resources and silver. Another reason not to rush camping the deer - we might lose it for a while early on if a creative civ is below us.

...and our stone......but my guess is further SW is where a settler will settle.

Here are my guesses:
I suspect Gyathaar was fairly careful to make this game winnable. Unlike SG5, where AH first was bountiful for the human capital and SG6, where BW first was, my guess is that both choices pay off. Furthermore, my guess is that no Industrious AI has stone, so those who go for Pyramids have a fighting chance of getting them. SOmeone may settle to the far W, 2 tiles from the sheep, but since they'll need Sailing, presumably, they won't steal our sheep. It might even be set up for us to keep our cattle, but I doubt it, but that just depends on Gyathaar's mood, the day before he finalized this map...

My guess is that Kubla will be on our continent, along with a bunch of HeredRule AIs and Mansa. I wouldn't be surprised if there were 1 or 2 other Creative AIs. That would make this competition very interesting, because who do you chooe between two equally powerful ones: The one nearby or the one far to the south? And I think we'll be able to Dominate with our continent alone, but that depends on Gyathaar's mood the day he finalized this map...

 

[LowtherCastle]

(other cities they will raze overseas unless it is the last city of that civ).

How does Kubla's 0 RazeCityProb factor in then?

 

[klarius]

How does Kubla's 0 RazeCityProb factor in then?

I has nothing to do with it.

There are first conditions checked where all leaders will never or always raze.
Only if none of these conditions hold true the RazeCityProb is checked against the RNG.

 

[Erkon]

I played my (modified) map to 100 AD, at which time I had learned Liberalism (75 AD). To me this is insanely early, but that's probably due to the map and not me

First question: How does this date compare to other tries on my map? I could not find any hint in the previous posts.

Observations: Oxford is built in 4 turns. No need to save trees for it.

Build path: Theatre, GLib, Globe

Second question: Is this a suitable build path, or does GLib go before Theatre?

I didn't manage to trade BW in first round, but in second when I had sailing as well. More important, I could not get Cathy to DoW because she had her hands full from 2200 AD. It turned out that she was having wet dreams about me, but it took 2300 years before she shook of her shyness and penetrated my borders with lots of pointy sticks.

Third question: How do you/we choose PA-partner? Is it more important to get an early war with someone, than getting the optimum partner?

 

[Erkon]

Are you saying I should go 20 turns now?

I'm saying I can't make up my mind. klarius suggestion to break at T20 makes sense. But then the next stop (T29 @ Writing + WB) will make LC's turn set very short and rather uninteresting. Perhaps we should skip the alignment and just follow the standard 20-15-10-10 turn sets.

 

[LowtherCastle]

I'm saying I can't make up my mind. klarius suggestion to break at T20 makes sense. But then the next stop (T29 @ Writing + WB) will make LC's turn set very short and rather uninteresting. Perhaps we should skip the alignment and just follow the standard 20-15-10-10 turn sets.

Trust me--I can make my turnset veeerrrrry interesting...

20-15-10 is perfect for finishing Alpha on T45 which looks most likely. Just make sure Gnejs pauses and uploads right before his mistakes.

 

[klarius]

I played my (modified) map to 100 AD, at which time I had learned Liberalism (75 AD). To me this is insanely early, but that's probably due to the map and not me
On various maps I had times form 100BC into the 200ADs. It's not really the map, but things like lux trades, religions and tech trades (CoL and philosophy traded or researched). These facts vary a lot even on the same map.

First question: How does this date compare to other tries on my map? I could not find any hint in the previous posts.
Observations: Oxford is built in 4 turns. No need to save trees for it.

On various maps I had times form 100BC into the 200ADs. It's not really the map, but things like lux trades, religions and tech trades (CoL and philosophy traded or researched). These facts vary a lot even on the same map.
I still used 1 or 2 forests for university-oxford.

Build path: Theatre, GLib, Globe

Second question: Is this a suitable build path, or does GLib go before Theatre?

I built GLib before theatre but noted lately that I have trouble to even get all resources improved in time.

Third question: How do you/we choose PA-partner? Is it more important to get an early war with someone, than getting the optimum partner?

Wait and see. Definitely don't go wildly into a war w/o having good knowledge of the map and all civs on our landmass. If there are a few turns missing when we get communism that's no big problem - we can still help our partner also w/o PA (but we risk duplicating research).
But, if we bet our money on a loser and piss off others that is a problem.