SGOTM 13 - One Short Straw

[RRRaskolnikov]

Meanwhile, her GNP fluctuates wildly between 20 and 23.

Maybe a noobish question, but how does the AI manage its slider? (maybe the problem comes from rounding issues? though that would fit better a 1 point variation)

edit: going with your explanation, she would have shifted how many times, twice? (meeting two other AIs?)

 

[babybluepants]

GNP fluctuation:
Don't know, but her food and hammer graphs fluctuate on the same turns - I think it's just tile assignment switches. Don't know much about AI behavior. Perhaps she's riding the happy limit, switching build to settler...

Why aren't we considering other techs as well?

Wasn't an exhaustive list.

TBH, I don't really trust that we can bulb CS or Machinery here, unless the AIs are seriously gimped by DS.

We're skipping Medi so as not to block the Astro bulb via Philo.

Avoiding CS for the same reason. I'd question this, maybe, as I'm not sure how we're supposed to get the multiple GS's in the short term, (without Caste/Pac/PHI, with GP+GM pollution, probably slow libraries in secondary cities), and it could still be a non-Astro map. We have plenty of time to evaluate this, though, assuming Oracle MC.

I suspect we'll wanna run Slavery pretty aggressively given the total lack of base production. And we're not SPI. Early COL didn't seem so good in this case. I guess we're ORG, but we still won't be building CH's until after Monument - WB - Granary - LH - maybe Library... Perhaps we won't build them at all.

Edit: I'd probably rather bulb Currency than COL, given multiple intercontinental TR's.

 

[RRRaskolnikov]

yeah courthouse savings won't be stellar, but coupled with a shrined religion (confu) it could be worth it. I am brainstorming mostly though.

Good point about poly/med, I had forgotten what was decided already.

OT: I just saw your Malian field of gold (in lieu of a WOTM starting position)... I might try it!

 

[babybluepants]

OT: I just saw your Malian field of gold (in lieu of a WOTM starting position)... I might try it!

You know I was just willing a gold to appear in this game. It totally worked!

 

[LowtherCastle]

1. I'd like to see a comparison test between MC+colossus and CoL+Mids. Unless MC+colossus gives us a huge headstart on teching, it's probably a lot weaker, plus, to me it's a gamble (astro?).
2. Of course we research masonry before building the Oracle! (Duh...wudamImissin?)
3. I think we should beeline the Oracle, so it's CoL or MC. We're already late on it, after pottery+fishing. We still need masonry+myst+poly+PH(+wtg) and we should also research AH asap, preferably before settling on any plains tiles.
4. This is a huge map, we're hammer-poor, and the AIs are clearly far away from us (> the usual 9-10 tile spacing). Galleons rock for this situation. 5:movement: + 3 passengers, impervious to suiciding caravels and triremes. We should probably be able to get Astro by 1AD with all this food. And then we're able to conquer rapidly or research more, as needed. Furthermore, that 3pop AI might be accessible only across ocean. The Representation also gives one more benefit: an instant +3 whipping cushion for a rapid mobilization.
5. And did I mention an earlier capture of the GLH, if we so choose? I say phooey on those humpty-dumpty, fiddle-dee-phumpty galleywalleys...

 

[shyuhe]

Currency will be powerful here -- is it possible to shoot for a currency oracle? It's not quite as powerful as Colossus but it frees us from having to settle the copper earlier than we would (presumably) like. Plus it puts the tech path as writing--myst--poly--priesthood--alpha--currency (Oracle). Not sure what kind of dates we'd be looking at with this path though.

I would prefer to avoid civil service until we are sure about the astronomy position. It will make a big difference in the mid-late game portion of this game and can save a lot of turns in the long run. Plus, Paris isn't so great as a bureau capital anyways.

 

[LowtherCastle]

Shyuhe, like bbp, I"m curious how you figure that Vicky knows other AIs.

 

[shyuhe]

Oh I thought Dhoom had posted at some point that Vicky's EP were split against us. I think I'm misremembering So never mind, maybe my alpha plan isn't so hot.

 

[Dhoomstriker]

AI_6 = Pop 3 City (21000 Rival worst)

Despite the text "rival" worst, I *think* the human player might be included in the results... we do have only a pop-3 City.

Vicky doesn't have BW yet (no slavery), so she probably either has 2 more archers, or Sailing + 2 warriors or something.

I can confirm that Vicky knows Sailing, as her knowing it gave us a bonus to research on that tech.

5 - We will want triremes if we're going to declare war I think.

I still think that it might be worth sending out a Pillaging Galley... the AI is unlikely to have a Galley in position to do anything about it and by the time that she arrives on our shores to retaliate, enough time should have passed that we'll be able to sue for peace before she can do any damage.

3) Possibly build Pyramids

They say that building The Pyramids costs you a few early-game Cities. Since we have a lot of spots to expand to, I would tend to think that we could skip building The Pyramids.

How do we know this? Espionage seems even.

If Vicky knew 0 other AIs, then we'd have equal Espionage. Instead, she'd been spending her Espionage Points on another AI. Only after we started spending Espionage Points on her did she turn around and spend some on us, beginning on the following turn.

If you're the first player that an AI meets, Espionage Points will be even (unless I am wrong yet again--it happens). In this case, we have 48 Espionage Points and she only has 44, which indicates that she didn't use the Espionage Slider to try and "catch up" to the first turn' worth of Espionage Points that we spent on her (or at least if she did raise the Espionage Slider, she only allocated 4 Espionage Points' worth to us).


Note that I was show-casing this fact at the bottom left of some of the screenshots by hovering the mouse over Vicky's name while taking the screenshots (or, alternatively, hovering over our name).

Is she switching back and forth between techs?

It is very possible that Vicky was researching Hinduism and just recently switched to another tech. I forgot to check the F7 screen to see if she was the founder, but likely another AI founded it and she switched techs as a result... but that would have been only for one turn and probably only for the last turn of play, so it doesn't really explain the graph.

Perhaps she is Farming over top of her Cottages or something? Switching between Coast and non-River Grassland Cottages?

Switching multiple squares such that Food or Hammers do go up or down but with a net effect of multiple switches making it appear that these values were constant?

A Barb unit pillaging a Cottage or temporarily walking over top of one?

She is Financial, which could mean that switching between a Coast and a non-Riverside Cottage could mean a difference of 2 Commerce.

 

[shyuhe]

Some random ramblings but my thoughts on wonders:

Oracle (duh)
Pyramids (maybe?) -- they're kind of useless without caste system or library+markets (both expensive for us). I'd prefer to stay in slavery since we'll be whipping most of our production.
GL -- a bit of a detour and I don't think the extra GS will arrive in time to help get to astro.
Parthenon -- ditto, too slow.
SoZ, ToA, Paya -- Useless for us.
MoM -- maybe worth building through whip overflow? It'll let us use 3 GP to run the equivalent of 3 golden ages.
Colossus -- strong with seafood, but obsoleting with astro may nerf its worth significantly.
GW, HG, Pizza -- out of the question.
Other medieval wonders -- hopefully we won't tech that far

 

[Dhoomstriker]

Currency will be powerful here -- is it possible to shoot for a currency oracle? It's not quite as powerful as Colossus but it frees us from having to settle the copper earlier than we would (presumably) like. Plus it puts the tech path as writing--myst--poly--priesthood--alpha--currency (Oracle). Not sure what kind of dates we'd be looking at with this path though.

Currency is definitely an option.

However, going for Metal Casting does not mean that we HAVE to settle on the Copper.

Metal Casting is useful for:
a) Forges = extra production when whipping and due to having a Gold for +1 Happiness with a Forge, it is Happiness-neutral to whip this building

b) Triremes


The AIs are not likely to get Metal Casting for a long time, even if we do Oracle it, so The Colossus should be "safe" to save for later.


EDIT: One thought on Oracling Currency is that doing so gives us the 20% discount on self-teching of Code of Laws for knowing both of the optional pre-requisites (Priesthood--which we would have to have learned if we are building The Oracle--and Currency).

Since our capital is not so good for Bureaucracy, a Trade Route economy + cheap Courthouses could be a good way to allow us to REX like mad.

 

[babybluepants]

If you're the first player that an AI meets, Espionage Points will be even (unless I am wrong yet again--it happens). In this case, we have 48 Espionage Points and she only has 44

Isn't there always a 4-point difference for some reason? Like I said, I don't really understand this - may be wrong.

so The Colossus should be "safe" to save for later.

But kinda useless if we beeline Astro by 1 AD.

Given our likely whipping cycles, we're probably only working 2 coastal tiles tops per city. If that's the case, Currency = TC. Like shyuhe said, though, I think a major reason to do so would be early Alpha, which only makes sense with more AIs met.

2. Of course we research masonry before building the Oracle! (Duh...wudamImissin?)

Maybe we don't really wanna settle marble this early or waste early beakers? Marble doesn't have food in the initial radius, for one thing. I would think Pyramids are more of a factor in deciding whether to research Masonry.

 

[shyuhe]

If we can spare 10(? I can't do epic math) worker turns, we can settle 1SW of marble and quarry + road it. It's not a great third city, but it'll be decent with marble + clam.

 

[Dhoomstriker]

HG-- out of the question.

What's wrong with The Hanging Gardens? Generally, it's a nice way for you to distribute production from one City to every City in the empire, by using Hammers in one City and then whipping the +1 population point in every other City. We do have Stone nearby, too.

If we REX, then it becomes even stronger.


Probably the toughest part about this Wonder would be beating the AIs to it. Is that why you would suggest avoiding it--for fear of missing it?

 

[shyuhe]

HG is a bit expensive and comes at a bad time (with strong AI competition for it too). Aqueduct is 133 hammers, which is a 6 pop whip with some change. Paris has 8 base hammers so it'll take a long time to build the aqueduct, much less HG.

Do we want to give any thoughts to Moai? It's not a bad wonder but I'm not sure how it fits into our plan. If we build it in Paris and work 3 coastal tiles, it'll take upwards of... 50 turns to break even. So not that bad of an investment but we need stone of course.

 

[LowtherCastle]

1. We have 44eps on Vicky. SHe has 48 on us, evidently because she met us during her turn and started accumulating them a turn before us. I doubt she's met anyone else.

2. Marble is perfectly situated to 1) take the worker from gold to chop a monument, and 2) build the Oracle all by itself. Its borders will expand about when we have a wb or two ready for it. It 2pops the granary with max overflow into the Oracle and then it's just one more poprush-overflow from completion. Basically, it solves bbp's wkr/settler transportation problem. Speed up the settler as many turns as needed (delay the lh a couple of turns) to pick up the worker. Then the worker goes to stone and is never needed again at marble. Settling any other city that needs the worker doesn't time well.

 

[RRRaskolnikov]

^^ I really like Lc's proposal about oracle build.

in other news, I have opened the save to check things out without screwing up everything (prolly thanks to Dhoom's multiple reminders) Still disgesting the micro already planned... will start writing ideas for the PPP in a couple of beers, I meant hours.

edit: and more seriously, tx to Dhoom for updating the test game and the spreadsheet (I have v10, hope that's the last one)! Really apreciated.

 

[babybluepants]

That sounds pretty excellent, LC. Let's test it out.

 

[Dhoomstriker]

edit: and more seriously, tx to Dhoom for updating the test game and the spreadsheet (I have v10, hope that's the last one)! Really apreciated.

Yes, you have the latest version, in that the only thing that really changed was that we learned Sailing a turn early.

Of course, as soon as we settle City 2, things will differ... plus, I did not record down Binary Science Slider movements, since we couldn't really be sure when City 2 would get settled.

However, now that you have knowledge of exactly how far City 2 will be located and have a test game to help plan things out, you can certainly add additional info, such as which turns you will switch your Science Slider.

Also, the Worker's actions aren't really listed anymore, since again, we couldn't be sure what the Worker would do.

However, you should probably record the timing of when you will put the Worker on the Galley in your spreadsheet or PPP (whatever you end up using), so that you don't forget to do so. If you do continue to use the spreadsheet format, then you could just use the Worker actions columns for this kind of movement.


As for Espionage... LC is right... the first number is how much the AI has spent on us and the second number is how much we have spent on the AI... which means that we are behind by 4 Espionage Points. It doesn't really seem fair to me that this method is how the game calculates it, but it does indicate that Vicky had not met any other AIs by the time that she met us.

From the Espionage Points alone, we can't know for sure if she met any AIs after that point and just decided that she hates us the most or if she still doesn't know any other AIs.

However, if she doesn't know anyone else, then:
If we want to get anything from Vicky via Alphabet trading, we might be better off declaring war with a Pillaging Galley and optional choking Archers. An AI will give up Monopoly techs for Peace. Parking 2 Archers on either a Forest or a Hills square (they are equivalent for Archers since Archers get +25% defensive bonus on Hills squares) within an AI's big fat cross, particularly if we do so not immediately adjacent to the City but still within the fat cross, will likely net us some techs for Peace, as well as some early Great General Points.

Of course, no one generally wants to research Archery, but it IS a pretty cheap tech if you do plan to leverage it.

 

[RRRaskolnikov]

If we want to get anything from Vicky via Alphabet trading, we might be better off declaring war with a Pillaging Galley and optional choking Archers. An AI will give up Monopoly techs for Peace. Parking 2 Archers on either a Forest or a Hills square (they are equivalent for Archers since Archers get +25% defensive bonus on Hills squares) within an AI's big fat cross, particularly if we do so not immediately adjacent to the City but still within the fat cross, will likely net us some techs for Peace, as well as some early Great General Points.

Of course, no one generally wants to research Archery, but it IS a pretty cheap tech if you do plan to leverage it.

Let's not assume the unprobable. We are more likely to met more AIs. And with marble, stone, and copper around, other opportunities are just better (seriously this is not a map asking for archery, which leads to the conclusion that archery must be skipped). that's my opinion at least

---

I will keep on with your spreadsheet, updating it once I have tested a few things.

First idea to work with:

-> we have 6 turns worth of exploration with one of our western wb. He has to be on fish tile by t75... what does he do? The currently foged tiles around marble island will be unfogged by the galley so where can he head? And which one is scheduled to net the fish (I vote the northern one)?