SGOTM 21 - Plastic Ducks

[Doshin]

Another possibility is that we gift (for 40G) Aesthetics and Literature to Elizabeth, now that MM is going to CoL.

---

Elizabeth is Philosophical and is likely to generate a GProphet from the ToA (her GMerchant was a bit unlucky). Judaism will be well spread, thanks to our continent. A shrine alone would be a good reason for us to attack.

The AI won't construct wonders unless they are cheap relative to other builds (why HC built a T10 Oracle > Settler in SGOTM19) and the ToA is very expensive. I think Elizabeth has Marble and/or has run out of space in which to expand.

Since she has Polytheism, she could build the Parthenon or GLibrary. Both would be good for us to capture.

If she gets the GLibrary in London, it becomes the best GP farm in the game: it would supply a passive 16 GScientist points per turn, 6 GProphet points, and 4 GMerchant points for Elizabeth or for us. That's 26 before running any GPeople, or building the NE.

I don't like to gift away techs, but Elizabeth's value as a trade partner is much lower now that we know Mansa in the game, and she is far closer to us than Louis. When we switch into Caste, we could also bribe her to join us.

Weak philosophical leaders with the GLibrary can make great targets. shakabrade might remember this old map, where (in my game) the AI Pericles built the GLibrary:

Spoiler :

"Hello, friend. My name is Pericles and I science!"

Spoiler :

Science 1:

Spoiler :

Science 2:

Spoiler :

Science 3:

Spoiler :

Science 4:

Spoiler :

Science 5:

Spoiler :

Welcome to the Republic.

 

[Doshin]

**Edit**
What was the way to bulb Engineering again? Double bulb Astro and then bulb Engineering with a Great Scientist?
The list always give me a headache.

Spoiler :

Great Scientist:

Writing
Mathematics
Scientific Method
Physics
Education
Printing Press
Fiber Optics
Computers
Laser (BTS)
The Wheel
Alphabet (BTS)
Philosophy
Chemistry
Fission
Fusion
Optics
Paper
Astronomy
Biology
Electricity
Flight
Genetics
Compass
Satellites
Aesthetics (BTS)
Sailing
Alphabet (Vanilla & Warlords)
Calendar
Medicine
Ecology
Advanced Flight (BTS)
Iron Working
Metal Casting
Engineering
Steam Power
Liberalism
Agriculture
Masonry
Bronze Working
Machinery
Gunpowder
Refrigeration
Superconductors (BTS)
Rocketry
Fishing
Combustion
Plastics
Composites
Stealth (BTS)
Mining
Military Science (BTS)
Radio
Meditation
Drama
Theology
Music
Civil Service
Democracy
Corporation
Communism
Economics
Hunting
Archery
Animal Husbandry
Construction
Robotics
Monotheism
Mass Media
Horseback Riding
Replaceable Parts
Rifling
Artillery
Future Tech

Yes. You avoid Meditation (leads to Philosophy) and Civil Service (leads to Paper/Edu/PP etc.).

---

Given that we have no Cottages and the GLH, I wonder if we could bulb our way to take Communism from Lib. That would require ~4 GScientists.

 

[Kaitzilla]

Another possibility is that we gift (for 40G) Aesthetics and Literature to Elizabeth, now that MM is going to CoL.

---

Elizabeth is Philosophical and is likely to generate a GProphet from the ToA (her GMerchant was a bit unlucky). Judaism will be well spread, thanks to our continent. A shrine alone would be a good reason for us to attack.

The AI won't construct wonders unless they are cheap relative to other builds (why HC built a T10 Oracle > Settler in SGOTM19) and the ToA is very expensive. I think Elizabeth has Marble and/or has run out of space in which to expand.

Since she has Polytheism, she could build the Parthenon or GLibrary. Both would be good for us to capture.

If she gets the GLibrary in London, it becomes the best GP farm in the game: it would supply a passive 16 GScientist points per turn, 6 GProphet points, and 4 GMerchant points for Elizabeth or for us. That's 26 before running any GPeople, or building the NE.

I don't like to gift away techs, but Elizabeth's value as a trade partner is much lower now that we know Mansa in the game, and she is far closer to us than Louis. When we switch into Caste, we could also bribe her to join us.

Weak philosophical leaders with the GLibrary can make great targets. shakabrade might remember this old map, where (in my game) the AI Pericles built the GLibrary:

Aesthetics for 40 would be ok with me.
We need to hang onto Literature for a while longer so we can trade Construction+Literature to Mansa for Code of Laws in 3 turns.

Liz probably has marble ya.
Temple of Artemis costs 350 without marble to help.
It requires the tech Polytheism.
Liz discovered Monotheism in 1920BC (turn 52) and Temple of Artemis was built in 825BC (turn 82).
It would be possible, but still impressive for a coastal city to make ToA without marble.

 

[Duckweed]

Liz might not have marble if we look at the current stone and marble location, they are all far away from us before Astronomy. 825BC ToA is fast without marble, but not impossible. Shall we wait a few turns to see what's her current tech, we could trade Math+Aes if she get Monarchy.

I'm thinking about invest 10% slider into CoL so that Construction alone is enough to trade CoL.

I have thought about Kaitzilla's suggestion of beeline and bulb Astro. Quick hard.

• We need ~40 turns to claim the land from William, Shaka and Liz, so Astro is not that urgent.
• Bulb to Astro+Eng needs at least 3 GSs, which require 20~25 turns to grow the GP farms and produce them.
• We might need to start a weak GA (with less mature GP farms and no Pacifism bonus) now to reduce the cost of civic switch since when the empire is more than 12 cities, switching 2 civics cost 2 turns of anarchy.
• Bulbing path means that we are likely losing the beakers that we could trade: Machinery, Compass, and the bureaucracy bonus.

 

[Doshin]

(1) Argos: Settler/GScientist management:

Spoiler :

Key points:

• We can grow on T95. Run 1 Scientist and build a Market or Forge, whichever you prefer.
• The Settler needs to be produced on T99. It's easy to do so while running 1 Scientist on T96–98.
• From T99 onwards, the city will need to run 2 Scientists. If it doesn't, Sparta's GProphet will outpace it.
• On T99, the Galley can be 1SW or 1NE of the city, as long as it can unload the Settler in Iron City on T100.
• Fish City is founded on T101.

I haven't thought much about what Argos should build after T99.

(2) Iron City

Spoiler :

• On T96, the city continues to work the Iron.
• On T97, the city switches to the Sheep.
• On T100, the Granary completes with 18/22F. Turn on Avoid Growth.
• On T101, the city has 22/22F. Turn off Avoid Growth.

The city could 1-pop whip a Galley, Trireme, or Work Boat @ size 2.

 

[Kaitzilla]

In the test game, barb galleys showed up in force starting T110.
We should seriously think about making a Trireme in Iron City if we are going to settle arctic fish cities to protect them.

The +1 circumnavigation is a huge help!
Fish City producing its own workboat can settle the fish as soon as it is produced.
bug with work boat in the queue can pop borders and produce a workboat at the same time with the proper chop.
Also, the 0xp phalanx is no longer stranded on Thebes island and can be put to good use garrisoning Knossos against any Shaka Impis or Chariots and for +1 to let us keep 2-pop whipping military.

I had a huge force of 12 swords, 9 catapults, chariot, and phalanx built by T105 in a test run.
With careful timing of chops and whips, the military production of Sparta, Knossos, and Mycenae is tremendous.

Spoiler :

 

[Doshin]

(3) Fish City

T98–100:

Spoiler :

T101:

Spoiler :

T102:

Spoiler :

T103:

Spoiler :

T104:

Spoiler :

T105:

Spoiler :

T106:

Spoiler :

----

I use 1 more Worker than before, because circumnavigation means our WB can be produced and improve the Fish on one and the same turn.

Building Culture, with a WB queued, lets us pop the border and work the Fish after 2 turns.

The old micro, with 2 Workers, before the circumnavigation bonus, led to:

Spoiler :

The new micro, with 3 Workers and the circumnavigation bonus results in:

Spoiler :

 

[Kaitzilla]

Wow, Fish City really sped up with all that micro.
Good work Doshin

Liz might not have marble if we look at the current stone and marble location, they are all far away from us before Astronomy. 825BC ToA is fast without marble, but not impossible. Shall we wait a few turns to see what's her current tech, we could trade Math+Aes if she get Monarchy.

I'm thinking about invest 10% slider into CoL so that Construction alone is enough to trade CoL.
...

Mmm, if we aren't gonna trade her any techs for that 40, let me do the beg calculation real quick.
We met Elizabeth on Turn 44.

http://forums.civfanatics.com/showpost.php?p=12041660&postcount=2744

Begging mechanics for non-vassal deals:

Maximum amount that is accepted is

(TURNS_KNOWN + 50) * 2 * (OUR_POWER / THEIR_POWER), further multiplied by 3 if that AI is our land target.

Note that this remembers earlier gifts granted, so if this formula gives a value of 1000, and we already have begged 700 during the game, the max gift value granted is only 300. Again, the value of gold is GOLD*2, or GOLD*3 if the AI is in financial trouble. Beakers are approx BEAKERS*1.5.

It is currently Turn 95, so we've known her for 51 turns.
She isn't a land target and I don't think she is in financial trouble.
We haven't begged her for anything yet.

So she should give us up to FLOOR[[(51+50)*2(power_ratio)]/2]= 101(power_ratio) (floor means round down to nearest 10 in this case)
If our power ratio is 0.5 with her, then we can beg up to 50 and have it be successful.
Since we want 40, the question is if our power ratio is above 0.4?
I think it currently is, so begging Liz for 40 should work.

Obviously, we could wait a few more turns for our Power Rating to skyrocket and definitely have a successful beg in that case.

 

[Doshin]

Wow, Fish City really sped up with all that micro work.

Thanks. Using the Corinth Worker lets us chop a WB 1-turn sooner, so +3F, and the new circumnavigation bonus lets us improve the Fish immediately, providing another +3F. And three Workers can chop a Granary 2-turns quicker than two Workers.

---

FWIW, using 3 Workers to chop a LH (built on T107) seemed ok to me, but we could just as well use 1 chop and 1-pop whip a LH on T107 (built on T108). That would let us send 2 of the Workers back towards Mycenae and Corinth on T104–5, and only cost –1F in Fish City. The food loss is, in fact, a little higher if you consider that 2 Workers could farm the Grassland before moving west. But I don't know how whether other cities need these Workers more.

So, it is beneficial to chop the LH and improve the riverside Grassland, but the benefits are small and need to be set against using the Worker turns elsewhere.

 

[shakabrade]

Harder than expected and with totally different (and more simple) approach than expected, GPU temp has dropped by 25°C in idle and is around 75°C at full load (it wasn't able to be fully loaded before).



Now, format C and fresh Windows.

Hopefully testing this evening.

Must read posts carefully first for some micro gems.

 

[Doshin]

Hopefully testing this evening.

Must read posts carefully first for some micro gems.

Good luck. My posts #945 & #947 should help with the NE corner of the map. For the mainland: Sparta/Mycenae/Knossos should produce 2-units every 3 turns (or better ), with an initial emphasis upon Swords > Catapults (Swords can piggyback a Shaka stack to attack Amsterdam). Athens and Thebes will eventually 3-pop whip Settlers for the ice islands.

The ice island cities should be settled after we reach Civil Service. 11 cities is the limit for a 2-civic switch with 1-turn of anarchy.

I'd prefer not to chop the Forests north of Sparta. They can be used to build the Heroic Epic.

I'm thinking about invest 10% slider into CoL so that Construction alone is enough to trade CoL.

If it's T95 now, we could research Meditation on T98. We turn the slider up on T96 so that we are 1 turn away from completion. On T97, we put 100% research into Meditation. Max overflow goes into the next tech (i.e. Civil Service).

We can sell Meditation to Louis and any other AI with gold, and trade Meditation + Construction to Mansa.

I don't think we need to bulb our way to Astronomy soon. Warring with Willem/Shaka/Elizabeth will take some time (haven't estimated how long exactly) so early Galleons are only marginally beneficial, while Universities are cheaper than Observatories.

 

[Duckweed]

If it's T95 now, we could research Meditation on T98. We turn the slider up on T96 so that we are 1 turn away from completion. On T97, we put 100% research into Meditation. Max overflow goes into the next tech (i.e. Civil Service).

We can sell Meditation to Louis and any other AI with gold, and trade Meditation + Construction to Mansa.

That's a good idea. But, there's still a potential problem -- whether we could get Monarchy in time.

It's likely that

William: Construction
Mao: Calendar
Toku: useless

Shaka is more likely going for Math->Construction, Izzy is too slow. Liz and Louis have good chance of researching it, or Mansa has chance of going for it after CoL.

Therefore, if we can't get the Monarchy trade in time, we waste some beakers on a tech that we could trade later. Of course, if we do get the Monarchy trade, we are happy to get bureaucracy bonus 1 turn earlier.

 

[shakabrade]

If we grow Argos to size 6 and whip a settler on T96, we can OF into WB for NFish and get NFish one turn sooner.

Also, Sheep workers can be separated on T95: one can go on Pasture and one to chop a forest for Granary and they'll finish at the exact time. Granary is completed with 11/22 at the bin, which is good, I guess. We skip building the road and also skip moving 2 workers into a forest, so 3 worker turns are saved + we save a forest for infra chop in NFish.
Argos will still get GScientist before Sparta gets a Prophet.

The only disadvantage is 1T delayed Phalanx (because galley has to pick him up) from Thebes.

Spoiler :

Edit:
T106

Spoiler :

One more good thing is faster Farm in NFish.



Also, if Sheep are building Granary, they should build LH too so we could whip those hammers back. It hurts me to see the Iron whipped away somehow.

Is there anything keeping us away from whipping a Trireme in Thebes after Odeon is completed?
I am still not all that familiar. Wanted to play a bit before reading the discussion.

Good night people.

 

[Duckweed]

I am still not all that familiar. Wanted to play a bit before reading the discussion.

I'd usually do the opposite, read, think, discuss, and finally test as that saves times.

I have not run any test, but if Argos could whip the settler without affecting the GS, that's good.

Sheep should not produce LH at least in short time since coastal tile is weak and there's many other more important builds. The Iron tile is OK to be whipped away if there's something we need right away as it's only slightly better than a grassland farm.

I'm not saying that we should not whip a trireme in Thebes. But, there are 4 fish islands south, Thebes could supply 1 settler.

Note, the tile 1SW of Fish city is a grassland, not a plain.

 

[Kaitzilla]

...
Note, the tile 1SW of Fish city is a grassland, not a plain.

Oh! Good catch.
Yes, the test game should be updated to make 1S1W of Fish City a grassland tile.

Here's some more T95 Real Game Screenshots to help us plan our war.
I think Amsterdam has Walls (+50% defense) now.

Spoiler :

I no longer think we should camp Shaka's horse tile.
I think we should DOW Willem on T99 and push for Amsterdam.
The rest of our forces should move to 2N2E of uMgungundlovu to invade Shaka around T105 while he focuses on The Hague.

Our scout should get a look at Amsterdam on T98, Nobamba on T100, and uMgungundlovu T101, and Ulundi on T102 noting all troops and defenses.

Once we capture uMgungundlovu, we can either move on Shaka's capital or Nobamba.

Utrecht can easily be toppled by our reinforcements streaming from Knossos.
So the 1st Shaka city would should attack should be uMgungundlovu after we have Amsterdam in my opinion.

 

[Doshin]

If we grow Argos to size 6 and whip a settler on T96, we can OF into WB for NFish and get NFish one turn sooner.

Can you 2-pop whip at size 5 (T95 or T96) and still produce the WB + Settler in time? If you are 2-pop whipping, size 5 is more efficient than size 6.

Also, Sheep workers can be separated on T95: one can go on Pasture and one to chop a forest for Granary and they'll finish at the exact time. Granary is completed with 11/22 at the bin, which is good, I guess. We skip building the road and also skip moving 2 workers into a forest, so 3 worker turns are saved + we save a forest for infra chop in NFish.

This all sounds good. When I was testing, the Argos WB was too slow, but we've gotten the circumnavigation bonus since.

Argos will still get GScientist before Sparta gets a Prophet.

The only disadvantage is 1T delayed Phalanx (because galley has to pick him up) from Thebes.

Also, if Sheep are building Granary, they should build LH too so we could whip those hammers back. It hurts me to see the Iron whipped away somehow.

If 1F = 2H = 2C (approximately), the Plains Iron is worth about as much as an ordinary riverside Grassland farm. So it's not a problem to whip it away, especially since the city regrows instantly. A Lighthouse is a 60H investment with a very low rate of return compared to (e.g.) a Trireme that lets us settle a city 1-turn sooner, so it's not a problem to delay this.

Is there anything keeping us away from whipping a Trireme in Thebes after Odeon is completed?
I am still not all that familiar. Wanted to play a bit before reading the discussion.

I think the plan was to grow to size 6 and 3-pop whip a Settler in Thebes.

Unhappiness in Athens/Corinth/Thebes/Argos is delicate: but if you 3-pop whip a Settler at the happy cap, you can manage anger and OF from the Settler will generally be enough to complete a WB (which can't be whipped).

Good night people.

I no longer think we should camp Shaka's horse tile.
I think we should DOW Willem on T99 and push for Amsterdam.
The rest of our forces should move to 2N2E of uMgungundlovu to invade Shaka around T105 while he focuses on The Hague.

Our scout should get a look at Amsterdam on T98, Nobamba on T100, and uMgungundlovu T101, and Ulundi on T102 noting all troops and defenses.

Once we capture uMgungundlovu, we can either move on Shaka's capital or Nobamba.

Utrecht can easily be toppled by our reinforcements streaming from Knossos.
So the 1st Shaka city would should attack should be uMgungundlovu after we have Amsterdam in my opinion.

I think I see this a little differently. I'd expect us to have a main stack, which we'd reinforce after capturing Amsterdam, and a secondary stack made up of late-comers.

The main stack deals with Willem and then moves along the western coast: 9 tiles from moment of invasion through to Ulundi, which needs to be attacked from the west due to the river. This stack can't really be reinforced and attacks two hill cities, so it needs to be large. A secondary stack would then deal with the eastern cities. This would be smaller, and put together with the newly built units:

Spoiler :

We can pre-road Utrecht ---> uMgungundlova to speed the eastern stack up

 

[Doshin]

That's a good idea. But, there's still a potential problem -- whether we could get Monarchy in time.

It's likely that

William: Construction
Mao: Calendar
Toku: useless

Shaka is more likely going for Math->Construction, Izzy is too slow. Liz and Louis have good chance of researching it, or Mansa has chance of going for it after CoL.

Therefore, if we can't get the Monarchy trade in time, we waste some beakers on a tech that we could trade later. Of course, if we do get the Monarchy trade, we are happy to get bureaucracy bonus 1 turn earlier.

It's not a sure thing, but the AI tends to go to Monarchy pretty soon after Iron Working.

In our game, there are three AI who could easily go to Monarchy before ~T102:

Louis: last researched Monotheism on T89. Monotheism discounts Monarchy, and Hereditary Rule is Louis' favorite civic. He would give 90G for any of our techs on T95. So Monarchy >> Aesthetics/Alpha > anything else.
Elizabeth: Iron Working on T90. Would give 40G for any of our techs. Monarchy = Alphabet > anything else
Mansa Musa: CoL on T98. Too hard to predict, but Monarchy is as likely as anything else.
​

And there are three long-shots:

Isabella: Iron Working on T89. Theocracy is her favorite civic, and we gave her Writing on T87. So Theology seems a likely target.
Willem: Mathematics (T87). He's at war, so Construction >>> Alphabet/Monarchy > anything else.
Catherine: ??? Favorite civic is Hereditary Rule.​

Mao just got Calendar; Toku is useless; Shaka is slow and going to Construction.

----

Regardless of this, if we can sell Meditation to Louis and Elizabeth on T98, and trade it to Mansa for CoL, I still think we've done ok. Meditation keeps us 1 tech away from an AI's WFYABTA limit, which is 8 for Izzy, Mao, Louis, and Shaka (and we've traded/hope to trade for: Alpha, Archery, IW, Monotheism, MC, CoL and Monarchy... 7 total). It would be nice to pick up some of the non-urgent early techs (e.g. Calendar) from AI that aren't Mansa.

Getting Meditation to Elizabeth and Mansa would also steer them towards Philosophy. This has some risks for the Taj and Lib., but I'd still rather one of these get Philosophy than Tokugawa.

 

[Duckweed]

@Doshin

We traded techs with 2 groups which have not met with each other. In my memory, WFYABTA limit counts after they meet each other. That's how we could still trade with oversea AIs in continent map. Mansa is pretty much our only trade partner. I think you are aiming for Louis cash for a long time, I don't mind of trade Aes to him as what Kaitzilla suggested before.

@Kaizilla

I prefer to take another William's city before going for Shaka. Depends on where's Shaka's SOD resting, we might be able to DOW him with 3 stacks.

 

[Doshin]

@Duckweed

I think you are missing my point. Delaying the WFYABTA limit is a small positive in a worst case scenario, i.e. that we cannot trade for Monarchy, so researching Civil Service 1 turn sooner does not lead to an earlier civic switch. It is not a good reason in and of itself.

Let me focus on the main issues:

(1) Do we think that an AI (not Tokugawa) will research Monarchy by T105/6?

There is no good reason to push hard for CS if we will still lack Monarchy in 10 turns.

If we have ~700G banked when we turn on the research slider, we can sustain 100% research for ~8.5 turns. Running a max slider for ~8 turns would research Civil Service (= T106), and ~9 turns would research Meditation + CS (= T105; spend 2 turns on Meditation, 7 turns on Civil Service). [edit: all these #s are untested].

To me, it seems very likely that Louis is researching Monarchy now, and somewhat likely a second AI will go to Monarchy in the next 10 turns. However, AI research is very unpredictable. There are tendencies (I said Mao would research Calendar, and he did) but RNG plays too large a role to be sure.


(2) Why push hard to research CS one turn sooner?

Meditation costs 196B. We need a reason to justify the expense.

Economics reasons are slight: we can revolt 1 turn sooner (cities are smaller) and we will benefit from Monarchy/Bureaucracy 1 turn sooner.

Trading reasons are slight: Aesthetics/Alpha + Meditation = Monarchy; Construction + Meditation = CoL. The latter keeps Literature out of Mansa's hands. Meditation can be sold for , but we can sell other techs as well.

Civic switch timing reasons are big: we are at 10 cities now, and 12 cities is the limit for a 1-turn, 2-civic switch.

How might the next few turns shape up?

• Fish City: settled on T100 with an Argos whip, T101 without.
• Amsterdam: captured T102?
• The Hague: captured T105?
• Ice Ball Cities: assuming 3-pop whips of Athens and Sparta, the first would be settled on T105; second would be settled T106 (untested).

Researching CS on T106 would delay at least one Ice Ball city (= city 13/14); postponing a double civic switch even further would delay a second Ice Ball city (= city 15). The Hague (= city 13/14) has 3 Archers defending it, minus whatever Shaka kills, so will fall as soon as we have the requisite numbers to attack.

We could delay the ice ball cities by claiming the furthest locations first, to enable easier backfill, but this is risky without a Trireme.

----

Combine the economic, trading, and city settling/conquering reasons to understand what is gained by pushing for an earlier CS.

 

[Duckweed]

@Doshin

That's a good idea. But, there's still a potential problem -- whether we could get Monarchy in time.

It's likely that

William: Construction
Mao: Calendar
Toku: useless

Shaka is more likely going for Math->Construction, Izzy is too slow. Liz and Louis have good chance of researching it, or Mansa has chance of going for it after CoL.

As you can see, I pointed out the major problem in the 1st reply. I agreed with your proposal if the condition is met.

The question is what's the odd of getting Monarchy in time? From experience, there's very high chance that Immortal AIs should complete Monarchy ~T100. From current situation, probably ~50% since only Louis and Liz has fair chance of researching it now. Mansa also has good chance of going for it after CoL.

Spoiler :

However, if the luck conservation law works, you know the result. Just a joke

1 turn earlier CS plus 1 turn earlier of Fish island is good, but might not be better than the cost of Med. We could settle the furthest south island instead of eastern island.