SGOTM 21 - Plastic Ducks

Indeed, Dosh, Argos can be whipped at size 5, build a WB and still be faster than Sparta with GPP.
Circumnavigation allows us to get NFish one more turn sooner than before (on T99) and WB gets there in time. We also save one more worker turn (half of unnecessary road) because worker is not chopping faster than NFish expansion.




Trireme can come from Corinth. We could settle South fish first (T101) because we can settle it without Trireme (if we stumble upon a galley, settler can be unloaded and picked up by Trireme+galley 2T later if our galley loses the battle). When SFish border pop, warrior can be transported to the West for fogbusting. We can also settle Eastern city (T103).



Now off to deal with headache and then think about strategy. I generally agree that Willem should be dealt with first and then we can attack with 3 stacks (if Utrecht stack from late reinforcements can be called a stack).

Edit:

We could start thinking about Argos building workers for chopping the West and maybe for buying a task...

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

Click to expand...

But I am an engineer, not a scientist.
I wouldn't make a rule of thumb of this. It depends on how deep you are into a game. If you can't recall the map layout right, no use of thinking. Also, when you try to time builds from 5 cities, testing gets you there faster.

@Kaitzilla

Thanks for screenies!


CS

One turn faster CS equals 10 more net . That is really not much. Hammers come from food and Athens won't whip itself on its own, so nothing is lost there except of a negligible opportunity cost. We also lack Academy to make commerce worth more. Really no biggie to wait with it.

Edit:
Why do we want so many ice island cities? 13th city is already a net loss.
I recall that in SGOTM16 we didn't settle better cities than these and we were Cre/Ind, so cheap forges.
We'll also capture 4 cities.
Teching at this point is the most delicate since we are faster than the AI but lack for gold. If we are slow, win date will show.

Each island city is only a bit loss on commerce with GLH as they are not far away from capital. They will quickly be commerce positive when growing bigger. Those island cities can all start to contribute after 30 turns of settlement, which is still far away from end of the game. We settled almost all reachable islands in SGOTM16.

Hope to see your 1st version of test game tonight.

The time of DOWing is not fixed, if you see Shaka's SOD start marching (send the cata and the sword produced next turn to scout 1st), we could even DOW with 1 or 2 swords to take the chance of stealing. I don't think we need to bombard Amsterdam as it's slow. Even without Shaka's help, we could just sacrifice a couple of catas and then use the swords to take the city.

The micro is very interesting this turnset!
Those ice islands are a headache to plan for. Do we really want them?
Hmm, I think I see a way to do it and be safe from barb galleys rampaging too much.

I don't think we should whip Argos anymore.
When we do our golden age, we want max population there to run a lot of scientists.

shakabrade said:

CS

One turn faster CS equals 10 more net . That is really not much. Hammers come from food and Athens won't whip itself on its own, so nothing is lost there except of a negligible opportunity cost.

Click to expand...

This isn't true, as you can see in post #959. Please respond to these points if you are against the idea. I tried to be as balanced as possible in laying out the benefits and risks.

The Bureaucracy bonus applies to whips, btw.

The Real Game Mansa has Elephants!
And Willem's 3rd city appears to be north of Amsterdam.
3N2E of Amsterdam there is an orange tile that shouldn't be there unless there is a city.

Here is my best T105 test game attempt.
Doshin's road 1N1W of Iron City idea is great.
It gets Fish City settled on T101 instead of T102 all by itself!

Used a galley chain @3S2E of Thebes to get Athen's warrior down to a south tundra island immediately.
The catapult didn't leave Athens until T97 and the Thebe's phalanx didn't leave Athens until T100.
Also, the warrior from Corinth had to come down to garrison Athens eventually.
The worker that unloaded in Athens T96 went north to chop the forest north of Corinth. Then went even further north to chop Fish City.
Corinth made Worker.
Iron City built Granary->Trireme->started a Galley.

I like this setup because it allows our galley to explore east after dropping off the 1st tundra fish island settler and there is nowhere for barb galleys to spawn between our cities (1 tile dark areas ).
I favor the closer tundra fish island vs. far east because it is 5 tiles closer to us and is easier to defend.

I've thought about a naval invasion of Elizabeth after we kill Willem instead of settling Arctic fish cities+killing Shaka, but capturing the Temple of Artemis is a bad idea in the short term.

Doshin said:

This isn't true, as you can see in post #959. Please respond to these points if you are against the idea. I tried to be as balanced as possible in laying out the benefits and risks.

The Bureaucracy bonus applies to whips, btw.

Click to expand...

We should be able to get Civil Service by T104 if we overflow tech Meditation or T105 if we don't ya.
All depends if Mansa can get Code of Laws done in 3 turns.
My iceball city was settled T102. Hmm

shakabrade said:

...

We could start thinking about Argos building workers for chopping the West and maybe for buying a task...

Click to expand...

Another task would be nice, but I'd rather keep the extra worker for now.
I can't wait to get those western forests. (And it will be terrifying with Impis prowling around waiting to pounce like Woodsman II units).
Plus, Amsterdam probably won't last long enough for a gift worker to get there in time.

shakabrade said:

...
Edit:
Why do we want so many ice island cities? 13th city is already a net loss.
I recall that in SGOTM16 we didn't settle better cities than these and we were Cre/Ind, so cheap forges...

Click to expand...

It certainly wouldn't be worth it normally, but it is a long game and we got The Great Lighthouse, so it's probably worth doing.
That is, unless barb galleys start pillaging fish causing us enormous losses!
Imagine an iceball city with its fish pillaged.
Useless dead weight!
If all goes well, the worst feeling will be at 3 population when the arctic cities grow at 3 per turn trying to 2-pop whip a granary at Size 4.

Kaitzilla said:

The Real Game Mansa has Elephants!
And Willem's 3rd city appears to be north of Amsterdam.
3N2E of Amsterdam there is an orange tile that shouldn't be there unless there is a city.

Spoiler :

Click to expand...

Ha. Nice catch. That is the first time I've ever seen this screen provide useful information (I wonder if this would provide a lazy way to find the "fear my Archer! / Axeman! / Modern Armor!" info...).

----

The save looks good... except...


How easy or difficult was it to reach Civil Service on T105? Is T104 too much too soon? We might be able to capture Amsterdam sooner than T105, which would delay any island cities.

We can be flexible with settling the islands. If Monarchy is unavailable on T104/5/6, we could push further east and take a less than optimal site. The economic hit comes sooner, but the city gets set up sooner in turn (a long wait for Granary + LH + CH).

What were troop numbers like around Knossos from T98 onwards?

Woops! I was wrong about T105, not sure what I was thinking/counting.
Plus, I didn't do the Meditation overflow in my test game at all.

We can have Civil Service done in 5 turns if we tech Meditation and 6 turns without.
The Real Game has a slightly stronger tech rate than the test game.
On T95, real game is +197 at 100% slider and test game is +181 at 100% slider (more trade routes, maybe a few of the worked tiles are different ).

If Mansa gets CoL traded to us on T98, then if we overflow tech Meditation T97 we will get Civil Service on T103.
Meditation is a 130 tech. Not sure how much gold we'd have to sell it for to come out ahead.
We could wait 1 extra turn to capture Amsterdam or settle an iceball city on T103 if it saves us 1 turn of Anarchy ya.

The save looks good... except...

Click to expand...

Here is the flow of military units from my test game run attempt:
1W of Utrecht is called the DOW LINE, because it is controlled by Willem in the Real Game.

DOW LINE (The turn a unit crosses into it)
T97 - 1 sword
T98 - 1 sword, 1 catapult
T99 - 2 swords, 1 catapult
T100 - 2 catapults
T101 - 1 sword
T102 - 1 catapult
T103 - 2 swords, 1 chariot (will hopefully be medic someday)
T104 - 2 swords, 1 catapult
T105 - 1 catapult

Kaitzilla said:

If all goes well, the worst feeling will be at 3 population when the arctic cities grow at 3 per turn trying to 2-pop whip a granary at Size 4.

Click to expand...

We can use OF from Workers to set these cities up.

Grow on a Granary to 20/22F @ size 1, and then start a Worker. Build this for 8 turns to 48H, then grow to size 2 and instantly whip. The city is now at size 1, 3/22F, with 5H in the Granary and 24H of overflow. Regrow to size 2 and whip the Granary.

We could also repeat the process with the Lighthouse.

Set up times (all untested):


With a Granary, a city needs 11/12/13F surplus to regrow from size 1 ---> 4 (i.e. average of 6F every 2 turns needed).
With a Lighthouse, the city has +6FPT surplus from the Fish. Coastal tiles are neutral.

Development of (C) is about 6 turns slower than (B).
Any city that builds a Worker stops growth for 2 turns. 4 turns are then needed to regrow the 2 whipped pop.
The actual output of (C) is therefore roughly the same as that of (B).

We could count food/hammers to be more exact, but it seems unlikely that we will want 8+ Workers from 4+ island cities when these cities lack infrastructure. So 1 Worker per island city (for Granary OF) seems best to me.

Kaitzilla said:

The micro is very interesting this turnset!
Those ice islands are a headache to plan for. Do we really want them?
Hmm, I think I see a way to do it and be safe from barb galleys rampaging too much.

I don't think we should whip Argos anymore.
When we do our golden age, we want max population there to run a lot of scientists.

Click to expand...

Hmm, we'll have Monarchy and a plenty of food to grow quickly. We whip at size 5, so not so much food is lost. Happiness (Archers) can come from Sheep city. They'll be useful later for more whipping. This is probably after this turnset.

Doshin said:

This isn't true, as you can see in post #959. Please respond to these points if you are against the idea. I tried to be as balanced as possible in laying out the benefits and risks.

I am against the one turn sooner CS since I really can't see the benefit. It'll only delay our troops for no substantial benefit. CS gives 16 and costs 6 gpt. Opportunity for trading anything useful as Meditation is worth the wait since we can't run the full slider for a long time and don't have the multipliers. Not having CS is not hurting our potential since we don't relly have any worth mentioning at the moment.

The Bureaucracy bonus applies to whips, btw.

Click to expand...

I'll quote myself to prove that I was aware:

Hammers come from food and Athens won't whip itself on its own, so nothing is lost there except of a negligible opportunity cost.

Click to expand...

Doshin said:

We can use OF from Workers to set these cities up.

Grow on a Granary to 20/22F @ size 1, and then start a Worker. Build this for 8 turns to 48H, then grow to size 2 and instantly whip. The city is now at size 1, 3/22F, with 5H in the Granary and 24H of overflow. Regrow to size 2 and whip the Granary.

We could also repeat the process with the Lighthouse.

Set up times (all untested):

(A) 30 turns. City @ size 2. Granary + Lighthouse. 17/24F.
(B) 31 turns. City @ size 2. Granary + Lighthouse + 1x Worker. 18/24F.
(C) 35 turns. City @ size 1. Granary + Lighthouse + 2x Worker. 19/22F.​

With a Granary, a city needs 11/12/13F surplus to regrow from size 1 ---> 4 (i.e. average of 6F every 2 turns needed).
With a Lighthouse, the city has +6FPT surplus from the Fish. Coastal tiles are neutral.

Development of (C) is about 6 turns slower than (B).
Any city that builds a Worker stops growth for 2 turns. 4 turns are then needed to regrow the 2 whipped pop.
The actual output of (C) is therefore roughly the same as that of (B).

We could count food/hammers to be more exact, but it seems unlikely that we will want 8+ Workers from 4+ island cities when these cities lack infrastructure. So 1 Worker per island city (for Granary OF) seems best to me.

Click to expand...

Wow, great and complicated .
Edit: Because of transporting the workers and ability to be messed up.


Other thoughts:

I didn't have problems settling the islands sooner. There is a small chance for only one turn to run into a galley, a case in which we can unload a settler not to risk it, and then fight with more than 60% odds of winning. Or postpone a city for 2T and cover behind a Trireme from Corinth.

Regarding too many island cities. I am talking about them slowing our tech pace. I have noticed in some games how Kakumeika managed to tech much better with much smaller empire (no shi* Sherlock...). Although we had 1 tech per turn in the end-game (problem is that is a limit), we couldn't compensate for our slower early, and especially mid-game teching. Island cities will slow our mid-game research and also force us to take smaller Lib prize. If those were double-fish cities, I wouldn't object, but these are slow and can't really whip all that much, maybe galleons later. At least they are Fish cities, though. Our aggression is beneficial when we take developed land and cities. The point is that expansion needs to be carefully thought of even for a long game.

I can't recall the game, but they could have won it with more aggressive approach and less mistakes. Or was it TSR?

Anyway, if you buy it, you buy it, if not, I'll accept.


DOW LINE. Love it!

Work, eat, test, sleep,...

Doshin said:

We can use OF from Workers to set these cities up.

Grow on a Granary to 20/22F @ size 1, and then start a Worker. Build this for 8 turns to 48H, then grow to size 2 and instantly whip. The city is now at size 1, 3/22F, with 5H in the Granary and 24H of overflow. Regrow to size 2 and whip the Granary.

We could also repeat the process with the Lighthouse.

Set up times (all untested):

(A) 30 turns. City @ size 2. Granary + Lighthouse. 17/24F.
(B) 31 turns. City @ size 2. Granary + Lighthouse + 1x Worker. 18/24F.
(C) 35 turns. City @ size 1. Granary + Lighthouse + 2x Worker. 19/22F.​

With a Granary, a city needs 11/12/13F surplus to regrow from size 1 ---> 4 (i.e. average of 6F every 2 turns needed).
With a Lighthouse, the city has +6FPT surplus from the Fish. Coastal tiles are neutral.

Development of (C) is about 6 turns slower than (B).
Any city that builds a Worker stops growth for 2 turns. 4 turns are then needed to regrow the 2 whipped pop.
The actual output of (C) is therefore roughly the same as that of (B).

We could count food/hammers to be more exact, but it seems unlikely that we will want 8+ Workers from 4+ island cities when these cities lack infrastructure. So 1 Worker per island city (for Granary OF) seems best to me.

Click to expand...

Nice thought. It's a more efficient way of generating hammers, with the loss of some food and commerce. It's good if we need those workers, but might not since they are quite late and we are going to capture some workers in the war. From experience, we usually don't need more workers after starting the war.

shakabrade said:

Regarding too many island cities. I am talking about them slowing our tech pace. I have noticed in some games how Kakumeika managed to tech much better with much smaller empire (no shi* Sherlock...). Although we had 1 tech per turn in the end-game (problem is that is a limit), we couldn't compensate for our slower early, and especially mid-game teching. Island cities will slow our mid-game research and also force us to take smaller Lib prize. If those were double-fish cities, I wouldn't object, but these are slow and can't really whip all that much, maybe galleons later. At least they are Fish cities, though. Our aggression is beneficial when we take developed land and cities. The point is that expansion needs to be carefully thought of even for a long game.

I can't recall the game, but they could have won it with more aggressive approach and less mistakes. Or was it TSR?

Click to expand...

While other teams have regretted that they have not understood the power of land, our team members have started to think in the opposite way. Settling those island city cost no more than 5cpt @size1 and will be less when it grows bigger since each pop gives 2C while the cost is less. For 30 turns, the commerce loss from those city will be probably 50~100C, all those cost won't affect the critical techs by more than 1~2 turns if you think about our tech rate 40 turns later and the cost of those techs.

Even though I'm very out of touch with the game, I feel the need to weigh in a bit. Feel free to disregard my post.

* I'm surprised buying the remaining hints has not been suggested for a while. I've even suggested ways to get our gold back the following turn. At least the ones for Cathy so we don't shoot ourselves in the foot.

* More cities is better with GLH, I agree. But too many isn't good either.

Doshin said:

shakabrade and kossin will enjoy seeing the number of 1-2 tile ice islands.

Click to expand...

kossin said:

Unless there are many AI traderoutes to obtain, settling several ice-ball-GLH-cities is not worth it with non-Organized. Well, not now at least.

Click to expand...

Too many of these slows down gpp production and unit production. While the commerce hit at first may be undesirable and the long-term effects are well known to the team, it's also no news that the land approach only catches up around T190~200.

In this case, we see some other teams went for war even earlier, so I'm not sure how they handled their economy.

Maybe somewhere in the middle with some focus on generating gold?

~~~

Again sorry to be useless, but it's all been unexpected around here...

Hello, kossin, hope that everything is ok.


Land is power, but one tile islands aren't land.
You can't workshop them, chop into infra too speed them up or use workers to make them worth more. WBoats cost hammers too. They take too long to grow and add maintenance to all cities and not just their own (number of cities). It gets worse until you have around 30 cities, a case in which number of cities maintenance is maxed. They don't build wealth. If they had more food, I wouldn't mind settling them all, but now I think we could stop at city 12, do a 1T switch to Monarchy and Bureau. If we were financial or had Colossus, I wouldn't mind too. But we don't.

And it is not that we haven't expanded...



Is there any chance that border pop from Sheep would actually open up a passage to Catherine?

It is very easy to go into the Worldbuilder, add eight 1-tile island cities at random, and judge their economic impact. It is ~2CPT per city, at size 1, with exclusively inland trade routes. And we are not settling these aggressively (zero 1-tile islands so far, 3-pop whips in Thebes and Athens are being used to produce the first 2 Settlers, we are not hurrying to settle these at the cost of more useful builds).

Workers and Settlers have been whipped to manage unhappiness. We look to be the 2nd quickest team to go to war (inc. Phoenix Rising). We have not whipped away improved tiles, and all new cities have claimed new resources. Without Monarchy or Pacifism or a religion, running specialists is inefficient. If we bulb our way to Paper/Education/Printing Press/Astronomy: what then? The other top teams that have gone this route in previous SGOTMs have burned out upon reaching the Renaissance. SGOTM 19 was the most recent example.

With 9 AIs, and above all Mansa, it is also very easy to pick up classical and medieval techs through trading or espionage. Tech trading is the single most broken feature in the game.

We enter expansion mode prior to spreading religion, then build infrastructure (Libraries + Universities + CHs, Forges & Observatories in some cities) and grow GP farms to the happy cap (why I prefer not to whip Argos). The first Golden Age is timed with the maturation of GP farms and large cities. 1-tile islands whip the initial navy (4–8 Galleons) while mainland cities stay big throughout the Golden Age.

kossin said:

* I'm surprised buying the remaining hints has not been suggested for a while. I've even suggested ways to get our gold back the following turn. At least the ones for Cathy so we don't shoot ourselves in the foot.

Click to expand...

I wouldn't mind doing this. We absorbed most of the gold back that we gave to Willem before.

shakabrade said:

Is there any chance that border pop from Sheep would actually open up a passage to Catherine?

Click to expand...

I don't think so.

Best thing before Optics (probably) would be to get the city's borders to 100culture, load a 5XP Chariot onto a Galley, and move the Galley to the culture's edge. Then give the Chariot Sentry.

shakabrade said:

But I am an engineer, not a scientist.
I wouldn't make a rule of thumb of this. It depends on how deep you are into a game. If you can't recall the map layout right, no use of thinking. Also, when you try to time builds from 5 cities, testing gets you there faster.

Click to expand...

I really hope that you did some test, read, and think carefully about my post before arguing more.

10 cities for current empire, 75gpt



settle 20 more island cities. 30 cities in total,@size1 26gpt, lose 49/20=2.5gpt per city. 5 turns to grow to size 2 with a Fish



30 cities @size2 57gpt, lose 18/20=0.9gpt per city 6 turns to grow to size 3



@Size3 about even of commerce.

Each island city cost 5*2.5+6*0.9 = 18C + 130H (1 settler + 1WB) ~33 turns later, when the city grows to size 3 with granary+LH, it starts to be commerce positive and produce 6F=12hpt. Each island city takes 40~45 turns to pay its cost. Therefore any island city with a Fish settled 45 turns before the end game is a gain in all aspects.

Keep in mind that earlier cities costs less or even a gain on commerce.

I have not convinced kossin about expansion in SGOTM19, so I won't try further again.

Doshin said:

Set up times (all untested):

(A) 30 turns. City @ size 2. Granary + Lighthouse. 17/24F.
(B) 31 turns. City @ size 2. Granary + Lighthouse + 1x Worker. 18/24F.
(C) 35 turns. City @ size 1. Granary + Lighthouse + 2x Worker. 19/22F.​

Click to expand...

That's brilliant!
Yes, Plan B is a genius way to get 1 free worker out of the iceball cities.
The excess workers can then be used to buy hints

Doshin said:

I wouldn't mind doing this. We absorbed most of the gold back that we gave to Willem before.

Click to expand...

Any more gold we give to Willem probably won't be recovered.
The amount of an AI will trade depends mostly on EmpirePopulation*TurnsKnown.

After we take Amsterdam and The Hague, I expect Willem to trade us 0 for a very very long time.

If Willem has 1 city and we wanted to use a spy to steal back , and he doesn't trade it away, we could get it back that way though.
I'd estimate we'd need the amount of gold x2 worth of espionage points to steal it with 50% wait bonus for spy.

A gifted worker might be more easily recovered.
Ideally, Willem and Shaka are still at war when we sign a cease fire with Willem.
Then when we DOW Shaka, we can open borders immediately with Willem and eventually sail back his own workers north to his new capital for hints.

shakabrade said:

Is there any chance that border pop from Sheep would actually open up a passage to Catherine?

Click to expand...

If I had to guess where Catherine was, I'd say far to our north. Or far to our east.
Isabella is in one and Catherine is in the other.
2nd border pop from Iron City? hmm

I don't know honestly. Does a +1 vision chariot Doshin mentioned make a difference in getting across with the Iron City 100 border pop?

Kaitzilla said:

Any more gold we give to Willem probably won't be recovered.
The amount of an AI will trade depends mostly on EmpirePopulation*TurnsKnown.

After we take Amsterdam and The Hague, I expect Willem to trade us 0 for a very very long time.

If Willem has 1 city and we wanted to use a spy to steal back , and he doesn't trade it away, we could get it back that way though.


A gifted worker might be more easily recovered.

Click to expand...

Do you think we could gift the gold on T95 and recover it on T96?

---

Also:

My amazing (almost definitely impractical/probably counter-productive ) way to gift Workers with the loss of just 1 turn per Worker:

Heh, I think the worker gift has to be done in Willem's capital.
We'd need a 2nd naval invasion to overwhelm the troops he'd have defending the city to do it in 1 turn.

Doshin said:

Do you think we could gift the gold on T95 and recover it on T96?

Click to expand...

No. We just traded for Willem's 110.
He won't have more for us until his population increases a lot or many more turns have passed.

The calculation is ((AI Total Population * (Turns since you met the AI + 10) * iMaxGoldTradePercent)/100) - total of all gold that AI traded to player previously

Click to expand...

However, now that you mentioned it, does gifting to the AI count against "- total of all gold that AI traded to player previously"?
It certainly needs testing!

I will check it in the test game now.

**Edit**
Nope, test AI after a big trade was only offering 10 even though I made sure it had 5000
Gifted it 100
Still only willing to trade 10

If we gift Willem on T95, we won't get it back T96.

If Mansa would just trade us copper, we could chop out the Colossus.
That would make the arctic fish islands more attractive yes?

It's not like we are going to rush for Astronomy now, so it might be worth it.

Regarding settling more cities:

It is much easier to WB 20 island cities than to produce them while also producing other needed stuff. Research is crawling, can you maintain a stack in the enemy territory composed of 1 movers who take cities slowly and still research? What do we get out of Lib with that? Is our military progress slower due to producing settlers and WBs? Also, maintenance would be higher because you can't settle 20 cities that close in a real game. If we run more expensive civics than in the test, add more turns to payback. With Mids and cheap libraries, you can escape most of these questions, but we don't have them. Yet.

The explanations you give are simply too linear. We are talking about trying to optimize something here. With those hammers and wealth, you can either get more army and open another front, invest wealth into research and get better units and other perks. Then you compare with that. 45 turns of payback can be doubled. What happened in the end game for us was getting beakers for more than 1Tech per turn and we couldn't return the turns wasted before, simply due to this limit.


Settling city 12 and 13 is still fine. But without Fin and Org or Colossus or Free trade civic, we can't stretch much more. 12 cities if we want 1T switch to Bureau and HR.

When cities become able to build a lot of wealth soon after being settled, we can spam them as we please. To get to that point, you need techs like Communism, Corporation, Medicine, Railroad. Anyway, if you can't see that following axioms does not always lead you to the right side, I don't intend to change that. Whatever the team decides, I am fine with that in the end.


Anyway, the test with lots of issues and some mismanagement:




Nighty night.

Edit: I added 13th city in the screenie via WB and should be ignored.