SGOTM 14 - Plastic Ducks

[Deity player]

Here is the PPP for the next 2 turns. I didn't include chopping into the PPP (= the last of the 3 workerturns spent on prechopping and chopping), this is considered as automatic.

PPP

T88

- worker 1S of Washington silk chops
- worker 1W of Washington cows chops
- worker 3E of Washington moves 1E of Washington corn

- NY works wheet and hires 4 scientests
- Boston works rice+crabs+oasis+stone+gold and hires a scientist

T89

- the new warrior from Seattle goes to Washington
- the new warrior from Atlanta goes to NY
- the new workboat from Philly to Philly crabs
- missionary goes 2S of Seattle

- worker on Chicago silver builds a mine
- worker 1S of Chicago moves to Chicago silver and builds a mine
- worker 1E of Washington corn chops
- worker 1W of Washington corn moves 1S of Washington corn

- Washington works sheep+corn+cows+(mined grassland hill)+(mined plains hill)+[2x(lake)]+(forested grassland)
- Philly works [2x(forested grassland)]+crabs

T90

- pause, save, upload, report (since maybe, we should do some tech trades before converting to Confu)

 

[kossin]

Diplo and tech screen.

Spoiler :

I guess we can trade Math next turn for IW as our remaining warriors complete this turn.

Going to take a look to PPP.

 

[kossin]

Ok first thing:

-we should convert to Confu 1t earlier because on T90 we're getting a new farm and a net on the crabs, this is 3F wasted. We can trade for IW next turn as we'll be done with warriors I believe. Don't think we need Monarchy but Trading Math could mean getting Calendar via trade vs not.
-doing the above and working a 2F1C tile in Boston for 2 turn would give us size 7 there for T92

EDIT: Misfit Gypsy Nuts seem to have whipped a ton in one turn, the graph looks so weird

 

[Duckweed]

Here is the PPP for the next 2 turns. I didn't include chopping into the PPP (= the last of the 3 workerturns spent on prechopping and chopping), this is considered as automatic.

PPP

T88

- worker 1S of Washington silk chops
- worker 1W of Washington cows chops
- worker 3E of Washington moves 1E of Washington corn

- NY works wheet and hires 4 scientests
- Boston works rice+crabs+oasis+stone+gold and hires a scientist

6th citizen on 2F1C tile

T89

- the new warrior from Seattle goes to Washington to Phi
- the new warrior from Atlanta goes to NY to 3S of silk to fogbust the 2 coastal tile until Toto arrives
- the new workboat from Philly to Philly crabs
- missionary goes 2S of Seattle

- worker on Chicago silver builds a mine
- worker 1S of Chicago moves to Chicago silver and builds a mine
- worker 1E of Washington corn chops
- worker 1W of Washington corn moves 1S of Washington corn

- Washington works sheep+corn+cows+(mined grassland hill)+(mined plains hill)+[2x(lake)]+(forested grassland)
- Philly works [2x(forested grassland)]+crabs

Do the following trade, we want AIs having Math as soon as possible

Mansa: Math<->IW He is researching Math
Gandhi: Alpha+Math<->Monarchy+Poly He is researching Alpha

T90

- pause, save, upload, report (since maybe, we should do some tech trades before converting to Confu)

Convert to Confu

No need for another stop, just propose the PPP to T92 now.

 

[Duckweed]

Ok first thing:

-we should convert to Confu 1t earlier because on T90 we're getting a new farm and a net on the crabs, this is 3F wasted.

We will be wasting worker turns around Washington or the hammer overflow to cash.

We can trade for IW next turn as we'll be done with warriors I believe. Don't think we need Monarchy but Trading Math could mean getting Calendar via trade vs not.
-doing the above and working a 2F1C tile in Boston for 2 turn would give us size 7 there for T92

EDIT: Misfit Gypsy Nuts seem to have whipped a ton in one turn, the graph looks so weird

The graph does not show correctly now, they submitted twice for the same turn, the lower score one was the 1st submission. So actually they did massive lightbulb in one turn. Pretty good without Oracling CoL. Their score is a lot lower than us, which means much smaller empire size.

 

[kossin]

Interesting, I'm puzzled how their power and score both shot down and then up even then.

 

[Bebekija]

Interesting, I'm puzzled how their power and score both shot down and then up even then.

It may have been because of starvation, to pop GPs...There's only 2 ways to "lose" score, that is losing pop or cities.. I can't imagine it's the latter.

Anyway, good job Deity!
The most important point of the turnset was to grab the Mids, and we did it!

I can only take a look at the game and at your PPP tonight, but meanwhile I agree with Duckweed that you can make it up until T92 just as well.

 

[Deity player]

Ok I will write the PPP until T92 and follow your suggestions.

I will convert on T90 since otherwise we lose many hammers in Washington by smaller overflow.
EDIT: After testing a bit, the only difference between converting on T89 and on T90 (except for the loss of those 3F) seems to be that we get more cash in the former case and more hammer overflow in Washington in the latter case (the difference is ~10). I don't know much about the overflow bug so this is what I found out only from my testing. Could you confirm/disprove that? I guess that the hammers would be better but then we lose those 3F...

- the new warrior from Seattle goes to Washington to Phi

I intend to send the new warrior to Washington and then the warrior from Washington to Philly. By doing this, Philly will have its military protection 1T earlier (anyway, this is no big deal).

 

[Deity player]

PPP

Changes until T90.
I didn't include chopping into the PPP (= the last of the 3 workerturns spent on prechopping and chopping), this is considered as automatic.
Toto to Atlanta, then the warrior from Atlanta to NY.


T88

- worker 1S of Washington silk chops
- worker 1W of Washington cows chops
- worker 3E of Washington moves 1E of Washington corn

- Washington works sheep+corn+cows+(mined grassland hill)+(mined plains hill)+[2xlake]+(forested grassland)
- NY works wheet and hires 4 scientests
- Boston works rice+crabs+oasis+stone+gold+[riverside grassland 2F1C]

T89

- the new warrior from Seattle goes to Washington
- the new warrior from Atlanta goes 3S of silk to fogbust the 2 coastal tiles until Toto arrives
- the new workboat from Philly to Philly crabs (and improve the crabs on the next turn)
- missionary to Seattle (and spread Confu there on T91)

- worker on Chicago silver builds a mine
- worker 1S of Chicago moves to Chicago silver and builds a mine
- worker 1E of Washington corn chops
- worker 1W of Washington corn moves 1S of Washington corn

- Washington works sheep+corn+cows+(mined grassland hill)+(mined plains hill)+[2x(lake)]+(forested grassland)
- Philly works [2x(forested grassland)]+crabs

- if possible, do the following trades:
Mansa: Math<->IW
Gandhi: Alpha+Math<->Monarchy+Poly

T90

- Washington builds Confucian missionary

- worker NE of Washington moves 1E of Washington spices and builds a farm
- worker 1S of Washington corn chops

- convert to Confucianism

T91

- worker 1S of Washington silk builds a farm
- worker 1W of Washington cows moves 1S of Washington silk and builds a farm
- both workers 1E of Washington spices move 1N of Washington and build a farm

- Washington works sheep+corn+cows+(mined grassland hill)+(mined plains hill)+[2xlake]+[2x(riverside grassland 2F1C)]
- Philly works (forested grassland)+crabs+(riverside grassland)

T92

- pause, save, upload, report

 

[kossin]

Take the hammers, they're better than 3F.

 

[Duckweed]

I intend to send the new warrior to Washington and then the warrior from Washington to Philly. By doing this, Philly will have its military protection 1T earlier (anyway, this is no big deal).

As you said, the result will be the same, but you can use "go to" command to do the work in one time.

PPP

T90

- Washington builds Confucian missionary

- worker NE of Washington moves 1E of Washington spices and builds a farm
- worker 1S of Washington corn chops

- convert to Confucianism

You can't chop the grassland forest this this turn, otherwise the overflow hammers will be translate to cash. So only do the chop on the plain forest and build a farm on the grassland tile. However, it there's a spread in either Silk or canal city, produce a settler in Washington instead, you can do the chop then.

T91

- worker 1S of Washington silk builds a farm

Need to cancel the action, so that it can finish the chop next turn.

- worker 1W of Washington cows moves 1S of Washington silk and builds a farm

Moves to Washington or 1E of Washington, either for the road to dye city or the improvement in Phi, I'll leave the choice to others.

- both workers 1E of Washington spices move 1N of Washington and build a farm

- Washington works sheep+corn+cows+(mined grassland hill)+(mined plains hill)+[2xlake]+[2x(riverside grassland 2F1C)]

1 grassland forest + 1 Scientist or Spice, again, I'm undecided

- Philly works (forested grassland)+crabs+(riverside grassland)

T92

- pause, save, upload, report

Comments.

 

[Bebekija]

I checked the PPP and everything seems fine.
There are some worker actions that are not in it but since these actions started on T87, that must be the reason why you didn't put those in.

As for the choices left :
- I'd convert to Confu on T90 for the reasons mentioned by others,
- I'd do the road to dye rather than moving the worker to Philly, cause there are already 2 workers near Philly, and a road to dye can save up to 2 turns before settling
- I'd hire a scientist in Wash. I didn't do the maths, maybe the spice is better in terms of raw yield, but a few more GPP + higher GS chances, even for a single turn, may compensate for the lost food and hammer, which are pretty marginal by now.

 

[Duckweed]

A couple more comments on the PPP.

1. Toto to 3S of Silk and then the warrior there goes back to NY. Our 1st target is very likely the eastern witches, so our fleet is going to sail from SE corner and Toto will go to the battlefield instead of acting as a MP.

2. You can delay the missionary to Silk city by 1 turn, T92 spreading is good enough.

Edit: 3. T91: Boston: 1 worker goes to spice road, 1 worker goes to 1W forest, otherwise the settler in Washington will take long time.

I checked the PPP and everything seems fine.
There are some worker actions that are not in it but since these actions started on T87, that must be the reason why you didn't put those in.

As for the choices left :
- I'd convert to Confu on T90 for the reasons mentioned by others,
- I'd do the road to dye rather than moving the worker to Philly, cause there are already 2 workers near Philly, and a road to dye can save up to 2 turns before settling

There are 2 workers from silver city, which will be sent to dye city for sure. There are also 2 workers irrigating the farm in Boston, which could be sent to road to dye city. Phi has 3 citizens already and 2 are working on unimproved tiles, so 2 workers are not enough in short time. Another question is whether we want a LH in phi, if so, then we need to send 2 workers to improve the hill as soon as possible.

- I'd hire a scientist in Wash. I didn't do the maths, maybe the spice is better in terms of raw yield, but a few more GPP + higher GS chances, even for a single turn, may compensate for the lost food and hammer, which are pretty marginal by now.

Probably ~1% higher odd of GS, that's quite small. Since I have not run test to produce GPs, so I don't know whether the extra GPP is necessary.

The reason why I'm undecided on these 2 issues is that they need longer tests to make it clear. Anyway, these issues are not big deals now. Our luck to T104 is the most important.

 

[kossin]

The reason why I'm undecided on these 2 issues is that they need longer tests to make it clear. Anyway, these issues are not big deals now. Our luck to T104 is the most important.

Very true.

3 problems:

-where will Taoism appear
-what GPs will we get
-can we trade for Calendar

Otherwise, it's smooth sailing

 

[Deity player]

There are some worker actions that are not in it but since these actions started on T87, that must be the reason why you didn't put those in.

Yes.

2. You can delay the missionary to Silk city by 1 turn, T92 spreading is good enough.

In fact, it is enough to spread the religion in Seattle on T93 so I will move the missionary only on T91.

~~~~~~~~~~

Thanks for all your comments. I liked especially the idea of not chopping one of the Washington forests right away but building a farm there for 2T. (I always thought that after finishing the chop, we would have to spend another whole 5T by building the farm.) And this lead to the following...

We can convert to Confu on T89 and do the same trick with the worker 1W of Washington cows, i.e. not chop the forest on T90 but build a farm (or a road) there for 1T and chop the forest only on T91 (then no hammers would be converted to cash). This would lose 1 workerturn since we probably won't finish this farm in a short time but it would save 3F as kossin mentioned. So the question is whether we want to invest 1 workerturn to 3F which I think is fair enough. If you like it, I will change the plan.

~~~~~~~~~~

The red text refers to the two points which is Duckweed undecided about. I wrote my first sight preferences (and the alternatives are in brackets). I could do some longer tests yet but I would need several more days to do it.

Updated PPP

T88

- Toto goes 3S of Washington silk then the warrior which will be there by then goes to NY

- worker 1S of Washington silk chops
- worker 1W of Washington cows chops
- worker 3E of Washington moves 1E of Washington corn

- Washington works sheep+corn+cows+(mined grassland hill)+(mined plains hill)+[2xlake]+(forested grassland)
- NY works wheet and hires 4 scientests
- Boston works rice+crabs+oasis+stone+gold+[riverside grassland 2F1C]

T89

- the new warrior from Seattle goes to Washington and then the warrior from Washington to Philly
- the new warrior from Atlanta goes 3S of silk to fogbust the 2 coastal tiles until Toto arrives
- the new workboat from Philly to Philly crabs (and improve the crabs on the next turn)

- worker on Chicago silver builds a mine
- worker 1S of Chicago moves to Chicago silver and builds a mine
- worker 1E of Washington corn chops
- worker 1W of Washington corn moves 1S of Washington corn

- Philly works [2x(forested grassland)]+crabs

- if possible, do the following trades:
Mansa: Math<->IW
Gandhi: Alpha+Math<->Monarchy+Poly

T90

- Washington builds Confucian missionary
- missionary to Seattle (to spread Confu there on T92)

- worker NE of Washington moves 1E of Washington spices and builds a farm
- worker 1S of Washington corn chops
- worker 1S of Washington silk does not finish the chop but builds a farm and cancels his mission (due to the possible religion spread on the next turn)

- convert to Confucianism

T91

- one worker 1E of Washington spices moves 1N of Washington spices
- the other worker 1E of Washington spices moves to spices and builds a road
- worker 1S of Washington silk builds a farm and cancels his mission
- worker 1W of Washington cows moves 1E of Washington (alternative: to Washington)
- Washington works sheep+corn+cows+(mined grassland hill)+(mined plains hill)+[2xlake]+(grassland forest)+spices (alternative: + 1 scientist)
- Philly works (forested grassland)+crabs+(riverside grassland)

T92

- pause, save, upload, report

 

[Duckweed]

In fact, it is enough to spread the religion in Seattle on T93 so I will move the missionary only on T91.

The action of saving worker turns in silver city led to a side effect (I did not realize before) -- The missionary can only arrive canal city on T93. Therefore, if Tao holy city is in Canal city or there's a Tao spreading there, there will be ~10% odd of failure of spreading of Confu. I don't want to take the risk on silk city, which is further to reach. So, send missionary on T90.

We can convert to Confu on T89 and do the same trick with the worker 1W of Washington cows, i.e. not chop the forest on T90 but build a farm (or a road) there for 1T and chop the forest only on T91 (then no hammers would be converted to cash). This would lose 1 workerturn since we probably won't finish this farm in a short time but it would save 3F as kossin mentioned. So the question is whether we want to invest 1 workerturn to 3F which I think is fair enough. If you like it, I will change the plan.

If we send the worker to Phi (I incline for this choice now), we can mine the hill 1 turn earlier (2H, could lead to finish LH 1 turn earlier, means another 1F) and have at least 1 farm up 1 turn earlier (1F), so that worker is probably better than 3F.

Regarding the Washington tile management, working on spice gives 1H to settler, so if that 1H is important to pump settler 1 turn earlier, then on spice, otherwise, I'm fine to hire a scientist.

Kossin should have clear opinions on those issue as these affect something during his set.

 

[kossin]

No opinions really as I have not tested in a long time, too busy. I have some time this weekend however.

My feeling is that it doesn't change much either way, if we grab 1H it's very unlikely to change settler date and if we add 2~3 beakers with the scientist it's very unlikely to change the teching.

Anyone have the code for where holy cities will spawn? Could be useful.

 

[Bebekija]

No opinions really as I have not tested in a long time, too busy. I have some time this weekend however.

My feeling is that it doesn't change much either way, if we grab 1H it's very unlikely to change settler date and if we add 2~3 beakers with the scientist it's very unlikely to change the teching.

Anyone have the code for where holy cities will spawn? Could be useful.

There you go, enjoy!........

void CvGame::doHolyCity()
{
	PlayerTypes eBestPlayer;
	TeamTypes eBestTeam;
	long lResult;
	int iValue;
	int iBestValue;
	int iI, iJ, iK;

	lResult = 0;
	gDLL->getPythonIFace()->callFunction(PYGameModule, "doHolyCity", NULL, &lResult);
	if (lResult == 1)
	{
		return;
	}

	if (getElapsedGameTurns() < 5 && !isOption(GAMEOPTION_ADVANCED_START))
	{
		return;
	}

	int iRandOffset = getSorenRandNum(GC.getNumReligionInfos(), "Holy City religion offset");
	for (int iLoop = 0; iLoop < GC.getNumReligionInfos(); ++iLoop)
	{
		iI = ((iLoop + iRandOffset) % GC.getNumReligionInfos());

		if (!isReligionSlotTaken((ReligionTypes)iI))
		{
			iBestValue = MAX_INT;
			eBestTeam = NO_TEAM;

			for (iJ = 0; iJ < MAX_TEAMS; iJ++)
			{
				if (GET_TEAM((TeamTypes)iJ).isAlive())
				{
					if (GET_TEAM((TeamTypes)iJ).isHasTech((TechTypes)(GC.getReligionInfo((ReligionTypes)iI).getTechPrereq())))
					{
						if (GET_TEAM((TeamTypes)iJ).getNumCities() > 0)
						{
							iValue = getSorenRandNum(10, "Found Religion (Team)");

							for (iK = 0; iK < GC.getNumReligionInfos(); iK++)
							{
								int iReligionCount = GET_TEAM((TeamTypes)iJ).getHasReligionCount((ReligionTypes)iK);

								if (iReligionCount > 0)
								{
									iValue += iReligionCount * 20;
								}
							}

							if (iValue < iBestValue)
							{
								iBestValue = iValue;
								eBestTeam = ((TeamTypes)iJ);
							}
						}
					}
				}
			}

			if (eBestTeam != NO_TEAM)
			{
				iBestValue = MAX_INT;
				eBestPlayer = NO_PLAYER;

				for (iJ = 0; iJ < MAX_PLAYERS; iJ++)
				{
					if (GET_PLAYER((PlayerTypes)iJ).isAlive())
					{
						if (GET_PLAYER((PlayerTypes)iJ).getTeam() == eBestTeam)
						{
							if (GET_PLAYER((PlayerTypes)iJ).getNumCities() > 0)
							{
								iValue = getSorenRandNum(10, "Found Religion (Player)");

								if (!(GET_PLAYER((PlayerTypes)iJ).isHuman()))
								{
									iValue += 10;
								}

								for (iK = 0; iK < GC.getNumReligionInfos(); iK++)
								{
									int iReligionCount = GET_PLAYER((PlayerTypes)iJ).getHasReligionCount((ReligionTypes)iK);

									if (iReligionCount > 0)
									{
										iValue += iReligionCount * 20;
									}
								}

								if (iValue < iBestValue)
								{
									iBestValue = iValue;
									eBestPlayer = ((PlayerTypes)iJ);
								}
							}
						}
					}
				}

				if (eBestPlayer != NO_PLAYER)
				{
					ReligionTypes eReligion = (ReligionTypes)iI;

					if (isOption(GAMEOPTION_PICK_RELIGION))
					{
						eReligion = GET_PLAYER(eBestPlayer).AI_chooseReligion();
					}

					if (NO_RELIGION != eReligion)
					{
						GET_PLAYER(eBestPlayer).foundReligion(eReligion, (ReligionTypes)iI, false);
					}
				}
			}
		}
	}
}

Here are the interpretations of above code taken from the religion strategy guide.

Spoiler :

Once the player has been determined, that player's cities are each checked in turn. A random seed is added to the city's population, which then has two negative factors applied to it - an 8x bias against the city with the Palace, and a smaller bias scaled with the number of religions already present in the city.

So while you cannot assign the holy city as you might wish, you can arrange circumstances so that the biases work in your favor (keep the population high, keep the population in other candidates low, refrain from spreading religions to your preferred city, aggressively convert the other cities).

I don't have much time for testing these days but maybe tomorrow night I'll be able to check the different options left.
Though if Deity is ready to play before that, I don't know if it's worth delaying playing the turnset...

 

[kossin]

Ok that's good, basically the highest chances are cities with no religion and small, most likely Philly.

 

[Snaaty]

Ok that's not good, basically the highest chances are cities with no religion and BIG, most likely not Philly.

some minor corrections

religion gets founded most likely in biggest city with no reli so far. so if we want to avoid the holy city in a certain city (most likely our gp cities) we need to spread the religion there before founding tao. having 3 cities at pop 6-9 makes it 100% guarantied the religion gets founded there, since percentage goes up with each pop and it goes up quite a bit with each pop. already with a city at pop 4 and a city with pop 1 religion goes in 4 out of 5 times in the bigger city (havent looked at the code, am speaking out of experience)