happiness quirk...

[Marquis de Sodaq]

While punishing the AI in the Delunda Sunt challenge (in the Strategy forum), an odd this happened - odd, to me at least, because I've not seen it before. I'd captured many cities and was in monarchy, and the martial law worked in reverse! This is while playing as the Seleucids, with the Hanging Gardens still in effect. I had approximately 30 cities.

In some size 2 cities, my citizens were actually happier while their sons and husbands were off to war than while they were stationed at home. With a chariot and phalanx defending the town, there was one happy citizen (HG), and one unhappy citizen. When the chariot rode off to slay Celts, the unhappy citizen became content. When the chariot died, the unhappy citizen became content. Apparently these were not very popular units...

This happened in several cities. Due to the race against the clock, I didn't spend much time testing. Does anyone have an explanation?

------------------
"There is no fortress impregnable to an ass laden with gold."
-Philip of Macedon

[Scouse Gits]

You don't play much ICS do you?

Once the two black face situation applies moving troops out of a city will increase its happiness (with the HG). There is probably a definitive explanation around somewhere, but the reality makes for lots of empty cities in most ICS games

------------------
Scouse Git[1] -- git1@scousers.net

"Staring at your screen in horror and disbelief when you open a saved game is one of the fun things of a succession game " - Hueij
"The Great Library has been built!"
"A short cut has to be challenging,
were it not so it would be 'the way'." - Paul Craven

[Marquis de Sodaq]

I didn't think 30 or so cities was ICS... I thought this meant scores more than I had. Anyway, it didn't seem like more than a normal game. Maybe it's just perception, but I think I usually have 30 to 40 cities in my normal quasi-perfectionism. Not intended, you know - this advanced tribe needs a port, how about one 2 squares from their capitol, oh and another on this gold mountain, and so on. This was the first time I'd seen the "happy to see them off" effect. Maybe because of all those colluseums I normally build...

This has been discussed in threads, I even did a search before posting. Just didn't think it applied - how many cities do ICSers settle in a game?

And if it turns out I ICS, have 8000g in the coffers, AND build everything under the sun, discover a new advance every other turn, - maybe I'm a better player than I'd thought...

------------------
"There is no fortress impregnable to an ass laden with gold."
-Philip of Macedon

[East Street Trader]

Marquis

I'm probably going to get this wrong but I'll have a go.

It's to do with the appearance of the black hatted double unhappy brigade.

30 cities in Monarchy is well enough to start getting these gentlemen. DaveV has helpfully posted (more than once) an authoritative set of figures which tell you the numbers of cities which result in the first citizen being unhappy and, later, double unhappy. From memory, it's 8 cities before the first citizen is unhappy and then, once you get to the number which triggers double unhappies some cities get to have double unhappies but other cities don't.

Anyway, the point about the double unhappies is that there is indeed a quirk in the game such that HG (and 2 luxuries also I think) has the unexpected effect of moving the double unhappy straight to happy - missing out both single unhappy and content.

So if the first citizen is a black hat then with no military units in residence HG makes the first citizen a happy, the second citizen stays unhappy but that doesn't send the city into riots.

If you leave one unit in the city then the game takes the martial law effect of that unit into account first and the double unhappy is lifted to being merely unhappy. Now HG operates on that citizen and only lifts it from unhappy to content. So you finish up with one content citizen and one unhappy citizen = riot.

That is what produces the counter intuitive result that if you move the military unit out - or it gets killed in situ, then the riot stops.

[Laertes]

This accords with my recent experience. I decided to try a game on ICS, and have noticed the same effect as described by EST. It is validated when you look at the "Happy" view in the city information screen. The HG has an impact, but not the troops.

As for city numbers, in early AD, I have got about 40 cities (2 conquered. I am not purist ICS - I think I take too many steps before starting a city with a new settler, and I have had to move onto 4 new islands as I started isolated. I am now getting a bit bored with continually building settlers and trying to find places to settle, and am tempted by some very nice city sites to actually stop at 40 an develop like mad.

[DaveV]

Study the "happy" window in the city screen to learn about how citizens are made happier. The order in which the various measures are applied is a key element of the process. That order is:

1. Luxuries
2. Improvements
3. Martial Law
4. Wonders

Two luxuries will convert a very unhappy citizen to happy, as EST reported. Elvis or some luxuries will allow your city to have a happy first citizen regardless of the number of garrison troops. In the case of HG, though, a martial law unit may change that very unhappy citizen to merely unhappy, and HG will now change him to content instead of happy.

As to the troops' effect not being shown in the happiness window: I think that display isn't always exactly right. I used to think that happiness measures became somehow less effective, but I think it's just a display problem. One martial law unit can change a very unhappy into an unhappy, but he cannot be changed to content by martial law alone.

Some examples of how I keep size two cities happy:

Red, blue: one (or more) units (or HG) changes to blue, blue.
Red, red: two (or more) units change to blue, blue. One unit changes to blue, red, then HG changes to blue, blue.
Black, red: two (or more) units change to red, blue. HG will change to blue, blue. One unit changes to red, red; HG will change to blue, red (riots). With no units, HG will change to cyan, red (city is content).
Black, black: two (or more) units change to red, red. HG will change to blue, red (riots). One unit changes to black, red. HG will change to cyan, red (content). With no units, HG will change to cyan, black (content).

[SlowThinker]

The happiness stuff summary:
I have tried to sort out things about the happiness, could more experienced players look if they are correct?

---

The effect of happinness items is provided in steps. There are steps of three types:
two positive:

A. "luxury" type:
	if C or VU exist then		C, VU -> H
else	if U exists then		U -> C

this is probably a bug, the idea was probably

	if C exists then		C -> H
else	if U exists then		U -> C
else	if VU exists then		VU -> U

B. "temple" type:
	if VU exists then		VU -> U
else	if U exists then		U -> C

and one negative:

C. "unit in the field" type:
	if H exists then		H -> C
else	if C exists then		C-> U

(VU - very unhappy, U - unhappy, C - content, H - happy)

---

"luxury" type:		luxuries, Courthouse and Palace in democracy, Hanging gardens, Cure for cancer
"temple" type:		improvements, Bach, Shakespeare, martial law (without VU->C)
"unit in the field" type:		units in the field

The order is done in accordance with the city window:
1. specialists are taken (they are chosen from H and VU preferentially),
2. luxuries,
3. improvements (incl. Michelangelo)
4. units (martial law or unit in the field)
5. wonders

Individual items have multi-step effect: For example temple with mysticism causes that the step B. is executed two times (and so two VU are not changed to U but one VU is changed to C).

Shakespeare is executed (city size + 1) times (i.e. needn't turn all citizens content).
Luxuries may do max. <city size> steps. Hanging Gardens may do max. <city size> steps (in the city where it was built).
The martial law may do max. <city size> steps and max. once on each citizen (in other words the martial law can't change a citizen from VU to C, only to U). Sometimes the happiness analysis window shows no effect by the martial law (if at least one VU is present ?), but this is not true.
Bach has no effect if at least 2 VU are present

Many units in the field causes all citizens to be U under republic, but always let one citizen C in democracy.

Specialists are always C, no effect of happiness items.

This post has been edited several times.

[SlowThinker]

Originally posted by DaveV
As to the troops' effect not being shown in the happiness window: I think that display isn't always exactly right. I used to think that happiness measures became somehow less effective, but I think it's just a display problem. One martial law unit can change a very unhappy into an unhappy, but he cannot be changed to content by martial law alone.

I am sorry, I didn't read this carrefuly. I will rebuild my last post probably.

[SlowThinker]

The long post is finished now.

[SlowThinker]

An addition: luxuries may do max. {city size} steps.

Do you know the order the Wonders takes effect? By the order in the rules.txt file?
It would be good if HG and Cure for C. acts as first so that black hats remain black.

Edited: the first sentence

[DaveV]

Mike's is treated as a cathedral and applied at the "temple" stage. All other happy wonders are applied in a fourth, "wonder" stage. This is why you need to move troops around to keep a Very Unhappy for HG to convert to Happy. It's also why JSBC is so beloved by democratic players: it will act to cancel out the unhappiness from one unit in the field, while MC's effects will be undone by units in the field.

[SlowThinker]

Originally posted by DaveV
It's also why JSBC is so beloved by democratic players: it will act to cancel out the unhappiness from one unit in the field, while MC's effects will be undone by units in the field.

But the placement of the MC to the "temple" stage has a negative effect only if a part of MC's effect is redundant, hasn't it?

[DaveV]

Originally posted by SlowThinker
But the placement of the MC to the "temple" stage has a negative effect only if a part of MC's effect is redundant, hasn't it?

An example: size 7 city under democracy, deity level, number of cities is under the first riot factor threshold. 1 content, 6 unhappy. 6 cups convert to 1 happy, 1 content, 5 unhappy. Temple + MC convert to 1 happy, 6 content. A unit in the field converts to 1 happy, 4 content, 2 unhappy; city riots.

But, if you add JSBC, its effects are added in after the penalty for having units afield, and it will change the city back to 1 happy, 6 content.

[SlowThinker]

Originally posted by DaveV
An example: size 7 city under democracy, deity level, number of cities is under the first riot factor threshold. 1 content, 6 unhappy. 6 cups convert to 1 happy, 1 content, 5 unhappy. Temple + MC convert to 1 happy, 6 content. A unit in the field converts to 1 happy, 4 content, 2 unhappy; city riots.

But if the MC would act under the "wonders" stage in place of the "temple" stage, then the result would be equal in your example:
temple converts to 1H 3C 3U
unit in the field 0H 3C 4U (I disagree with you, units in the field turns H to C preferably. But it has no meaning in our example)
MC converts to 0H 6C 1U, city riots.

Only if the city would be of size 6, then 1 positive step of MC would be lost in the "temple" stage (the city would riot) but wouldn't be lost if used in the "wonder" stage.

[SlowThinker]

Originally posted by DaveV
Mike's is treated as a cathedral and applied at the "temple" stage. All other happy wonders are applied in a fourth, "wonder" stage.

But works Gardens and Cure for Cancer before Bach and Shakespeare?

[SlowThinker]

I have run some tests to find out which cities get additional red/black hats due to too many of cities:

	despotism
number of cities	1	2	3	4	5	6	7	8	9	10	11	12	13	14
					4	3	2	1	+8	+7	+6	+5
							7	8	9	10	11	12
							6	5	+4	+3	+2	+1
											10	9
											+11	+12
	monarchy
number of cities	1	2	3	4	5	6	7	8	9	10	11	12	13	14
							6	5	4	3	2	1	+12	+11
										9	8	7	+6	+5
										10	11	12	13	14

How to read the table:
For example, under despotism, when you build your 10th city, city nr. 7 gets black hat, city nr. 10 gets red hat, city nr. 3 gets black hat.

It looks the system is understandable.
Who will add other governments?

[SlowThinker]

I take it back! In my "Campo" game it didn't work.

It might be map-dependent..., or year-dependent...

I did my tests with small map, deity, everything in one turn via cheat menu. I changed the build order and it worked...

[Scouse Gits]

It is certainly map-size dependent ...

SG[1]

[DaveV]

Another thing to consider is city sizes. A hard-core ICSer may well have dozens of cities with none larger than size two. I've found the relationship between riot factor and number of cities to be very predictable under those circumstances. I think that a city of a certain size (size 3?) counts as two cities for riot factor calculation purposes.

[Julius Brenzaida]

It has definitely something to do with mapsize and the number of citizens. Indeed if you have, let's say 5 cities on despotism and they grow red and black hats come quickly. You have to be careful at the beginning of games, not to grow too fast before getting monarchy or republic.

[SlowThinker]

Originally posted by Julius Brenzaida
It has definitely something to do with mapsize and the number of citizens. Indeed if you have, let's say 5 cities on despotism and they grow red and black hats come quickly. You have to be careful at the beginning of games, not to grow too fast before getting monarchy or republic.

My goal is to be able determine the city where I can expect riots after building a new city.

[SlowThinker]

About the question which cities get additional red/black hats due to too many of cities:
After my past tests, recent reading of A pattern governing supply and demand changes and changes of happiness after English were destroyed by Zulus (and so one city was destroyed) I have a feeling that there may be a pattern based on all cities (from every civ) in the game.

[DrSpike]

There may be a pattern. Though noone has ever come up with it I'm not sure many tried that hard.

As an additional resource I recall that a pattern was hypothesised for SMAC that some players believed in......you might want to check there and see if it helps.

[SlowThinker]

After reading A pattern governing supply and demand changes , I realized that the Civ2 engine work also with a list of all cities of all civs. So I run some tests that ended up in the following results. Tests were done in deity.
Please verify it, I tested extensively only despotism.

Unhappiness due to number of cities (UDNC) is an additional kind of unhappiness above the base unhappiness (base unhappiness = 2nd unhappy citizen in each city in deity, 3rd unhappy citizen in emperor etc.). It is supposed an extra unhappiness may be caused also due to large cities.

Arrangements
The Civ2 engine maintains a list of all cities of all civs in the order in which they were created as the game was played. For every city I will name this numerical order as the all-inclusive city number (ACN).
There is also the order in which your own cities were built: this is represented by the city number (CN).
Please don't confuse ACN (placings among cities of all civs) with CN (placings among your own cities).

I will use an abbreviation C for a content citizen (a blue hat), U for a unhappy citizen (a red hat), VU for a very unhappy citizen (a black hat). I will suppose cities are 1-sized, so the term citizen will always be equal to the first citizen.

Every new city created by your civilization may cause UDNC in one or more your cities. The question is 'in which cities?' There is a pattern that says in which cities the new city with a given CN causes an unhappiness. The pattern depends only on ACNs of your cities:

---

The effect of governments
Communism and fundamentalism have no UDNC.
Each other government has its own government constant (GC). This number is equal to the maximal number of cities that have never an UDNC:

government	GC
anarchy	4
despotism	4
monarchy	6
communism	-
fundamentalism	-
republic	8
democracy	10

Now let us see despotism for example.

Cities are divided in 4 groups: Group 1 are cities with CN 1-4, group 2are cities with CN 5-8, group 3 are cities with CN 9-12, group 4 are cities with CN>12.

Step 1) Base unhappiness (in cities of groups 3 and 4)
Every city in group 3 gets an U citizen, every city in group 4 gets an VU citizen.

Step 2) Additional unhappiness (caused by cities of groups 2 and 3)
First steps 2a) and 2b) are processed for group 2, then again for group 3. Cities of group 2 cause citizens to be U, cities of group 3 cause citizens to be VU.
(Group 1 causes no unhappiness. Group 4 causes no unhappiness too, because all citizens are already VU.)

Step 2a)
The added unhappiness is determined by the following table. This is the pattern similar to 'xxxo' known from Oedo years etc. Each city falls into one of 4 categories; cities from every category get additional unhappiness together.
(Realize that only some of cities listed in the lowest row are yours. Naturally foreign cities get no unhappiness from creating your own cities.)

	group 2				group 3
kind of additional unhappiness	U				VU
CN of city that causes unhappiness	5	6	7	8	9	10	11	12
ACN of cities that get unhappiness	4, 8, 12, 16, 20...	3, 7...	2, 6...	1, 5...	4, 8...	3, 7...	2, 6...	1, 5...

Step 2b)
Rule 1 must be effective. So cities that didn't get any additional unhappiness from Step 2a) get unhappiness too, furthermore immediately after creating.

Rule 1)
After creating first 4 cities all citizens must be C, after creating first 8 cities all citizens must be U, after creating first 12 cities all citizens must be VU.

---

An example:

Suppose all cities are yours, so CN=ACN for every city.
Cells with citizens that became U (red hat) are red, cells with citizens that became VU (black hat) are black. For example when you build your 9th city under despotism, city nr. 9 gets red hat, cities nr. 4 and 8 get black hat.

despotism		group 1				group 2				group 3
	CN of city that causes unhappiness	1	2	3	4	5	6	7	8	9	10	11	12
Step 1	CN of cities that get unhappiness									9	10	11	12
Step 2a	ACN (=CN) of cities that get unhappiness					4	3	2, 6	1, 5	4, 8	3, 7	2, 6, 10	1, 5, 9
Step 2b	CN of cities that get unhappiness							7	8			11	12

 

monarchy		group 1						group 2						group 3
	CN of city that causes unhappiness	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17	18
Step 1	CN of cities that get unhappiness													13	14	15	16	17	18
Step 2a	ACN (=CN) of cities that get unhappiness							6	5	4	3, 9	2, 8	1, 7	6, 12	5, 11	4, 10	3, 9, 15	2, 8,14	1, 7, 13
Step 2b	CN of cities that get unhappiness										10	11	12			15	16	17	1 8

 

---

Remaining work:
All tests were done with cities of size 1. Large cities may cause an additional extra unhappiness.
What happens if a city is destroyed, then another city is build (own or foreign)?
Tests were done for deity only.

 

[DrSpike]

Neat....now how do I practically use this to tell in which city the unhappy first citizen due to empire size will appear?

[SlowThinker]

DrSpike,
Hm, I didn't think about utilization very much yet. (Also there may be still some errors. I am waiting for some confirmation)

You must know the ACNs of your cities. From the point where you build Marco Polo it is easy (only problem may be formerly destroyed cities - they may shuffle the order). Without MPE it will be more difficult. Especially for the first city with an unhappy citizen due to empire size (city nr. 5 under despotism) you have no clue. But once you know the effect of the city nr. 5 you know which city have ACN=4 (or 8 or 12 - it depends also on the number of civs in the game). Then you can guess which city have ACN=3 and so will be the next one.

From other side you may find out the ACNs of your cities from the way how cities get red and black hats and then use the knowledge of ACNs for commodities S&D changes.

[DrSpike]

Much as this is very clever it doesn't look as practically useful yet as some of your research. But keep trying.

[Ben Kenobi]

What about goto?

I'm playing a 250 city game with AI zoos. Won't the goto for the units list the order in which the cities, including AI cities are listed?

I've noticed this extra unhappiness occurring in pairs of cities in my democracy on the Attitude advisor.

I can't really see the pattern emerging unless you have a significant number of the total cities, or knowledge of all or almost all the AI's.

[SlowThinker]

Originally posted by obiwan18
I'm playing a 250 city game with AI zoos.

zoos??

What about goto?...Won't the goto for the units list the order in which the cities, including AI cities are listed?

I don't think so. Goto and Find city are in the alphabetical order

I can't really see the pattern emerging unless you have a significant number of the total cities, or knowledge of all or almost all the AI's.

You can get the pattern also from commodity changes. Commodity changes distinguish 16 'types' of cities and happiness only 4.

[SCG]

Originally posted by SlowThinker

zoos??

What does the western world do to keep species from becoming extinct? They put that species in a cage, preventing them from escaping and getting themselves killed off - a zoo . How do we keep AIs under control and keep from being killed off? we leave them caged up in one city, and prevent them from escaping - hense being called a zoo