Simulating Citizens

This is for directed discussions on immediate questions of game design. Only moderators can create new threads.
Locked
Message
Author
User avatar
Bigjoe5
Designer and Programmer
Posts: 2058
Joined: Tue Aug 14, 2007 6:33 pm
Location: Orion

Re: Simulating Citizens

#241 Post by Bigjoe5 »

RonaldX wrote:I don't really see a need for a growth value, all of the things you have set as factors in growth could simply be set as factors in an integer target value, but I understand what you're getting at. The reason I don't like growth is because it doesn't work asymptotically.. If you act very slightly bloodthirsty for a long time, then a hugely bloodthirsty race will eventually come to love you for it because you slowly but surely grow in that direction, even though you aren't acting in a particularily bloodthirsty manner.
This is the known and desired outcome of the model I propose. The idea is that over the course of the game, an empire's alignments will gradually polarize towards one side or the other unless explicit effort is made to keep them around the middle, for a particularly diplomatic empire, for example. This is desirable because as the game progresses, each empire will inevitably develop its own distinctive character, which is one of the goals of this system to begin with. In addition, as I've mentioned, if over many turns, a large, difficult-to-reverse shift in alignment is observed, this makes the decision of whether or not to start down that path at all even more significant.
I don't see any reason why I should get a bonus to my allegiance if another empire treats a race badly, or vice versa. I believe that is an overcomplication. I have enough to worry about designing ships, producing fleets, fighting battles, conducting diplomacy, defending myself from espionnage, etc. etc. to have to worry about how my planets are going to react if my opponent starts enslaving his people. I agree with the factors regarding ennobling/enslaving my own people, or killing billions of a given race, or destroying their homeworld, but the things that affect allegiance to MY empire should be MY actions, not my opponents'.
Geoff seems to agree with you, but I do not. However, I can see the benefit of not having diplomatic statuses such as treaties and wars affect allegiance in this way. Instead, a race will either like a certain type of treaty or not, and that will affect the growth of alignment in a certain way, regardless of that species allegiance to the empire in question.

However, in terms of one-time trade negotiations (trading resources, ships, tech, etc.) I believe it would be useful if the one-time shift in current alignment was based on the species' allegiance towards both the empires in question. This would give an incentive to give good treatment to empires towards whom your species have high allegiance, which will greatly enhance the political/diplomatic aspect of the game, since if your opponents have higher allegiance towards you than towards him, he'll be more likely to give you presents to try to curry favour with his people. This helps to make being the diplomatic empire an interesting and viable way to play.
Not necessarily. Only the things that happened that still have an effect need to be displayed. In my example, bombarding an enemy planet and killing 15 billion of a race causes a 15 point drop in the target allegiance of that race to your empire. Each turn, they forget 1 billion of those deaths (a defined property of the "killed lots of the race" factor). After 15 turns, as long as you havn't killed anyone else, they completely forgive you and this factor disappears from the list of effects on target allegiance.
That means that each allegiance altering event would need to have both a magnitude, and a "rate of diminuation", which is more complicated than a single event just altering the alignment scale once. In addition, this means that the "growth" of allegiance will essentially be dependent on how many distinct factors are contributing to it, which is kind of weird, and would lead to predictable, but weird fluctuations in allegiance growth.
As long as you define certain factors as diminishing, there is no reason to keep track of them beyond the period where they are relevant.
They are, however, still mixed in with current object-effects, which is still awkward and potentially confusing, as well as easily avoided by using alignments.
The rate of growth or decay can simply be tied to the factors themselves. They have a rate at which they accumulate (blockading a planet causes target happiness to drop by x per turn) or decay (a race suffers a -100 initial hit to target allegiance if you destroy their homeworld, but it climbs back up 1 per turn for 100 turns, at which point you are forgiven). In this way the factors themselves are the only determinants of target allegiance AND target happiness, both values can be easily calculated at any time. A player can quickly see which factors are affecting his allegiance, or on a planet, happiness, which are the most important, and how long they are going to stick around for.
The fact that it's tied into the factors themselves just makes the idea of allegiance even more complicated. Suppose the initial hit of 100, for example, occurred when the allegiance of that species was only 50. You would see everything that adds up to 50, and you would see Destroyed Homeworld: -100 (+1), whereas the player's actual rate of allegiance growth will be 0, due to the fact that that +1 only applies to the diminishing factor of that specific allegiance factor (i.e. the penalty will be -99 next turn, which still puts the player below 0). That's what happens when you try to combine diminishing factors and current factors and historical factors into the same set of values - weird things happen that can be avoided very easily by using only alignments as I have proposed.
Krikkitone wrote:Also, as for affects to allegience Never fading away? are you serious, no actions that affect allegiance only for a while? If I kill some population with collateral damage in the first 100 turns It will be remembered till the end of the game?
As long as there are many ways to alter the alignment values over the course of the game, including continual growth towards one extreme or the other, I don't see a problem with this, since it's still possible to max out the scale even if you did something horrific in the early game; you just need enough time to get them to like you again, which I think is fine. This "diminishing" of the effects of certain events is essentially built right into the system this way, since after a long while, you'll get to a point where it won't matter what you did in the early game, you'll still either be at 0 or 100.
Geoff the Medio wrote:Allegiance of a species to an empire does not depend on the allegiance of the species to other empires, or what diplomatic relationships the empire has with other empires
This is essentially the only thing I would object to. As I've mentioned before, a direct exchange of resources should affect allegiance (via alignment) based on relative allegiance to the two empires. For example, if Blue has 20 more allegiance than Green from the Egassem, Green would get a bonus to Egassem allegiance for giving Blue a present of technology or ships or minerals or whatever. If Blue gave Green a present, and Green accepted, that would be a penalty to Egassem allegiance for Green for taking something from an empire they liked (it doesn't matter who initiated the deal, whether it was Blue giving something to Green, or Green demanding it from Blue). I'm not entirely sure at this point if there should be any impact on Egassem allegiance to Blue in these situations.

Edit:
Geoff the Medio wrote:* Target happiness of a planet can depend on the foreign relations of the empire that owns the planet, and the species on the planet's allegiance to other empires
This is also somewhat questionable... This is essentially what you just said allegiance shouldn't depend on. Why is it better for happiness to be affected directly in this way?

Edit2:
Geoff the Medio wrote:
RonaldX wrote:
Most species would be indifferent or mildly opposed to bad things being done to most other species, but particularly friendly or hateful species could have "ethos" that strongly likes or dislikes empires for doing good or bad things to specific other species.
This sounds like it's getting overcomplicated and nitpicky. It could be relatively easily implemented as a modifier to allegiance like, "Race X gets a bonus to allegiance for every 1 billion of Race Y that you vaporize," but this has the negative connotation that it is by design better for certain races to be selected as opponents than others. While not as in-depth, the level of pacifism or bloodthirst of a race could produce similar modifiers without the added complexity level of inter-race compatibility.
Perhaps it would be "nitpicky" to have one race's allegiance to an empire depend on the actions of that empire towards another race. But, if each empire's actions towards each race are tracked and rated like any other "alignment" scale (eg. values like pacifism, expansionism, environmentalism, etc.) then this inter-species sympathy or hatred can be added to species-empire relations essentially for free. It doesn't necessarily need to be used, but could be made easily modifiable with content files.
This could potentially be used in specific campaign scenarios, where the races are predefined and uncustomizable, and there may be specific rules which are slightly different from standard mode, to make the prescribed path to victory more interesting.
Warning: Antarans in dimensional portal are closer than they appear.

User avatar
eleazar
Design & Graphics Lead Emeritus
Posts: 3858
Joined: Sat Sep 23, 2006 7:09 pm
Location: USA — midwest

Re: Simulating Citizens

#242 Post by eleazar »

With the exception of the items commented on below, i agree with Geoff's summary. Apologies if i've missed the intervening discussion, but things seem to be moving so fast right now.
Geoff the Medio wrote:* Planets have a happiness meter, which has a current and target value with min and max values of 0 and 100 (like standard planet meters).
I realize all our meters so far go from 0 to 100, and that (at least the old) design documents indicate that 20 should always be the break point where bad things start happening. But 20 as a "neutral point" is awkward for happiness, since there are probably quite a lot of causes for unhappiness, and jamming all the action into the bottom 20 points looses a lot of resolution.

There's a major difference between happiness meters and normal planetary meters. While it nonsense to have a population of -10 or a mineral production of -44, a negative happiness can be meaningful.

In fact, i think it is more easily understandable if the neutral point for happiness (and allegiance, and possibly alignment scales) were zero. It's easier to see that a happiness of -10 signifies slight unhappiness (on a -100 to 100 scale) than a happiness of 45 (on a 0 to 100 scale). Or if you want to keep the thing 100 long, a -50 to 50 scale is fine. I strongly feel that zero is an easier to understand neutral point for the scale.

Geoff the Medio wrote:* Current happiness is modified each turn to move up to one point towards its target value, or less than one point if the difference between them is smaller than one point.
Surely it is premature to define the precise speed that current moves toward target. That's something that will likely be need to be adjusted after testing. And much earlier in the thread we did some work on nice exponential decays.
Geoff the Medio wrote:* Allegiance of a species to an empire may also depend on "social engineering" type settings or choices the player can make, such as what government style the empire uses, which the "ethos" of the species may make favourable or unfavourable to the species
I realize this item is a possibility, but i'll note that i think it's quite possible that "civics" (i.e. social engineering choices) could instead effect the imperial alignments directly-- depending on what civics we devise. For instance, choosing the "Democracy" civic would move an empire further toward the "Egalitarian" end of a "Egalitarian<>Hierarchical" scale.
Geoff the Medio wrote:* A species' "rank" in an empire could alternatively affect the target happiness of planets of that species in the empire, and not affect allegiance
* Target happiness of a planet can depend on the foreign relations of the empire that owns the planet, and the species on the planet's allegiance to other empires.
I think it is important that the things that effect happiness be entirely local, with the exception of allegiance. Otherwise we have a 3 tiered system, with the middle tier blurring the line between the two others, like so:
  • Allegiance: collective feeling of a whole species
    Empire-wide Happiness: collective feeling of a large fraction of a species
    Local Happiness: individual feeling of a particular planet
I really don't think it's worth our trouble to make the player distinguish between local and empire-wide happiness, especially since very similar functionality can be achieved by rolling up empire-wide causes of happiness into allegiance. The terminology and functioning is clearer with only two tiers.
Geoff the Medio wrote:* Allegiance of a species to an empire does not depend on the allegiance of the species to other empires, or what diplomatic relationships the empire has with other empires.
I've recently been wondering about this item. It was a big part of earlier discussions, and i supported it, but i haven't been able to remember any compelling reasons to keep it, beyond it being a way to somewhat simulate "repulsive" or "charismatic" pics, which IMHO is not enough reason on it's own. Unless there is some compelling reason that hasn't occurred to me (or i've forgotten) i'm happy dropping this.

User avatar
Krikkitone
Creative Contributor
Posts: 1559
Joined: Sat Sep 13, 2003 6:52 pm

Re: Simulating Citizens

#243 Post by Krikkitone »

Let me see if I can sum up the major ideas

Each of these models avoids the problem of a "target with a target" which is counterintuitive to the idea of a target (if my 'target' is where my happiness will be if nothing affects it, then what about the fact that my target itself is moving?)

limited version of Geoff's model*...
Allegiance only changes due to "events" there is no 'drift' to a target over time.
Happiness however does drift to a target dependent on Allegiance
Happiness has the important effect on the world

Big Joe's...
Allegiance 'Has' a target, but that target is always 0 or 100, the variation is in how fast it gets there
Happiness drifts to a target dependent on Allegiance (it drifts fairly fast compared to the normal movement of allegiance)
Happiness has the important effect on the world

Krikkit1's...
Allegiance drifts to a target
(un)Happiness drifts to a target independent of Allegiance.
(un)Happiness and Allegiance interact in some simple way (add or subtract) for any important effects on the world


Each of these have particular traits
1. In Geoff's, Happiness+Allegiance is very stable, rarely will something alter it.. if something does suddenly alter it, Allegiance alters suddenly and permanently, Happiness (and planet behavior) alters gradually

2. In Big Joe's, both are continually steadily moving.. Large sudden Allegiance movements due to events (as opposed to the continuous movement) has a non-fading effect on Allegiance, and a gradual nonfading effect on Happiness (and planet behavior)

3. In Krikkitone's, both move toward a target... sudden adjustments to Current Allegiance values due to events produce a sudden effect on planet behavior (the same as sudden adjustments to (un)happiness)

In both #1+2 a change of ownership needs either a special rule to adjust Happiness OR allow change in ownership to have only a gradual effect on planet behavior
In #3 a change of ownership has an inbuilt immediate effect on planet behavior


I prefer mine for the reasons that
1. It still allows Target Allegiance (events can push you away from a 'normal' that you gradually return to)
2. It has the built in planet ownership change effect
3. It allows instant effects from instant allegiance altering events
4. It allows any type of "movement to target" exponential, regular, variable based on circumstances
5. It works with two competing meters so you get a greater range for possible responses (effectively if one meter is positive and the other is negative, the "effect range" goes from -100 to +100)
6. Two competing meters also means there doesn't have to be "thresholds" like 20, the "threshold" is the other meter.



There is also the issue about what to do with empire-specific diplomacy, but that is a separate subject and the best way to deal with it depends on the overall model chosen

*The 'expanded' version of Geoff's model where allegiance has a target that is seperate from its current value gives the "Target of a Target" problem... ie my world says currrent Happiness 40, target 60... but if my Allegiance is current 60, target 40... so on the long term my Happiness does NOT reach 60 [like a 'target' should], but 40

PS I think geoff intended for 'social settings' to directly affect alignment.. and thereby affect allegiance

Also a seperate thought on ho to get "Higher ranks" to work.

A species can 'be' a certain rank in your empire , but only gets the benefits of the rank if they meet certain criteria

The population of all Overlord species in your empire must be less than 1/10 of the population of all "less than citizen" species (or they only get 'Master race' effects)

The population of all Overlord+all Master Race species in your empire must be less than 1/3 of the population of all "less than citizen" species (or they only get 'Aristocrat' effects)

The population of all Overlord+all Master Race+all Aristocrat species in your empire must be less than the population of all "less than citizen" species (or they get only 'Citizen' effects)

Essentially this is the "there must be enough servants" rule
So you could designate your starting race as an Overlord race... but it would have no effect until you conquered and enslaved some other races, the greater your 'slave race' population compared to your starting race, the greater the Effects would be.

User avatar
Geoff the Medio
Programming, Design, Admin
Posts: 13587
Joined: Wed Oct 08, 2003 1:33 am
Location: Munich

Re: Simulating Citizens

#244 Post by Geoff the Medio »

Bigjoe5 wrote:However, in terms of one-time trade negotiations (trading resources, ships, tech, etc.) I believe it would be useful if the one-time shift in current alignment was based on the species' allegiance towards both the empires in question. This would give an incentive to give good treatment to empires towards whom your species have high allegiance...
Why does this need to happen by altering alignment or allegiance? My summary post has diplomatic status affecting planets' target happiness, which would also provide that incentive to players.
Geoff the Medio wrote:* Target happiness of a planet can depend on the foreign relations of the empire that owns the planet, and the species on the planet's allegiance to other empires
This is also somewhat questionable... This is essentially what you just said allegiance shouldn't depend on. Why is it better for happiness to be affected directly in this way?
It's exactly what I said allegiance shouldn't depend on, which is why the point were grouped together.

The motivation for having allegiance not depend on other allegiances is to make it more predictable and understandable. A speces' alleiance to an empire should be a species opinion of that empire, not of other empires, for which the species has separate allegiances. I don't see a need for the extra complications that allegiance interdependence via diplomatic relations would bring, which similar results can be achived by modifying target happiness instead.
eleazar wrote:Surely it is premature to define the precise speed that current moves toward target.
The rate of movement towards target can be modified later, as I suspect will also be done for resource meters. For now, we just need a value to start with which is simple. "This will be reviewed and possibly revised" can be added to the summary, though.
Krikkitone wrote:I think geoff intended for 'social settings' to directly affect alignment.. and thereby affect allegiance
That's correct. My summary post doesn't actually mention "alignment" specifically, and the means by which "social settings" affect allegiance could be via "alignment".
eleazar wrote:I think it is important that the things that effect happiness be entirely local, with the exception of allegiance. Otherwise we have a 3 tiered system, with the middle tier blurring the line between the two others, like so:

Allegiance: collective feeling of a whole species
Empire-wide Happiness: collective feeling of a large fraction of a species
Local Happiness: individual feeling of a particular planet

I really don't think it's worth our trouble to make the player distinguish between local and empire-wide happiness, especially since very similar functionality can be achieved by rolling up empire-wide causes of happiness into allegiance. The terminology and functioning is clearer with only two tiers.
We can't really restrict happiness altering effects to be entirely localized, as it's (probably?) inevitable that there will be a variety of "scopes" of happiness altering effects...
* Empire techs could affect all planets in the empire, regardless of speces.
* Species effects would affect all planets of a particular species, regardless of empire.
* Species-empire effects would affect all planets of a particular species in a particular empire.
* Species-system or species-species effects could affect all planets of a particular species in a parituclar system.
* Buildings could affect all planets within some local area.
* Specials would affect a single planet.
* Events could also affect a single planet, or multiple planets depending on the nature of the event.

However, I don't see these as many different "tiers", and I think making a strong distinction between empire-wide and purely-local is a bit of an arbitrary cutoff when there are a variety of groupings for (happiness) meter altering effects. Each planet has a set of properties and situation that give it a potentially unique combination of effects that alter its target happiness. Some of these act on other planets as well, but all "act locally" on every planet they affect.

That said, I can see some potential confusion about a particular effect would modify the "empire-wide" allegiance or the "local" happiness. However, we don't need to emphasize "empire-wide" and "local" about allegiance and happiness, but rather can make them primarily disinctive due to the the different consequences of the values.

User avatar
eleazar
Design & Graphics Lead Emeritus
Posts: 3858
Joined: Sat Sep 23, 2006 7:09 pm
Location: USA — midwest

Re: Simulating Citizens

#245 Post by eleazar »

Krikkitone wrote:What makes the above unworkable (or at least highly undesirable without massive kludges) is

Target Happiness being affected by Current Allegiance

That means that Target happiness would have a "current" and "target" value (ie it would have a sudden change and then a gradual return to normal).
This makes it behave completely differently than all other meters and in a way that is counter to its description.

It Also means that a Current Allegiance altering event will not have an Immediate fading impact on Current Happiness, instead current happiness will Slowly feel the impact and then slowly fade away.
If I mass bomb Psilon worlds, MY Psilon worlds won't start feeling the effect until several turns later.
If i understand your complaint, there is a much simpler means to avoid it than the creation of a new "discontent" meter, and whatever else.

Simply add in Allegiance to the happiness equation at a different point.
So to determine the happiness of a planet on a particular turn, you increment (by whatever means) "tracking happiness" towards "target happiness". ("Tracking happiness" works just like "target happiness" except with this system, it's not the final answer.) Then you add "Current Allegiance" onto "tracking happiness" which produces the value of "happiness" that you actually use. Thus any fluctuation in allegiance would immediately be reflected in happiness.

I am not sure that this is desirable. But it is possible.

Bigjoe5 wrote:
RonaldX wrote:I don't really see a need for a growth value, all of the things you have set as factors in growth could simply be set as factors in an integer target value, but I understand what you're getting at. The reason I don't like growth is because it doesn't work asymptotically.. If you act very slightly bloodthirsty for a long time, then a hugely bloodthirsty race will eventually come to love you for it because you slowly but surely grow in that direction, even though you aren't acting in a particularly bloodthirsty manner.
This is the known and desired outcome of the model I propose. The idea is that over the course of the game, an empire's alignments will gradually polarize towards one side or the other unless explicit effort is made to keep them around the middle, for a particularly diplomatic empire, for example. This is desirable because as the game progresses, each empire will inevitably develop its own distinctive character, which is one of the goals of this system to begin with. In addition, as I've mentioned, if over many turns, a large, difficult-to-reverse shift in alignment is observed, this makes the decision of whether or not to start down that path at all even more significant.
It may be desirable for imperial alignments to have some sort of durability, i don't think we want a system were empires can rapidly fluctuate from being extremely bloodthirsty to extremely pacifistic. However i don't think we need a mechanism to force an empire to extremes, weather or not it's actions are extreme. There are already a lot of factors in play which will tend to make it difficult (and sometimes suicidal) but not technically impossible to reverse an extreme alignment.

Consider. If a player starts out with a Pacifistic species, he'll probably begin the game in a Pacifistic manner: not focusing much on the construction of warships, nor the research of offensive techs. He'll tend to incorporate and populate his empire with other Pacifistic species (or at least species that aren't bloodthirsty), and promote them to higher rank. The wonders he builds and civics he choses will tend to enhance (or at least not obstruct) a pacifistic path to victory. Should the empire want to suddenly shift to a bloodthirsty path, there are natural obstacles in the path. Development up to this point has not well provided the empire for this path, and the bulk of the citizenry will either object to bloodthirstiness, or at best not be enthusiastic about it. I see no need for additional means to push an empire toward one extreme or another.

However, i don't think an empire should naturally advance to an extreme on all of the alignment scales at once. Some empires might embark on a strategy midway between Bloodthirstiness and Pacifism, and while it won't garner them any allegiance points, it will keep them from upsetting ethoi on either side. But for the purposes of winning high allegiance to at least some of citizens, and because it indicates a focused strategy, i expect the vast majority of empires will have at least one extreme alignment scale.

RonaldX wrote:I don't see any reason why I should get a bonus to my allegiance if another empire treats a race badly, or vice versa. I believe that is an overcomplication. I have enough to worry about designing ships, producing fleets, fighting battles, conducting diplomacy, defending myself from espionnage, etc. etc. to have to worry about how my planets are going to react if my opponent starts enslaving his people. I agree with the factors regarding ennobling/enslaving my own people, or killing billions of a given race, or destroying their homeworld, but the things that affect allegiance to MY empire should be MY actions, not my opponents'.
To clarify, i don't think anyone is proposing what you seem to be concerned about:
For example, Red, Purple and Green empires all have a population of Pifiltrigi, with a neutral allegiance. Red declares war on Purple and glasses several Pifiltrigi planets. The allegiance of the Pifiltrigi toward Red in all empires will go down. But nobody, i think, is proposing that the allegiance of the Pifiltrigi should go down toward any other
empire due to Red's bombing. However, it has been sometimes proposed that if Green makes a favorable treaty with Red (an empire Pifiltrigi have lower allegiance to) that this will decrease Pifiltrigi allegiance toward Green.

BigJoe wrote:That means that each allegiance altering event would need to have both a magnitude, and a "rate of diminuation", which is more complicated than a single event just altering the alignment scale once.
If allegiance factors fade away, it should occur at a single uniform rate, not at a rate specific to each kind of event.

Geoff the Medio wrote:
bigjoe wrote:Allegiance of a species to an empire does not depend on the allegiance of the species to other empires, or what diplomatic relationships the empire has with other empires
This is essentially the only thing I would object to. As I've mentioned before, a direct exchange of resources should affect allegiance (via alignment) based on relative allegiance to the two empires.
This seems like the least justifiable circumstance for treaties to effect allegiance. Surely declaring war on a beloved empire is a more obvious circumstance for rousing citizen disapproval, than trading spare minerals with a hated empire. Not that i'm sure that we really need either.

User avatar
Krikkitone
Creative Contributor
Posts: 1559
Joined: Sat Sep 13, 2003 6:52 pm

Re: Simulating Citizens

#246 Post by Krikkitone »

Geoff the Medio wrote:We can't really restrict happiness altering effects to be entirely localized, as it's (probably?) inevitable that there will be a variety of "scopes" of happiness altering effects...

* Species effects would affect all planets of a particular species, regardless of empire.
This one I strongly disagree with.. something that makes members of a species more/less likely to revolt against ALL empires would be bad

All "Revolt" inducing/avoiding activities should be targeted against a particular empire or at a particular location
* Empire techs could affect all planets in the empire, regardless of speces
Handle by increasing/decreasing allegiance of all species present in your empire to you
* Species-empire effects would affect all planets of a particular species in a particular empire.
This should be handled Purely by Allegiance (which is a species-empire effect) either by having allegiance impact happiness or by having allegiance directly impact amount of "Rebel activity' etc.

The others are just different variations on "Local effect" (System wide, Range, etc.)
If i understand your complaint, there is a much simpler means to avoid it than the creation of a new "discontent" meter, and whatever else.

Simply add in Allegiance to the happiness equation at a different point.
So to determine the happiness of a planet on a particular turn, you increment (by whatever means) "tracking happiness" towards "target happiness". ("Tracking happiness" works just like "target happiness" except with this system, it's not the final answer.) Then you add "Current Allegiance" onto "tracking happiness" which produces the value of "happiness" that you actually use. Thus any fluctuation in allegiance would immediately be reflected in happiness.

I am not sure that this is desirable. But it is possible.
That was my earlier idea, (happiness is a composite value)
However I think "Discontent" v. "Allegiance"
is simpler than
"Allegiance" +otherthings->"Happiness" v. Threshold values (20, 0 or whatever)

Note: discontent would not be 'new' it would totally replace 'happiness', by being negative it would more easily compare to allegiance
(Easier to check if A or D is greater than if A+H add up to something)

User avatar
eleazar
Design & Graphics Lead Emeritus
Posts: 3858
Joined: Sat Sep 23, 2006 7:09 pm
Location: USA — midwest

Re: Simulating Citizens

#247 Post by eleazar »

Krikkitone wrote:
Geoff the Medio wrote:* Empire techs could affect all planets in the empire, regardless of speces
Handle by increasing/decreasing allegiance of all species present in your empire to you
Allegiance is a property of a species as a whole. Otherwise were back dealing with the various problems we had a month or so back when we were trying to make allegiance work on an empire-wide, not species-wide basis.

User avatar
Bigjoe5
Designer and Programmer
Posts: 2058
Joined: Tue Aug 14, 2007 6:33 pm
Location: Orion

Re: Simulating Citizens

#248 Post by Bigjoe5 »

eleazar wrote:I realize all our meters so far go from 0 to 100, and that (at least the old) design documents indicate that 20 should always be the break point where bad things start happening. But 20 as a "neutral point" is awkward for happiness, since there are probably quite a lot of causes for unhappiness, and jamming all the action into the bottom 20 points looses a lot of resolution.
Just because 20 is the threshold for bad things happening doesn't mean that it has to be the "neutral" point. It's actually not entirely clear that there will even be a meaningful "neutral" point, beyond that which is derived from allegiance.
In fact, i think it is more easily understandable if the neutral point for happiness (and allegiance, and possibly alignment scales) were zero. It's easier to see that a happiness of -10 signifies slight unhappiness (on a -100 to 100 scale) than a happiness of 45 (on a 0 to 100 scale). Or if you want to keep the thing 100 long, a -50 to 50 scale is fine. I strongly feel that zero is an easier to understand neutral point for the scale.
I don't have a huge problem with happiness being displayed to the player on a -50 - 50 scale, but I don't really see a huge need for it either. I think it makes a lot of sense (particularly for the purposes of this discussion) if 50 is the middle-range "neutral" point, and 20 is the bad-things threshold, for both happiness and allegiance. I also don't think that bad things should start happening as soon as happiness is below the middle point, since that's half of the entire meter, and 20 makes more sense for a threshold than -30 does.

Geoff the Medio wrote:* Allegiance of a species to an empire may also depend on "social engineering" type settings or choices the player can make, such as what government style the empire uses, which the "ethos" of the species may make favourable or unfavourable to the species
I realize this item is a possibility, but i'll note that i think it's quite possible that "civics" (i.e. social engineering choices) could instead effect the imperial alignments directly-- depending on what civics we devise. For instance, choosing the "Democracy" civic would move an empire further toward the "Egalitarian" end of a "Egalitarian<>Hierarchical" scale.
I would certainly reccomend that such options affect alignment, rather than affecting allegiance directly.
Geoff the Medio wrote:* A species' "rank" in an empire could alternatively affect the target happiness of planets of that species in the empire, and not affect allegiance
* Target happiness of a planet can depend on the foreign relations of the empire that owns the planet, and the species on the planet's allegiance to other empires.
I think it is important that the things that effect happiness be entirely local, with the exception of allegiance. Otherwise we have a 3 tiered system, with the middle tier blurring the line between the two others, like so:
  • Allegiance: collective feeling of a whole species
    Empire-wide Happiness: collective feeling of a large fraction of a species
    Local Happiness: individual feeling of a particular planet
I really don't think it's worth our trouble to make the player distinguish between local and empire-wide happiness, especially since very similar functionality can be achieved by rolling up empire-wide causes of happiness into allegiance. The terminology and functioning is clearer with only two tiers.
I wouldn't rule out the possibility of there possibly being a tech that increases max happiness on all worlds... that's something that can't really be rolled up into allegiance because each species will have a different allegiance. The things Geoff is describing above are species-specific, and can (and should) be rolled up into allegiance. I believe that it should be possible to have non-species-specific empire-wide bonuses to happiness, however, since this allows another way to add a buffer to your happiness if you have to make some politically compromising decisions.
Geoff the Medio wrote:* Allegiance of a species to an empire does not depend on the allegiance of the species to other empires, or what diplomatic relationships the empire has with other empires.
I've recently been wondering about this item. It was a big part of earlier discussions, and i supported it, but i haven't been able to remember any compelling reasons to keep it, beyond it being a way to somewhat simulate "repulsive" or "charismatic" pics, which IMHO is not enough reason on it's own. Unless there is some compelling reason that hasn't occurred to me (or i've forgotten) i'm happy dropping this.
It's fine if the ongoing effect due to diplomatic relationships is not dependent on the species' allegiance to the other empire, but as I've mentioned above, having effects due to one-time actions such as giving gifts to other empires which are dependent on the species' allegiance to the other empire adds an advantage to playing the diplomatic empire, who gets high allegiance from everyone, and to whom all the other empires will suck up and give gifts to curry favour with certain species. This makes the diplomatic path to victory more viable and interesting, which IMO, is a pretty compelling reason.
Krikkitone wrote:Big Joe's...
Allegiance 'Has' a target, but that target is always 0 or 100, the variation is in how fast it gets there
Happiness drifts to a target dependent on Allegiance (it drifts fairly fast compared to the normal movement of allegiance)
Happiness has the important effect on the world
Strictly speaking, allegiance doesn't always have to have a "target" of 0 or 100, since the empire may have some alignments which conform to the species' values and some which go against them. A Xenophobic Pacifist for example, would never reach 0 or 100 from a Xenophobic Bloodthirsty race, for example. But the point, I suppose, is that the "target" is always a composite value comprised of Alignment values which are all either 0 or 100. This just means that near the end game, certain species will always act predictably when under the control of a specific type of empire, and this behaviour will be predictable from game to game, so that each species a particular "level" of allegiance will acquire a strategic niche in the end-game, which I think is a good thing.
Krikkitone wrote:I prefer mine for the reasons that
1. It still allows Target Allegiance (events can push you away from a 'normal' that you gradually return to)
Why is this better than alignments gradually polarizing towards the extremes, and species of a specific ethos gradually aquiring a characteristic allegiance level towards an empire of a certain type?
2. It has the built in planet ownership change effect
I don't think it's been fully determined that any modification to current happiness will actually be required; it might be better if there was a slight period of adaptation, during which there are still small amounts of rioting going on, even when a planet is captured by an empire towards which it has high allegiance.
3. It allows instant effects from instant allegiance altering events
This isn't necessarily a good thing either. Also, the happiness growth rate can easily be tweaked so that allegiance effects (within a reasonable range) on target happiness have an effectively instantaneous effect on current happiness, due to the magnitude of happiness growth.
4. It allows any type of "movement to target" exponential, regular, variable based on circumstances
To what sort of circumstances are you referring, and why couldn't they affect happiness growth in the same way in any system where happiness has a current and target value?
5. It works with two competing meters so you get a greater range for possible responses (effectively if one meter is positive and the other is negative, the "effect range" goes from -100 to +100)
Mathematically, theres no distinction between this and anything else whe've been discussing. It just means that the same change in happiness has half the effect.
6. Two competing meters also means there doesn't have to be "thresholds" like 20, the "threshold" is the other meter.
I don't follow. There will always have to be a threshold value at which riots and rebellions start to happen, and as I've already expressed, I don't think it's a good idea for this threshold value to be the half-way point.
Geoff the Medio wrote:Why does this need to happen by altering alignment or allegiance? My summary post has diplomatic status affecting planets' target happiness, which would also provide that incentive to players.
I'm not talking about diplomatic status, which as an ongoing status, would have an effect on alignment growth rather than the current value. I'm talking about one-time events such as signing a trade agreement. I think we both agree that target happiness should not be affected by one-time events. It makes more sense if such actions affect either current happiness or current alignment. I feel that since a bonus to alignment is more permanent and affects allegiance as well as happiness, this is a better incentive to suck up to empires towards whom your species have high allegiance, and as I've stated, general empire-wide species-specific bonuses should usually be rolled up into allegiance, since summarizing empire-wide species-specific bonuses is pretty much the point of allegiance.
It's exactly what I said allegiance shouldn't depend on, which is why the point were grouped together.

The motivation for having allegiance not depend on other allegiances is to make it more predictable and understandable. A speces' alleiance to an empire should be a species opinion of that empire, not of other empires, for which the species has separate allegiances. I don't see a need for the extra complications that allegiance interdependence via diplomatic relations would bring, which similar results can be achived by modifying target happiness instead.
I see what you mean, but this complexity can be avoided by making alignment modifications due to diplomatic statuses not modified by a species' allegiance to the other empire (meaning that they would affect Ethical Compatibility Alignments rather than Species-Empire Alignments). This way, allegiance is not dependant on other allegiances, and the complex web of allegiance altering doeesn't exist.
Consider. If a player starts out with a Pacifistic species, he'll probably begin the game in a Pacifistic manner: not focusing much on the construction of warships, nor the research of offensive techs. He'll tend to incorporate and populate his empire with other Pacifistic species (or at least species that aren't bloodthirsty), and promote them to higher rank. The wonders he builds and civics he choses will tend to enhance (or at least not obstruct) a pacifistic path to victory. Should the empire want to suddenly shift to a bloodthirsty path, there are natural obstacles in the path. Development up to this point has not well provided the empire for this path, and the bulk of the citizenry will either object to bloodthirstiness, or at best not be enthusiastic about it. I see no need for additional means to push an empire toward one extreme or another.
Not all of the factors you mentioned will necessarily be in the game or implemented in a fashion that will necessarily put up a roadblock to changing alignment. I believe that the difficulty in changing from extreme alignments should be built in to the allegiance/alignment system itself, and content can be created afterwards which supports/balances this system.
eleazar wrote:However, i don't think an empire should naturally advance to an extreme on all of the alignment scales at once. Some empires might embark on a strategy midway between Bloodthirstiness and Pacifism, and while it won't garner them any allegiance points, it will keep them from upsetting ethoi on either side. But for the purposes of winning high allegiance to at least some of citizens, and because it indicates a focused strategy, i expect the vast majority of empires will have at least one extreme alignment scale.
There is nothing "forcing" the player to advance to extremes on all the scales; if an empire wants, for some reason, to stay in the middle of the pacifist/bloodthirsty alignment scale, he can choose his actions carefully and not veer towards one extreme or the other, but the point is that this must be a deliberate decision because it is an important decision.
eleazar wrote:This seems like the least justifiable circumstance for treaties to effect allegiance. Surely declaring war on a beloved empire is a more obvious circumstance for rousing citizen disapproval, than trading spare minerals with a hated empire. Not that i'm sure that we really need either.
Declaring war itself is an event, seperate from war as a diplomatic status, and therefore can affect current alignment in the same way as the exchange of resources.

Speaking of which, I should clarify a bit what I mean by this.
Bigjoe5 wrote:If Blue has 20 more allegiance than Green from the Egassem, Green would get a bonus to Egassem allegiance for giving Blue a present of technology or ships or minerals or whatever. If Blue gave Green a present, and Green accepted, that would be a penalty to Egassem allegiance for Green for taking something from an empire they liked. It doesn't matter who initiated the deal, whether it was Blue giving something to Green, or Green demanding it from Blue.
So suppose Blue gets 500 minerals from Green. Green now gets a bonus to current Egassem alignment. Now Green gets 300 food from Blue. Green gets a penalty to current Egassem alignment, but not enough to counter the gain he got from giving Blue 500 minerals.

Now suppose that these two transactions were a single exchange instead of two gifts. The net change in alignment would be exactly the same as if they had happened separately, and it would just be as if Green had given Blue 200 of some resource. So in terms of calculating the change in current alignment, each component of the deal is calculated separately.

What I'm getting at is that if species in your empire has high allegiance towards another empire, that doesn't give you an incentive to trade with them in general, but it does give you an incentive to enter trade agreements which are more advantageous for the other empire, in order to curry favour with a particular species. Similarly, if you declared war on an empire whom a particular species likes, current alignment with that species will immediately decrease, since that is a diplomatic action which can modify current values. The war would have no ongoing effects on the growth of Species-Empire alignments though, only on the specific Ethical Compatibility alignments, such as Pacifist/Bloodthirsty, and whatever other alignments are relevant.
Krikkitone wrote:
Geoff the Medio wrote:We can't really restrict happiness altering effects to be entirely localized, as it's (probably?) inevitable that there will be a variety of "scopes" of happiness altering effects...

* Species effects would affect all planets of a particular species, regardless of empire.
This one I strongly disagree with.. something that makes members of a species more/less likely to revolt against ALL empires would be bad
Such effects should probalby be rolled in which allegiance, since then, even if the effect is over the whole species, regardless of empire, the only planets it affects would be those owned by the empire.
Warning: Antarans in dimensional portal are closer than they appear.

User avatar
Krikkitone
Creative Contributor
Posts: 1559
Joined: Sat Sep 13, 2003 6:52 pm

Re: Simulating Citizens

#249 Post by Krikkitone »

Bigjoe5 wrote: This just means that near the end game, certain species will always act predictably when under the control of a specific type of empire, and this behaviour will be predictable from game to game, so that each species a particular "level" of allegiance will acquire a strategic niche in the end-game, which I think is a good thing.
Krikkitone wrote:I prefer mine for the reasons that
1. It still allows Target Allegiance (events can push you away from a 'normal' that you gradually return to)
Why is this better than alignments gradually polarizing towards the extremes, and species of a specific ethos gradually aquiring a characteristic allegiance level towards an empire of a certain type?
This is better because it allows Alignment to be significant early game, it allows the player to more (but not totally) easily shift alignment throughout the game.

Essentially a Player can Start with a particular alignment (the 'target' alignments based on their starting Government/social picks is also their Starting Alignment)
3. It allows instant effects from instant allegiance altering events
This isn't necessarily a good thing either. Also, the happiness growth rate can easily be tweaked so that allegiance effects (within a reasonable range) on target happiness have an effectively instantaneous effect on current happiness, due to the magnitude of happiness growth.[/quote]
That is true with your system.... but if Happiness has a massive growth rate, then Temporary Local effects (this world was bombed -50) will be rapidly swamped (but we have +5 per turn)

Also, I would argue that it IS a good thing... any alteration in a current meter should have some pretty immediate effects, and the system shouldn't force
Allegiance changes to have slow effect
or
Happiness not able to have any significant "memory"
4. It allows any type of "movement to target" exponential, regular, variable based on circumstances
To what sort of circumstances are you referring, and why couldn't they affect happiness growth in the same way in any system where happiness has a current and target value?
I'm refering to Allegiance/Alignment Growth,
In your system the Growth is the key variable (rather than the 'target') and also, in your model Allegiance/Alignment events do not "fade" really at all, unless they move you away from that 0 or 100 you are moving towards

Also... your model still allows the player to get a "target" of 60 on Elitism... by Micromanaging... ie consistently switching governments from Democracy to Monarchy... wouldn't it be easier to just let them go Aristocracy and converge in on the 60?
6. Two competing meters also means there doesn't have to be "thresholds" like 20, the "threshold" is the other meter.
I don't follow. There will always have to be a threshold value at which riots and rebellions start to happen, and as I've already expressed, I don't think it's a good idea for this threshold value to be the half-way point.
There does not always have to be a Fixed Threshold value when revolts happen, any more than there is a Fixed Threshold needed when a ship will be detected.

The model is Discontent v. Allegiance-> triggers revolts like Detection v. Stealth->triggers detection of object (although even simpler, as no range is involved).. no Thresholds are needed.

Rebel Activity= larger of (Discontent-Allegiance,0)

Enemy Espionage Suppression (or whatever the "excess Happiness" bonus is)
=larger of (Allegiance-Discontent,0)



Also the advantage of my system is it can easily be adapted into one of the other systems.(because of the ease of adapting the growth rate)

If you want Allegiance/Alignment to have extreme targets, but have Growth towards the Target be strongly affected by conditions, then that is entirely possible (and "rebel" ness will exactly match it without having to give a very fast Happiness 'decay'

if you want Allegiance/Alignment to me effectively immobile except when events hit them or when 'circumstances' Change... then that's also possible give everything "Current effects instead of Target effects", and make movement to target 0.

User avatar
Bigjoe5
Designer and Programmer
Posts: 2058
Joined: Tue Aug 14, 2007 6:33 pm
Location: Orion

Re: Simulating Citizens

#250 Post by Bigjoe5 »

Krikkitone wrote:This is better because it allows Alignment to be significant early game, it allows the player to more (but not totally) easily shift alignment throughout the game.
I think it's good if alignments aren't as significant in the early game; it allows the player to be more flexible in the early game when he's still learning about the galaxy and developing his strategy. If, in the early-mid game, the player decides that the conditions of the galaxy call for a total reversal of one or more of his alignments, that should be possible without causing total upheaval.

Also, the fact that allegiance is usually closer to 50 in the early game means that even if happiness drops below threshold, there will be only riots, not rebellions, which is good, because the player can't have to worry about his planets rebelling against him when he only has five or ten. Instead, dealing with simple riots caused by low happiness (due to bombardment/espionage/whatever) is enough to worry about.
Essentially a Player can Start with a particular alignment (the 'target' alignments based on their starting Government/social picks is also their Starting Alignment)
It may be possible to have certain empires start with certain alignments based on the ethos of their starting race, but that's a different issue which is possible with just about any system, and I would prefer to have all alignments start at the middle point anyway, unless some special racial pick (warlord?) is chosen.
That is true with your system.... but if Happiness has a massive growth rate, then Temporary Local effects (this world was bombed -50) will be rapidly swamped (but we have +5 per turn)
That's fine. Bomb it again next turn. If the effects of bombardment on happiness are severe enough compared to the amount of population that's destroyed, a planet can be easily made to stay in a state of extreme unhappiness for quite a period of time (if that ends up being desirable, that is).
Also, I would argue that it IS a good thing... any alteration in a current meter should have some pretty immediate effects, and the system shouldn't force
Allegiance changes to have slow effect
or
Happiness not able to have any significant "memory"
Why is it so bad if happiness doesn't have a significant "memory"? Isn't the point of happiness to reflect the current, local conditions of a planet? I don't think it's a bad thing at all if 10 turns after a significant bombardment event, the planet's happiness is back to normal. Bombing a planet once shouldn't make it an open target for espionage for the next 50 turns.
I'm refering to Allegiance/Alignment Growth,
In your system the Growth is the key variable (rather than the 'target') and also, in your model Allegiance/Alignment events do not "fade" really at all, unless they move you away from that 0 or 100 you are moving towards
Alright. Then why is this:
4. It allows any type of "movement to target" exponential, regular, variable based on circumstances
better than a rate of growth which is predictable and clearly defined by specific, identifiable contributing factors?
Also... your model still allows the player to get a "target" of 60 on Elitism... by Micromanaging... ie consistently switching governments from Democracy to Monarchy... wouldn't it be easier to just let them go Aristocracy and converge in on the 60?
It's undetermined whether or not the player will be able to change governments in-game at all, much less with no cost to trade or some other resource. Moreover, the player can get a "target" of 60 by regular managing - when alignment reaches 60 (give or take a few decimal points), the player can change his policies so that growth is equal to 0. This is presumably how a diplomatic empire would keep all of his (ethical compatibility) alignments scales more or less level and try to please everyone (obviously such an empire would employ some sort of strategy to try to max out species-empire alignments).
There does not always have to be a Fixed Threshold value when revolts happen, any more than there is a Fixed Threshold needed when a ship will be detected.

The model is Discontent v. Allegiance-> triggers revolts like Detection v. Stealth->triggers detection of object (although even simpler, as no range is involved).. no Thresholds are needed.
Why is this better than Number of Rebel Units = (20 - Current Happiness)^X and Probability of Rebel Units trying to Take Over the Planet = (20 - Current Allegiance) / 20? The advantage of this is that

-Empires have a way to keep even very low allegiance species at bay (by keeping happiness above 20 with techs and buildings, even if allegiance is 0), but only if happiness is not disrupted in some way, in which case, major trouble ensues. Such planets could be interesting strategic targets, as the player might have an important strategic reason for wanting that species to be part of his empire, and an enemy player could target it with bombardment or espionage to incite a severe rebellion, or alternatively, take advantage of the low happiness (read: espionage defense) and use it to gain easy access to enemy techs and ship designs with espionage.

-Planets with high allegiance but low happiness can show their discontent by rioting. This will disrupt production in the on the planet, and require the player's attention, which will add additional interest in the early game, when this type of situation is likely to occur, and when a full scale rebellion would be too much for the player to be expected to deal with.

-The way the two thresholds interact is simple, elegant (IMO) and easy to understand.
Also the advantage of my system is it can easily be adapted into one of the other systems.(because of the ease of adapting the growth rate)

If you want Allegiance/Alignment to have extreme targets, but have Growth towards the Target be strongly affected by conditions, then that is entirely possible (and "rebel" ness will exactly match it without having to give a very fast Happiness 'decay'

if you want Allegiance/Alignment to me effectively immobile except when events hit them or when 'circumstances' Change... then that's also possible give everything "Current effects instead of Target effects", and make movement to target 0.
If you're saying that it's possible to make a hybrid of two systems, I'm not sure that really counts as an advantage for your system, since the value of creating such a hybrid is unclear.
Warning: Antarans in dimensional portal are closer than they appear.

User avatar
Krikkitone
Creative Contributor
Posts: 1559
Joined: Sat Sep 13, 2003 6:52 pm

Re: Simulating Citizens

#251 Post by Krikkitone »

Bigjoe5 wrote: Why is it so bad if happiness doesn't have a significant "memory"? Isn't the point of happiness to reflect the current, local conditions of a planet? I don't think it's a bad thing at all if 10 turns after a significant bombardment event, the planet's happiness is back to normal. Bombing a planet once shouldn't make it an open target for espionage for the next 50 turns.
Happiness Should be Able to have a significant memory
Alright. Then why is this:
4. It allows any type of "movement to target" exponential, regular, variable based on circumstances
better than a rate of growth which is predictable and clearly defined by specific, identifiable contributing factors?
The point is it can be, My system is not dependent on any particular means of adjusting Current values to meet Target values

They could be exponential (my favorite as it allows historical effects to be displayed)
They could be regular (value changes by one each turn)
They could depend on other factors (similar to yours)

This seperates the "requirements of the system" from the 'time element'.
Also... your model still allows the player to get a "target" of 60 on Elitism... by Micromanaging... ie consistently switching governments from Democracy to Monarchy... wouldn't it be easier to just let them go Aristocracy and converge in on the 60?
It's undetermined whether or not the player will be able to change governments in-game at all, much less with no cost to trade or some other resource. Moreover, the player can get a "target" of 60 by regular managing - when alignment reaches 60 (give or take a few decimal points), the player can change his policies so that growth is equal to 0. This is presumably how a diplomatic empire would keep all of his (ethical compatibility) alignments scales more or less level and try to please everyone (obviously such an empire would employ some sort of strategy to try to max out species-empire alignments).
That is Still more micromanagy.

Using Target values, the player just adjusts settings and makes tradeoffs to achieve the desired Target values, and then leaves the system alone.

Using your method, the player must First adjust settings to "move into position" and then adjust settings to "stall" [once they are at the desired values]

Your method works well for Tech Levels where 'more is always better' and so you determine what trade offs will lt you get there faster... for social settings, where "more" is sometimes "worse" and you want a balance, then you need to be able to 'set your goal' (with trade offs of course) and leave it there.
Why is this better than Number of Rebel Units = (20 - Current Happiness)^X and Probability of Rebel Units trying to Take Over the Planet = (20 - Current Allegiance) / 20? The advantage of this is that
1. Compare the complexity of your equations
You have 2 equations for Rebel activity both of which involve a seperate threshold, I have 1 for Rebel activity with no Separate Thresholds

2. Why must Allegiance and Happiness do different things?.. all systems but mine have Hapiness dependent on Allegiance (and other factors) Why should I need to pay ANY attention at all to Allegiance for an individual world... why not just Happiness (since it depends on Allegiance)

3. As for the cases of Happiness/Allegiance mismatch.. why should rioting be anything different than a small rebellion? ie things are a little bit bad so you have some minimal 'combat' that disrupts production.
Perhaps Happiness might be more susceptible to enemy influence for a particular world (since they only need to target a single world and not a whole species)... but that just means it is the flexible component of the whole

I don't think they should have separate Effects...

Allegiance= all the species wide effects on "Citizen behavior"
Discontent [my model], Other factors in Happiness [other models]=all the local effects on "Citizen behavior"

If they are actually having Different effects on Citizen behavior (except in cases where one doesn't apply like military units or a new colony), then I think the system IS crossing over too much into being a simulator.

Because as a player what am I concerned with is
Using my planets to defeat my enemies
"Citizen reactions" of planet make them more (rebellion) or less(espionage defense) a contribution to my enemies

User avatar
Bigjoe5
Designer and Programmer
Posts: 2058
Joined: Tue Aug 14, 2007 6:33 pm
Location: Orion

Re: Simulating Citizens

#252 Post by Bigjoe5 »

Krikkitone wrote:Happiness Should be Able to have a significant memory
Why? A planet on which nothing special is happening should require minimal attention from the player. If happiness has a significant memory, more planets at a time will have low happiness, and this will include planets on which nothing special is happening (aside from unhappiness reduced riots), to which the player will possibly have to devote extra attention. If something bad happened and it's over, happiness should return to normal fairly quickly so that the player doesn't have to devote undue attention to it.
The point is it can be, My system is not dependent on any particular means of adjusting Current values to meet Target values

They could be exponential (my favorite as it allows historical effects to be displayed)
They could be regular (value changes by one each turn)
They could depend on other factors (similar to yours)

This seperates the "requirements of the system" from the 'time element'.
Since it's not clear that any of those options are preferable to what I've proposed, it's not clear that this is necessarily an advantage for your system.
It's undetermined whether or not the player will be able to change governments in-game at all, much less with no cost to trade or some other resource. Moreover, the player can get a "target" of 60 by regular managing - when alignment reaches 60 (give or take a few decimal points), the player can change his policies so that growth is equal to 0. This is presumably how a diplomatic empire would keep all of his (ethical compatibility) alignments scales more or less level and try to please everyone (obviously such an empire would employ some sort of strategy to try to max out species-empire alignments).
That is Still more micromanagy.

Using Target values, the player just adjusts settings and makes tradeoffs to achieve the desired Target values, and then leaves the system alone.

Using your method, the player must First adjust settings to "move into position" and then adjust settings to "stall" [once they are at the desired values]

Your method works well for Tech Levels where 'more is always better' and so you determine what trade offs will lt you get there faster... for social settings, where "more" is sometimes "worse" and you want a balance, then you need to be able to 'set your goal' (with trade offs of course) and leave it there.
It requires a bit more management yes, but I don't think it even begins to qualify as micromanagement. It's fine, IMO, if the player has to devote a bit of care and attention to his alignment meters if he wants to keep them at a particular level. They're an important part of the game, after all.
3. As for the cases of Happiness/Allegiance mismatch.. why should rioting be anything different than a small rebellion? ie things are a little bit bad so you have some minimal 'combat' that disrupts production.
I've explained that; in the early game, the player shouldn't have to worry about actually losing one of his precious few planets in a rebellion.
Warning: Antarans in dimensional portal are closer than they appear.

User avatar
Geoff the Medio
Programming, Design, Admin
Posts: 13587
Joined: Wed Oct 08, 2003 1:33 am
Location: Munich

Re: Simulating Citizens

#253 Post by Geoff the Medio »

Bigjoe5 wrote:I feel that since a bonus to alignment is more permanent and affects allegiance as well as happiness, this is a better incentive to suck up to empires towards whom your species have high allegiance
Why do we need to add incentives for empires to "suck up" to other empires to boost species allegiance? This seems like it's trying to use the whole system backwards; players should be pampering or pandering to species that are present in or will come into contact with other empires, to manipulate what the other empires can do. If an empire wants to boost allegiance of a species to the empire, the empire should do so by being more like that species' ideal empire.

Also, having allegiance of a species to an empire depend on that empire's actions towards another empire and that species' allegiance to the other empire ends up mixing allegiances between empires in a complicated and confusing way. It also makes the conceptual meaning of allegiance less clear; allegiance should be an indication of a species' opinion of what an empire is or ideals the empire holds, or things the empire does that directly affect the species.
...general empire-wide species-specific bonuses should usually be rolled up into allegiance, since summarizing empire-wide species-specific bonuses is pretty much the point of allegiance.
As above, I think more emphasis should be put on what allegiance and happiness mean and do for and to a player, and less on how they accumulate the results of different scopes of effect (which don't have as clean a break between empire-wide and single-planet as has been implied earlier in this thread). If the only difference between allegiance and happiness to players is that one is "empire wide" and the other isn't, that's not really a compelling reason to have them as separate features. More important, and interesting, is what they mean and what they do.
Bigjoe5 wrote:
Geoff the Medio wrote:The motivation for having allegiance not depend on other allegiances is to make it more predictable and understandable.
I see what you mean, but this complexity can be avoided by making alignment modifications due to diplomatic statuses not modified by a species' allegiance to the other empire (meaning that they would affect Ethical Compatibility Alignments rather than Species-Empire Alignments). This way, allegiance is not dependant on other allegiances, and the complex web of allegiance altering doeesn't exist.
This is fine, and is consistent with my original post:
Geoff the Medio wrote:* Allegiance of a species to an empire does not depend on the allegiance of the species to other empires, or what diplomatic relationships the empire has with other empires
* A possible exception to this is species that like or dislike certain types of treaties in general as part of their ethos, but this would be a general response to treaties or lack thereof, and not a specific response to a particular treaty with a particular empire
If a species likes empires that make lots of treaties, that would fall under "like or dislike certain types of treaties in general". That species can have a higher allegiance to an empire that makes lots of those treaties, without making things complicated due to inter-empire allegiance dependencies via treaty calculus.

RonaldX
Space Kraken
Posts: 107
Joined: Thu Oct 29, 2009 12:40 am
Location: Hamilton, Ontario

Re: Simulating Citizens

#254 Post by RonaldX »

Wow.. that's a lot of posts since last night.

A few quick things:
* I agree that there is no pressing need for a long term memory of most happiness/allegiance/alignment factors.
* I believe that changes to happiness/allegiance for special events such as a planet takeover or homeworld loss can be handled easily enough by special rules if required. I don't see a couple special rules as being burdensome if the framework is solid otherwise.
* One-time events that alter allegiance/alignment/happiness should have a defined decay rate so that they cease to be relevant after a given time. They can decay at different rates depending on the importance of the event, or all decay at the same rate, it doesn't really matter. The first is more realistic, the second is simpler.
* As far as allegiance of races to other empires affecting their allegiance to your empire, hell, it might be realistic, but it's too confusing and complex for me to want to put any time into drafing up math for. I wouldn't consider the tiny bit of added realism worth the huge jump in complexity. "This race likes when you are trading/at war/at peace/etc." is sufficient, without considering how much they like the other empire involved. You guys seem to be really tied to the idea so I'll leave the number crunching to you.

Alright going just quickly, I'm going to grab a few points and throw together a kind of summary and try to keep it as simple and modular as possible. I'm basically just going to go through these and present them as definitions of how they would fit into the system, from the highest level down.

I won't give any numbers because numerical scaling is more of a balancing issue. The hard part is agreeing on a general concept. Once that is threshed out, appropriate math is easy to generate.

Alignment = several scales detailing your level of elitism, bloodthirst, or expansionism. Declaring wars, Being at peace, Constructing Buildings, Colonizing new planets, and setting government types can adjust these scales.
Species Ethos = the hardwired preferences for elitism, bloodthirst, or expansionism of a given race
Ethical Compatibility = Your Alignment compared to the Species Ethos of a given race
Species Treatment = A collection of factors which describe how you treat a given race, things like genocides, homeworld destructions, diplomatic treaties with empires the race loves/hates, the rank or status of the race within your empire, etc. (a.k.a Species-Empire Alignment)
Allegiance = A measure of how loyal a given race is to your empire. It is derived from the Ethical Compatibility and Species Treatment in some manner. Extremely low or high levels of Allegiance cause various effects such as revolts.
Local Effects = A collection of factors which affect only a specific planet, such as blockades, extra rations, bombardment, sabotage, etc.
Happiness = A measure of how happy the citizens of a given planet are, derived from the Allegiance of the planet's race to your empire and the Local Factors affecting the planet, in some manner. Extremely high or low levels of Happiness cause various effects such as rebellion.


The simplest way to represent these is just as basic integer addition. ie:
- Actions you take have a pre-defined effect on your alignment (being at war always increases your bloodthirst by x)
- Allegiance is simply Ethical Compatibility, plus or minus the "Species Treatment" modifier
- Happiness is simply Allegiance, plus or minus the "Local Effects" modifier.
Pros: Dead-nuts simple, predictable.
Cons: Prone to rapid fluctuations, not very realistic.

In order to get rid of the rapid fluctuations, we introduce the concept of "target" and "current" values. ie:

- Actions you take have a pre-defined effect on your alignment (being at war always increases bloodthirst by x)
- Current Allegiance is a tracked value which moves at some defined rate towards the Target Allegiance value. The Target Allegiance is simply Ethical Compatibility, plus or minus the "Species Treatment" modifier.
- Current Happiness is a tracked value which moves at some defined rate towards the Target Happiness value. The Target Happiness is simply Allegiance, plus or minus the "Local Effects" modifier.
Pros: More realistic, changes to Allegiance and Happiness are made more gradual.
Cons: More complex, depending on how fast the "defined rate of change" is, rapid spikes may only be made possible by special rules.
Could go Either Way: Players are able to effect significant changes to their Allegiances over the course of a game in any direction, (this may or may not be a good idea strategically.. If you establish a large power base in one type of play style, you could change your playstyle but you would likely face significant backlash).

BigJoe offers a slightly different take on this in order to prevent players from shifting allegiances easily:

- Alignment is a tracked value with limits, it changes based on a growth rate defined by your decisions (how many empires you are at war with vs. at peace with, what governmental style you are in, etc.) It does not function asymptotically but pushes a player towards the extremes of alignments at all times unless they change their policies to reverse the trend.
- Allegiance is a tracked value with limits, it changes based on a growth rate defined by a combination of Ethical Compatibility and Species Treatment. It does not function asymptotically but pushes a player towards the extremes of Allegiance at all times unless they change their policies to reverse the trend.
- Happiness is a tracked value with limits, it changes based on a growth rate defined by a combination of Allegiance and Local Factors. It does not function asymptotically but pushes a player towards the extremes of happiness at all times unless they change their policies to reverse the trend.
Pros: Most realistic model, represents the trend of a generally well treated people to grow more enamored of a leader, and of a generally poorly treated people to grow more and more unhappy with a leader.
Cons: More complex, less intuitive. Values are never stable unless at one extreme or the other without significant micromanagement (note 1).
Could go either way: Players in the end-game are essentially locked into an alignment.

***

The question at the end is, to what extent are you willing to sacrifice simplicity for realism?

(note 1) I say significant micromanagement here because I'm not entirely sure on how the best way to handle this is, math-wise. The concept sounds good until I try to apply numbers to it. For example If I make the bloodthirst scale shift +1 for every race I'm at war with, and -1 for every race I'm at peace with, then I have to willingly invite more enemies than I can maybe handle just to keep my bloodthirsty people happy with me, and the inverse is true for pacifistic races. A system like this would require heavy duty balancing in order to make it actually functional. Keeping it simple programming-wise makes it extremely difficult for the player (tons of micromanagement), and keeping it simple player-wise makes it extremely difficult for the programmer (and hides the math, making it less intuitive for the player).

If you disagree that the management of a constant growth rate is a big deal, ie "Just adjust your policies so that growth is zero," then consider what that might actually mean...

I am in a 2-player game, and half my planets are a warlike/elitist race, half are pacifistic/egalitarian. I want to keep my war/peace alignment scale neutral. The other player has declared war on me so we are in a state of war. My alignment meter then constantly drifts towards warlike because I am involved in a war. I need to offset this somehow, so I need to go into an entirely different factor and perhaps give the peaceful race a higher social status than the warlike one. I'm still getting more warlike, (and possibly also getting more elitist, which might actually be making my situation worse), but at least the peaceful race is staying relatively content because of their rank bonus. However, now I've set other scales into motion which are shifting away unchecked and could have other consequences later, for example I'm still getting more and more elitist so I need to offset this with something else, introducing even more runaway variables. Or perhaps one bonus isn't growing as fast as the other and I will eventually need to make more changes later to further offset this, adding in even more variables running off on their own as well. Now we're starting to get complicated, and we're only talking about 2 races.

Compare this to an integer based system.
At war with 1 empire: Warlike race Ethical Compatibility +5, Peaceful race -5
One race favoured over another: Elitist race Ethical Compatibility +5, Egalitarian -5
Peaceful race is favored: +10 Peaceful race Species Treatment, -10 all other races

Static, obvious, and no runaway variables that can come back and bite me in the ass later.

Is it as realistic? No. But it's a whole lot simpler, and I could draw up math to implement this in an hour.

-Ty.

Edit: Correction to example at end.

User avatar
Krikkitone
Creative Contributor
Posts: 1559
Joined: Sat Sep 13, 2003 6:52 pm

Re: Simulating Citizens

#255 Post by Krikkitone »

RonaldX wrote:The simplest way to represent these is just as basic integer addition. ie:
- Actions you take have a pre-defined effect on your alignment (being at war always increases your bloodthirst by x)
- Allegiance is simply Ethical Compatibility, plus or minus the "Species Treatment" modifier
- Happiness is simply Allegiance, plus or minus the "Local Effects" modifier.
Pros: Dead-nuts simple, predictable.
Cons: Prone to rapid fluctuations, not very realistic.

In order to get rid of the rapid fluctuations, we introduce the concept of "target" and "current" values.
That is the system I am proposing...
basic system +SOME Target, Current values (not All)

Alignments->Target+Current
Species Treatment->Target+Current
Local Events->Target+Current

To allow for fluctuations (since these are all INDEPENDENT variables)

But the DEPENDENT variables
Allegiance
"Happiness"

Stay as straightforward addition calculations (in my model Happiness does not even need to be displayed)
because otherwise you add in unnecessary, and hard for a player to Simply predict, complications

Basically value that Depends on something with a Target + Current value should not have be a Target value, because that gives you a fluctuating target... a BAD thing for a player that wants to know what things are going to be like 10-20 turns down the road.

I rename Local Events to "Discontent" (especially since its most significant factors seem to be negative ones)
"Happiness" then is either
Rebel Activity (Discontent-Allegiance)
OR
Espionage Defense (Allegiance-Discontent)

Depending on which is greater.
Why do we need to add incentives for empires to "suck up" to other empires to boost species allegiance? This seems like it's trying to use the whole system backwards; players should be pampering or pandering to species that are present in or will come into contact with other empires, to manipulate what the other empires can do.
If the Blue Empire has Psilons

How does increased Red Empire-Psilon Allegiance limit the Blue empires actions towards Red?

Unless Psilon-Blue Allegiance depends on on the Psilon Allegiance to Empires that Blue acts towards.

If it impacts "Local Conditions"/"Discontent" then it can have some effect, but that is complicated
Now I do agree that Allegiance impacting Allegiance can lead to complicated feedback loops that are also complicated, so Diplomacy should probably be left out of our "First Draft" model.
It DOES need to be in though.

Locked