India + Theocracy = Awesome?

[alpaca]

Looking at Arioch's site once again, I realized that I would always go for Theocracy in the Piety policies when playing India.

The way bonuses work in almost all games (and nothing leads me to believe it will be different in Civ5), is by adding the different bonuses together. That is, if you have a -20% bonus and a -50% bonus, you get a -70% bonus. This sounds logical on a first glance but the result is a counter-intuitively large bonus given by Theocracy when you already have a pretty big free bonus for playing India. Let me give you an example:

You're playing a non-Indian faction, and have 10 population unhappiness. Then, -20% will be 2 unhappy faces gone and you will only have 8. All nice and proper. But when you play India, you will only have 5 population unhappiness, but Theocracy will still make 2 unhappy faces go away, so you are reduced to 3. From your different vantage point, this is actually a -40% bonus rather than a -20% one. To put it another way: If you have 10 pop unhappiness with India and choose Theocracy, you get -4 unhappy faces. Pretty awesome.

If you also combine it with Legalism, you will completely eliminate pop unhappiness from your capital.

 

[Tomice]

It was already similar with the HRR courthouses in BTS, which gave awesome -75% city maintanance. I doubt it's gamebreaking, though.

What I do think is that some combos of UAs and SPs might end up so strong that the devs need to patch them. Luckily, Firaxis cared well about Civ4 (Patch 3.19 was really late!).

We might even end up having certain economic strategies that only work for one faction, which could be awesome (e.g. coastal tiles were only useful for fiancial leaders in civ4)

 

[snipperrabbit!!]

Sorry, but the 20% is only applied after the 50% so it's only a 60% overall.That's because modifiers applies to the civ itself not the standard average rule.

 

[Olleus]

Yes, with the way that bonuses stack in Civ (add the percentages, and then apply; rather than cumulatively apply the percentages on top of each other) means that +x% get comparatively weaker as you stack them, but -x% get much much stronger as they get stacked.

 

[Olleus]

Crosspost:

Sorry, but the 20% is only applied after the 50% so it's only a 60% overall.That's because modifiers applies to the civ itself not the standard average rule.

Are you sure? Thats not how it worked in every other iteration of civ.

Example. If a library and university both give +50% science; then having both gives you +100% science and not +125% science. Using the same maths implies that theocracy+india would give -70% happiness, not -60%.

 

[alpaca]

Are you sure? Thats not how it worked in every other iteration of civ.

Example. If a library and university both give +50% science; then having both gives you +100% science and not +125% science. Using the same maths implies that theocracy+india would give -70% happiness, not -60%.

I concur. Multiplying the bonuses (what sniperrabbit described) is not something that happened in any Civ game, nor any other game I can remember right now. On the other hand, it could be Firaxis make a special exception for Civ bonuses and treat them as a different "normal" level. Sniperrabbit, could you provide a link to this info?

Usually, games take a base value, then add all modifiers and apply them to it in a single pass. The alternative, which I actually prefer, is to take modifiers as multipliers (so -30% is a factor of 0.7), then multiply them on top of the value. That would mean -20% and -50% is 0.8*0.5 = 0.4 which is equivalent to a single -60% modifier.

This is more intuitive if you usually are at a certain level of modifiers, then get another modifier, but it's also harder to calculate because obviously you can't just add multipliers anymore. For designers, it has the advantage that it's easier to balance bonuses and penalties because they will always have the same impact on your current economy, no matter where your normal level is. I specifically refrained from posting about this in the OP because I thought it irrelevant to the topic, though.

 

[snipperrabbit!!]

Well, I will just take the example of a 20 pop Indian capital ! Unhappiness= (2x2)+(20x0.5)=14
Then, under Theocracy it will be either (A) (2x2)+(20x0.3)=10 or (B) (2x2)+(20x0.4)=12
Now, with Legalism on top of that it will be one of the following (AA) (2x2)+(-0.03x20)=3.4 or (BB) (2x2)+(20*0.27)=9.4
At this point, you can see that additional stacking can result in not only eliminating unhappiness generated by the city itself but even for other cities (with Freedom specialists and Military Caste garrisoning) while the multiplicative stacking is just right from a gameplay AND realism perspective.

 

[IdleEnergy]

I guess the question for those doubting multiplicative bonuses is this:

Hellenic League + Patronage = -100% influence degradation????

I could be totally wrong here, but multiplicative bonuses seems MUCH more likely when combining SP and SAs. But like most things we post about a yet-unreleased game, I'm guessing.

 

[Gre_Magus]

As IdleEnergy suggests, it is possible that unique ability modifiers are applied before other modifiers and change the base level of some attribute for a civ rather than stacking with the other modifiers.

The other possibility is that negative modifiers will work differently than positive ones. I'm too busy at work to think of how that might happen, but the problem is that adding together enough negative modifiers can quickly get you close to or at zero for an attribute, while adding positive modifiers prevents the exponential growth of attributes.

 

[Ahriman]

Most strategy games have usually had a weird and often unintuitive way of having both multiplicative and additive bonuses, depending on the particular bonus.
I would imagine that strength modifications and yield modifications will be additive, but its totally possible that some things like UAs will be multiplicative.

I think it would be very unlikely for negative modifiers and positive modifiers to work differently, because then something with +20% and -20% will not necessarily be 0% different.

Eg: if positive are additive and negative are multiplicative (and applied last) then +20% and -20% is (0.8)*(1.2) = 0.96 != 1

This would be very confusing.

 

[alpaca]

Well, I will just take the example of a 20 pop Indian capital ! Unhappiness= (2x2)+(20x0.5)=14
Then, under Theocracy it will be either (A) (2x2)+(20x0.3)=10 or (B) (2x2)+(20x0.4)=12
Now, with Legalism on top of that it will be one of the following (AA) (2x2)+(-0.03x20)=3.4 or (BB) (2x2)+(20*0.27)=9.4
At this point, you can see that additional stacking can result in not only eliminating unhappiness generated by the city itself but even for other cities (with Freedom specialists and Military Caste garrisoning) while the multiplicative stacking is just right from a gameplay AND realism perspective.

I guess the question for those doubting multiplicative bonuses is this:

Hellenic League + Patronage = -100% influence degradation????

I could be totally wrong here, but multiplicative bonuses seems MUCH more likely when combining SP and SAs. But like most things we post about a yet-unreleased game, I'm guessing.

Well, without an official answer (or an answer from a previewer) there's no way to reliably tell. All I can do is cross-comparing with different games and from that analysis I would be pleasantly surprised if Civ5 deviated from the "industry standard". The most relevant direct comparison I have is Civ4 where both Philosophical and Industrious as well as the production bonuses worked like I described - making additional bonuses through, for example, the National Epic less worthwhile.

I can't remember having played a game that multiplies bonuses, which at least means I haven't played one in the last few years after becoming conscious of the problem. I don't see a real reason so far that leads me to believe that they will be in Civ5, the Greeks might just receive such an awesome bonus with potentially no relation decay. It's not as if it would be game-breaking because in game terms it's a continuous progression from a small decay to no decay, even though it feels very different.

Edit: Positive and negative modifiers working differently probably won't happen, it'd just be complicated. And there's a good reason why, for example, there was only one maintenance-reducing building in Civ4 for most factions. Of course it's possible that UA effects are treated special but it's not a reasonable assumption without seeing evidence, because it adds complexity and takes work to implement.

 

[Louis XXIV]

Sorry, but the 20% is only applied after the 50% so it's only a 60% overall.That's because modifiers applies to the civ itself not the standard average rule.

While I think you might be right, please learn to use the conditional tense unless you have specific info to back it up. The word "might" or "could" every once in awhile wouldn't hurt anything and would actually give strength to your arguments because people will listen to your reasoning rather than ask you for proof.

 

[alpaca]

I think it would be very unlikely for negative modifiers and positive modifiers to work differently, because then something with +20% and -20% will not necessarily be 0% different.

Eg: if positive are additive and negative are multiplicative (and applied last) then +20% and -20% is (0.8)*(1.2) = 0.96 != 1

This would be very confusing.

Just saw your post: Actually this argument is true for just multiplying modifiers, which might be a reason contributing to very few (or no) games using it. Assuming that positive modifiers are treated like +x% represents a multiplier of 1 + x/100 the dilemma is this:

Case 1: -20% = (1 - 0.2) = 0.8 This would treat negative modifiers the same as positive modifiers in a sense. Say we start at level 1. Then we obtain a -20% modifier because we unlock a new social policy, this sets us at 0.8. We get a +20% modifier, equal to 1.2 which puts us at 0.8*1.2 = 0.96 (never mind the order because multiplication is commutative). Obviously, 0.96 is not 1 so -20% and +20% don't cancel out.

Case 2: -20% = 1/(1 + 0.2) = 5/6 = 0.833... This would treat negative modifiers different than positive ones but preserve returning to 1 when you apply "equal" positive and negative modifiers. So when we start at 1, and get a -20% modifier we are at 0.833 and then go back up to 6/5 * 5/6 = 1 when applying the second modifier. Here, going to 5/6 instead of 4/5 is the thing that's counter-intuitive and needs mathematics.

Neither implementation is perfect because in the second case you treat the modifiers differently, which is counter-intuitive, while in the first case "equal" modifiers don't cancel out, which is also counter-intuitive.

Of course, you could discard the "modifier" thinking and just refer to them as multipliers out of the box. So instead of "adds 20% production" you say "increases your production by a factor of 1.2" but this is somewhat aesthetically unpleasing, a bit harder for most people to understand and would scare people who have Mathematicaphobia.

All things considered, I would sacrifice the "equal modifiers cancel out" condition as the less evil one when broken, and in fact, producing another counter-intuitive effect. That is due to an act of renormalisation our brain performs unless the modifiers are very visible all the time: We get used to paying half price for things when we do it long enough, so when it's supposed to get 50% more expensive, we assume that this will put us at 1.5 times the current price, rather than at the one we started with. You can try this effect for yourself by playing with Industrious or Philosophical in Civ4 - when I play with Industrious I'm always a bit confused why wonders and forges are so expensive in the next game I play without it, even though I know why.

 

[Ahriman]

Whats even worse about using iterative multiplicative multipliers is that -40% is not the same as -20% and -20%. 1*0.6 != 1*.8*.8

This is why I suspect the game will use additive modifiers, except for rare occasions like some of the UAs.

Agree that there are already some counterintuitive things if you look closely at Civ4 mechanics, as in the "builds 50% faster" mechanic. Builds 50% faster doesn't actually mean builds in half the time, when other modifiers are present.

I'm also impressed that Civ5 appears to be avoiding a few of the problems created by iteratively stacking bonuses by making bonuses work differently; eg, the library gives extra research per population, rather than just a multiplicative bonus.

Compare:
Situation A: Size 10 city, libraries give +50% beakers, university gives +50% beakers, additive.
10 base research, 15 with library, 20 with university. So the +50% from the university is really only +33.3%.

Situation B: Size 10 city, libraries give +0.5 research per citizen to base beaker income, university gives +50% beakers.
10 base research, 15 with library, 22.5 with university.

 

[KrikkitTwo]

Well, I think it depends on how the bonus is Listed.

eg.
India says "half unhappiness from population". Which I think means 1/2 before or after any +- modifiers

so pop 11 Indian freedom theocracy with 2 specialists=9 unhappiness from regular pop+1 unhappy from 2 specialists=10 unhappy -20% unhappy=8 base pop unhappiness
for 4 pop unhappiness (Indian bonus applied last)

Wheras the American Ability (-25% gold for buying tiles) would stack with Monarchy.

So an tile that normally cost 40 gold would be -10 gold (American), -20 gold (Monarchy) for a total of




So India, Greece, Freedom SP, 2x Experience SP=all multiplicative

Almost everything else (Persia's combat boost, America's cheap gold, most SP and building effects, increased SP/GP costs from cities, are additive)

Not sure:
China's Great Generals


Basically %->additive
Half, double->multiplicative

 

[Ahriman]

Thats plausible - but of course, that's not how it worked in Civ4. Wonders that said build double speed with <resource> added +100% to base hammers while producing.

12 hammers with a forge and making a wonder at double speed had 12*2+3=27 hammers per turn, not
(12+3)*2 = 30 hammers per turn.

 

[KrikkitTwo]

Thats plausible - but of course, that's not how it worked in Civ4. Wonders that said build double speed with <resource> added +100% to base hammers while producing.

12 hammers with a forge and making a wonder at double speed had 12*2+3=27 hammers per turn, not
(12+3)*2 = 30 hammers per turn.

True, but otherwise a Theocratic, Legalistic Indian capital gets a total of Negative 0.03 unhappiness from every population member.

 

[alpaca]

Whats even worse about using iterative multiplicative multipliers is that -40% is not the same as -20% and -20%. 1*0.6 != 1*.8*.8

This is why I suspect the game will use additive modifiers, except for rare occasions like some of the UAs.

Agree that there are already some counterintuitive things if you look closely at Civ4 mechanics, as in the "builds 50% faster" mechanic. Builds 50% faster doesn't actually mean builds in half the time, when other modifiers are present.

I'm also impressed that Civ5 appears to be avoiding a few of the problems created by iteratively stacking bonuses by making bonuses work differently; eg, the library gives extra research per population, rather than just a multiplicative bonus.

Compare:
Situation A: Size 10 city, libraries give +50% beakers, university gives +50% beakers, additive.
10 base research, 15 with library, 20 with university. So the +50% from the university is really only +33.3%.

Situation B: Size 10 city, libraries give +0.5 research per citizen to base beaker income, university gives +50% beakers.
10 base research, 15 with library, 22.5 with university.

Well a lot of that particular part of being unintuitive can be solved by providing a good and clear UI representation. Just display all the modifiers and the end result without expecting the player to add them by herself and she'll be fine.

Avoiding the issue like that works, but only for a game that is as streamlined (or, should I say, flavourless) as vanilla Civ. You can't easily find different bonus classes for each building you add, and if you did this would increase the game's complexity by lots.

Well, I think it depends on how the bonus is Listed.

eg.
India says "half unhappiness from population". Which I think means 1/2 before or after any +- modifiers

so pop 11 Indian freedom theocracy with 2 specialists=9 unhappiness from regular pop+1 unhappy from 2 specialists=10 unhappy -20% unhappy=8 base pop unhappiness
for 4 pop unhappiness (Indian bonus applied last)

Wheras the American Ability (-25% gold for buying tiles) would stack with Monarchy.

So an tile that normally cost 40 gold would be -10 gold (American), -20 gold (Monarchy) for a total of




So India, Greece, Freedom SP, 2x Experience SP=all multiplicative

Almost everything else (Persia's combat boost, America's cheap gold, most SP and building effects, increased SP/GP costs from cities, are additive)

Not sure:
China's Great Generals


Basically %->additive
Half, double->multiplicative

I agree with Ahriman, this is not unplausible - but still conjecture. It's not how it worked in previous games, or similar games, so why do you assume that's how it works in Civ5? Furthermore, doubling something having a different meaning than +100%ing something would be confusing.

True, but otherwise a Theocratic, Legalistic Indian capital gets a total of Negative 0.03 unhappiness from every population member.

Negative modifiers are usually capped at 0. I have seen 0 reached in some complex mods, though.


To say something new for the topic:Our discussion about the nature of modifiers notwithstanding - what do you think are the indirect effects of this combination (which apply no matter what way you apply the mods)?

I could well imagine that, when you're India, the 20% (or 40%) bonus might actually not be so useful for you. Your unhappiness from population will be low anyways, so the theocracy bonus might just be unnecessary and, like building Colosseums even though you already have good happiness, you might be better off choosing something else.

On the other hand, it could also be that this bonus would be what pushes a metropolis strategy for India just over the top from being possible (i.e. not so harmful you're going to suffer large problems because of it) to being viable (actually being on pair with expanding).

From what I've read so far, the Firaxians are trying to balance the game so that a metropolis strategy is viable with any faction, so the first doesn't seem entirely unreasonable. Of course, this could be mitigated if Golden Ages via excess happiness are useful enough.

 

[Ahriman]

what do you think are the indirect effects of this combination (which apply no matter what way you apply the mods)?

I could well imagine that, when you're India, the 20% (or 40%) bonus might actually not be so useful for you. Your unhappiness from population will be low anyways, so the theocracy bonus might just be unnecessary and, like building Colosseums even though you already have good happiness, you might be better off choosing something else.

I think it really depends; after India's bonus, how much of your civ-wide unhappiness is coming from population? If its still high (which it might be, if you have a large population), then the bonus is worthwhile. We don't really have a good feel for this yet.
The real question is: "by how much does adopting this policy reduce my overall unhappiness", not "by how much does adopting this policy reduce my unhappiness from population".

The other thing is; to what extent is happiness still your binding constraint on growth? If food is a more important constraint, then you might want to focus on other things. Like conquest....

 

[KrikkitTwo]

I think it really depends; after India's bonus, how much of your civ-wide unhappiness is coming from population? If its still high (which it might be, if you have a large population), then the bonus is worthwhile. We don't really have a good feel for this yet.
The real question is: "by how much does adopting this policy reduce my overall unhappiness", not "by how much does adopting this policy reduce my unhappiness from population".

The other thing is; to what extent is happiness still your binding constraint on growth? If food is a more important constraint, then you might want to focus on other things. Like conquest....

However, India is likely to be focusing more on population.

For all other civs
1 Food= 1/2 useful pop + 1/2 unhappiness
For India
1 Food=1/2 useful pop + 1/4 unhappiness

Food has a greater net yield for India, So I can see India focusing on food more than others.

If India has 2x the population, then
1. It will have the same change in happiness
2. India will need the Happiness more since they have more unhappiness from the cities.