December Balance Beta - December 3rd (12/3)

[Gazebo]

Classic case is something like the Hanging Gardens. The food and the Garden are restricted to the 1 city. This is well understood because local yields are annotated in a universally recognizable way, and the free Garden explicitly says "in this city"

Temple of Artemis has a global +10% food, and then says "+15% production towards archery units". This is difficult to understand because there are numerous policies and beliefs which provide bonuses to miliary unit production, and they affect the entire empire. Those policies do not explicitly say "in all cities" or "on empire", and neither does ToA. Furthermore, pairing the local bonus production with a global bonus food, and explicitly saying the food is global, but saying nothing about whether the production is local or global, makes a reader interpret both bonuses as global.

In other words:

• This problem is specific to military unit production bonuses, because global bonuses of that type are given elsewhere using the same annotation
• This problem is exacerbated by the wording of ToA's other effect, which undermines a user's ability to comprehend the building as not being strictly a global modifier.

So yeah, at a minimum, the help text needs to be more clear. Knowing now that the bonus production to archery units is only in that city, I would imagine that boosting that to +25% would be more than fair, but the global food bonus and free herbalist are what people build that wonder for.

I opened a ticket 15 hours ago on this subject. Last time I suggested a minor change to the DLL I had to make the change multiple times because branches were merged out of order. Someone with more free access to the github should make the change, because that convinced me that making my own pull requests isn't worth the hassle, unless I'm going to make several changes at once.

You're describing vanilla buildings, though - don't make this out to be a bug with VP, it's a change from vanilla descriptions. Let's just be clear on that. I'm happy to look at it, but there's a lot of grief being thrown around for what are vanilla holdovers.

Re: Github, your branch was out of date, I'm sorry that it caused problems but don't blame the master branch if your fork isn't up to date!

G

 

[pineappledan]

Didn’t mean for this to become such a bugbear, just a text improvement to one sentence on 4 different buildings and mayyybe a conversation about whether the buildings in question were avoiding scrutiny over their balance because of misleading text descriptions.

My branch was out of date because someone with writer’s access merged their own branch before mine was reviewed. I’m not blaming anyone, except for maybe Github for making it a pain in the ass to re-download and update branches, but it’s an unnecessary hassle

 

[Gazebo]

Didn’t mean for this to become such a bugbear, just a text improvement to one sentence on 4 different buildings and mayyybe a conversation about whether the buildings in question were avoiding scrutiny over their balance because of misleading text descriptions.

My branch was out of date because someone with writer’s access merged their own branch before mine was reviewed. I’m not blaming anyone, except for maybe Github for making it a pain in the ass to re-download and update branches, but it’s an unnecessary hassle

I don't disagree with the end goal, but let's make sure we're framing it correctly.

Also, yes, Github can be a menace, but it's the best we've got.

G

 

[CrazyG]

mayyybe a conversation about whether the buildings in question were avoiding scrutiny over their balance because of misleading text descriptions

I can't speak for everyone, but I'm aware of what is global and what is local, and those wonders look balanced to me.

 

[Leondegrance]

I can't speak for everyone, but I'm aware of what is global and what is local, and those wonders look balanced to me.

Please do not downplay the lack of transparency like Gazebo does. Matter of fact you have no idea. There are players who need accuracy in their decisions. Otherwise it aint fun.

 

[Leondegrance]

This is how VP is today. 43 Civs. Early era oh wow I have 64 happiness. Late era oh wow I have minus 80 unhapiness. In one word - broken. And nobody knows why the differential.

 

[ashendashin]

Please do not downplay the lack of transparency like Gazebo does. Matter of fact you have no idea. There are players who need accuracy in their decisions. Otherwise it aint fun.

Dude, this is getting out of hand. If the info presented in-game is such a problem to you, just open some issues on github and maybe update the wiki while you're at it.
Acting as if G is going out of his way to obscure game mechanics makes you look like another BiteInTheMark.

This is how VP is today. 43 Civs. Early era oh wow I have 64 happiness. Late era oh wow I have minus 80 unhapiness. In one word - broken. And nobody knows why the differential.

Well there it is. I suppose that counts as one of those reports G was asking for, at least.

 

[Leondegrance]

Are you throwing hot potatoes in my yard ?

 

[Leondegrance]

I am here about 4 months, or 3. All I seen is bug patches. Havent seen a thorough mechanical explanation. You blame me for it ? You asking me to cut it out ?

 

[ashendashin]

Are you throwing hot potatoes in my yard ?

You've suddenly been very aggressive about a lack of transparency. Ignore everything besides my first statement regarding github, if you like.

 

[Phuc Nguyen]

This is how VP is today. 43 Civs. Early era oh wow I have 64 happiness. Late era oh wow I have minus 80 unhapiness. In one word - broken. And nobody knows why the differential.

I’ve never got any problems with happiness. U just need a little of micromanaging imo. Like look at what causes unhappiness in every city and try to fix it. Connect cities asap to not get unhappiness from isolation. Rush for wonders that globally reduce Boredom, Distress, illiteracy, religious unrest... The only thing that is bad for happiness in the new patch is how AI value their luxuries. Before you only need to trade 1-1, plus a little gold if necessary, but now it’s harder for you to get a deal even if you’re willing to trade 2-1.

The mechanic of happiness in VP is quite complicated, but if u get used to it and pay enough attention then it’s very manageable imo.

 

[Leondegrance]

I dont need generic advise. I need the formula.

 

[Phuc Nguyen]

I dont need generic advise. I need the formula.

Why though? Do you need it to make better play? I don’t think it will help much because it’s much more complicated than what is described in the game already. Or do you need it just to argue with the mod developers? Then you don’t need to, because BiteInTheMark will do it for you anyway.

 

[Leondegrance]

Equations speak more than words. That why

 

[Minh Le]

Chill dude. Its a game, not mathematics

 

[Leondegrance]

Chill dude. Its a game, not mathematics

Fine. But if your empire goes from 64 happiness to minus 80 in just an era time then something is very wrong.

 

[Minh Le]

Seriously its not the problem everyone have. It appears that only you are suffering it. And im sure its not the problem with the mod either, you can check my current let's play I posted on the forum using the current patch. Happiness is tough, as it should be, but managable.

 

[Moi Magnus]

Equations speak more than words. That why

The code is accessible on Github. I've quickly read it, and tried to undestand it, but that's not a "simple formula":

For each yield, compute "iThreshold" and "iCityYield", and return the difference divided by 50. (0 if negative, at least 1 if positive, and with a max to prevent having more unhappiness than citizens).

"iCityYield" is simply the yield produced per turn by the city divided by the population.

"iThreshold" is computed from:
* The global static yield G
* A modifier M
With "iThreshold = G x (M+100)/100"

G is updated at each population growth to the new "delayed global average" DGA
The "delayed global average" is computed from the "global average" GA, but in not directly equal: each turn, the update does: DGA = 5% GA + 95% DGA. So the "delayed global average" increase slowly.
The "global average" is not a standard average. Take compute N = "number of cities x 40%, rounded down, +1", and take the Nth lowest city for that yield.

(The 40% can be changed, it is named "BALANCE_HAPPINESS_THRESHOLD_PERCENTILE")

M = P x (A+S+100) / (100+P+P/2) with
P = population
A and S are computed as follows:

"S" come from all the abilities/need reduction buildings/... It should be a negative value.

"A" has a base value of 420 (can be changed with BALANCE_UNHAPPY_CITY_BASE_VALUE and BALANCE_HAPPINESS_POP_MULTIPLIER).
It is increased by a value depending on the yield (34 or 25)
It is increased by a value "T" depending on the distance of the player to the tech average.
It is also increased by 25 (BALANCE_HAPPINESS_CAPITAL_MODIFIER) if it is the capital.

"T" is computed from the global tech average GTA (as before, it is the number of tech of the Nth lowest civ, were N = 40% of number of civs, rounded down +1) and the number of tech of the player TP:
T = (TP - GTA)^3 x 2
(The 2 is BALANCE_HAPPINESS_TECH_BASE_MODIFIER)

@Gazebo: the value "BALANCE_HAPPINESS_POP_MULTIPLIER" of 420 seems quite strange. It is added to all the other modifiers, making them mostly irrelevant. But maybe I've missed something.

 

[Leondegrance]

The code is accessible on Github. I've quickly read it, and tried to undestand it, but that's not a "simple formula":

For each yield, compute "iThreshold" and "iCityYield", and return the difference divided by 50. (0 if negative, at least 1 if positive, and with a max to prevent having more unhappiness than citizens).

"iCityYield" is simply the yield produced per turn by the city divided by the population.

"iThreshold" is computed from:
* The global static yield G
* A modifier M
With "iThreshold = G x (M+100)/100"

G is updated at each population growth to the new "delayed global average" DGA
The "delayed global average" is computed from the "global average" GA, but in not directly equal: each turn, the update does: DGA = 5% GA + 95% DGA. So the "delayed global average" increase slowly.
The "global average" is not a standard average. Take compute N = "number of players x 40%, rounded down, +1", and take the Nth lowest player for that yield.

(The 40% can be changed, it is named "BALANCE_HAPPINESS_THRESHOLD_PERCENTILE")

M = P x (A+S+100) / (100+P+P/2) with
P = population
A and S are computed as follows:

"S" come from all the abilities/need reduction buildings/... It should be a negative value.

"A" has a base value of 420 (can be changed with BALANCE_UNHAPPY_CITY_BASE_VALUE and BALANCE_HAPPINESS_POP_MULTIPLIER).
It is increased by a value depending on the yield (34 or 25)
It is increased by a value "T" depending on the distance of the player to the tech average.
It is also increased by 25 (BALANCE_HAPPINESS_CAPITAL_MODIFIER) if it is the capital.

"T" is computed from the global tech average GTA (as before, it is the number of tech of the Nth lowest civ, were N = 40% of number of civs, rounded down +1) and the number of tech of the player TP:
T = (TP - GTA)^3 x 2
(The 2 is BALANCE_HAPPINESS_TECH_BASE_MODIFIER)

@Gazebo: the value "BALANCE_HAPPINESS_POP_MULTIPLIER" of 420 seems quite strange. It is added to all the other modifiers, making them mostly irrelevant. But maybe I've missed something.

This is awsemo attempt to clarity. I endorse. Problem is it is almost incomprehensible to folks like me. This is coder language. Can we have it old fashion way with sigmas and signs please ?

I was thinking about a making a thread asking people to share their formula knowledge but I was afraid of similar "coder" problem.

 

[Minh Le]

My impression with the current patch is happiness is more punishing the wider you are, which is reasonable. However, I only play on standard map size so its not really a problem for me. For people play on larger map maybe it can be a problem. Like at one point you have huge possitive happiness but at other point you will have huge negative.