Amenity Formula

[King Jason]

I am on my phone right now so have to way to even start verifying this, but here I got an idea. Could it possibly be based on continents?

I can confirm that it is not based on continents. In this video, at the point where player the acquires 3 cities through peaceful trade with china and egypt, his empire has 5 luxuries (Incense, Silk, Truffles, coffee, and silver). His first two cities are on Vaalbara, and are receiving +5 to amenities. China's cities, Chengdu and Taiyuan are on America, yet Chengdu is recieving a +5 (shown as +4 because it's at 3 population) where Taiyuan, at pop 1 (no negatives modifiers from pop) is at +3.

4 of his cities including Chengdu and Nekhen are recieving +5, where the 5th city is recieving +3.

He's researching governments at this particular point in the video and none of his policies offer amenities.

So what is required, in this particular case, is an explanation as to why Taiyuan is only recieving a +3 instead of a +5. Other videos display similar inconsistencies. The example I decided to start off with, in Quill18's video, just happened to have a solution i was able to find. So far, for this one, and the others, I can't observe anything else that explains the drop off.

 

[Ulthwithian]

King Jason: If 4 of his cities are getting 5 Amenities, and his 5th city is getting 3 Amenities, then this suggests an 'N nearest cities' mechanic for Amenities, which makes the Aztec UA make more sense. For multiple copies, this is generalizable to 'N nearest cities per copy', with or without diminishing returns as others have suggested. I might point out that if this true, and diminishing returns are additive, then the Aztec UA is even crazier than it looks right now.

Let's say that there is a simple 4-3-2-1 progression for 'cities impacted by a Luxury', and the Aztec UA adds 2 to each of these. That means that the 'normal' Civ would get 7 cities of coverage from 2 copies of a luxury resource, while the Aztecs would get 11 cities.

As more of a headache, it is possible, but unlikely, that different Luxury resources have varying cities they affect.

There is the other issue that this system is in flux, and so it may not make sense yet from data.

The only thing I can think of in your exact example is tile distance, possibly.

 

[Seek]

Good stuff Jason, thanks! I feel like we are close, if not there yet. The main thing I see an issue with in your theory is: if number of cities affects how many luxuries are applied, why wouldn't the UI reflect this with it's own negative amenity category? It seems like a logical thing to do. But then, it's all rather odd!

There's also some strange things going on in Marbs later videos where he lost amenities after building the Arena (+1 amenity) in his cap; here the city is at -3 amenities, size 11 (so -5 for size) and has -1 for war weariness and he has 3 luxuries on turn 89, and then here, the turn he completes the Arena Rio has -6 amenities! All the same factors are in play: the city is still the same size, he has the same number of luxuries and there was no government/policy change.

Some sort of escalating war weariness perhaps? But then when he looks at Recife a moment later it has -2 amenities at size 8 (so -3 for size) which would suggest a -2 WW penalty. I'm starting to think maybe some of the UI display bugs might be showing up here, I don't know how else to explain it..

 

[King Jason]

The only thing I can think of in your exact example is tile distance, possibly.

I don't think it's distance - Taiyuan is about 26 tiles from the capital and Chengdu is about 20 tiles. There are only two luxuries that are physically close to Taiyun, they're in Chengdu, the last one that is physically harvested from the empire is all the way back in the capital. So that means we're missing the 3rd luxury bonus. If the other two are too far away to give Taiyuan +5, then the 3rd amenity bonus would have to be too far away too.

The Luxury count in the empire currently looks like this;

2 Coffee, 2 Silk, 2 Incense, 1 Truffle, and 1 Silver. 1 Incense is in the Capital, the other is from trade. 1 silk and coffee is from chengdue, the others are from trade. Truffle and Silver are from trade.

So if we were to say that Luxuries affect N cities, All 5 luxuries are affecting the first 4 cities in the empire. If we then imagine that those luxuries have been "used up" on those 4 cities - we could theorize that if the empire didn't have multiple copies, perhaps Taiyun would have 0 amenities. However, 3 luxuries have duplicates, and Taiyun is recieving +3 exactly. I'd need to cross-reference this but perhaps what is happening.

Also, though obvious, I think it's a logical assumption that resources from trade are "sourced" at the capital.

 

[KrikkitTwo]

This means no reasonable resource trading between civs. It's quite sad, because resource trading is the biggest part of diplomacy.

No it means plenty of reasonable trading if many copies of a resource are concentrated in one area.

If every 4-8 city empire has 20 copies of their 2 resources.... then there is trading

 

[stealth_nsk]

No it means plenty of reasonable trading if many copies of a resource are concentrated in one area.

If every 4-8 city empire has 20 copies of their 2 resources.... then there is trading

I said "reasonable" - so I mean there will be much less trading than in Civ4 and Civ5. If something is worthless for you and valuable for another civ, you're likely to trade it away. If the difference is not that high (both selling resources which could become useful and getting back resources which have really limited use), chances for trade drop a lot.

Not to mention it would require quite crazy map setup. How will you implement resource distribution so 1 civ have enough resource to trade away, while others have clear need in them? The only way it will actually work is to have each resource appear on very small territory so generally 1 city has it all. I don't think I like that.

 

[KrikkitTwo]

I said "reasonable" - so I mean there will be much less trading than in Civ4 and Civ5. If something is worthless for you and valuable for another civ, you're likely to trade it away. If the difference is not that high (both selling resources which could become useful and getting back resources which have really limited use), chances for trade drop a lot.

Not to mention it would require quite crazy map setup. How will you implement resource distribution so 1 civ have enough resource to trade away, while others have clear need in them? The only way it will actually work is to have each resource appear on very small territory so generally 1 city has it all. I don't think I like that.

Luxury Resources are already clustered. (Just imagine each tile of a civ v resource gives 4-8 copies)

The key thing is that civ 1 has more than it can use and civ 2 has less.
If a civ only needs 1 copies of a resource to get full benefit or 20 it doesn't matter.

 

[King Jason]

I said "reasonable" - so I mean there will be much less trading than in Civ4 and Civ5.

Pretty sure the the possible implementation you're addressing is almost the exact way it worked in civ4. Unhappiness was accrued locally, but luxuries granted happiness globally. So once an empire in civ4 had dye, it never needed more dye. Though, admittedly, it's been quite awhile. I may be wrong.

If that proves to be the case in civ6 as well, I don't see how less trading will occur.

 

[Ulthwithian]

Though the rest of your post may have tackled it. The Luxury count in the empire currently looks like this;

2 Coffee, 2 Silk, 2 Incense, 1 Truffle, and 1 Silver. 1 Incense is in the Capital, the other is from trade. 1 silk and coffee is from chengdue, the others are from trade. Truffle and Silver are from trade.

Also, though obvious, I think it's a logical assumption that resources from trade are "sourced" at the capital.

Yeah, this has to be it. So the understanding of the Amenities from Luxuries works as follows:

1) Each Luxury gives 1 Amenity to 4 cities by default.
A) Immediate Reaction: Luxuries are 'worth' 50% more to the Aztecs in terms of coverage which implies Aztecs want to be 'wide'. (Just here for completeness.)
2) It is still unclear how the 4 cities are chosen. Possibilities:
A) Nearest 4 to the luxury resource.
i) In the case of luxuries granted from something other than a tile, the 'home tile' of the luxury is considered to be your capital. (Assumption)
B) First N cities
C) Nearest 4 to the capital (including capital)

This has very specific impacts on trade:

1) The wideness of your Civ will impact how many copies of Luxuries you want. This is very intuitive. If I have 12 Cities, I should want more Luxuries than if I have 6 cities, in terms of copies.

2) The tallness of your Civ will impact how many different Luxuries you want. This is less intuitive but still effective.

So, in terms of trade, a 'tall'/'smaller' Civ will want to trade extra copies of Luxuries it has in exchange for Luxuries it doesn't have. A wide Civ, in the meantime, could very well trade a Luxury it only has one more copy of in exchange for another copy of a Luxury it does have.

This also works well with the way tall vs. wide actually plays out. A tall Civ is much more likely to have a smaller area, and if luxuries are very regionally placed (i.e., 'clumped') then they are likely to have more copies of fewer luxuries. A wide Civ is more likely (though not necessarily very likely) to have the opposite issue. As such, it looks like this system would certainly promote trade or... other avenues of increasing Luxury Resource access.

The only issue here is that there exists hard breakpoints, apparently. This may be a case of 'best way to handle it' but there you go.

Now I'm wondering if there will be a Civ that gets more Amenities from Luxuries, perhaps to fewer cities. (A way of creating CiV India in CiVI would be to say it gets more Amenities from Luxury Resources, but their Luxuries only reach to 1 fewer city.)

 

[King Jason]

Yeah, this has to be it. So the understanding of the Amenities from Luxuries works as follows:

1) Each Luxury gives 1 Amenity to 4 cities by default.

Well the civ in question was the aztecs, so luxuries likely give 1 amenity to 2 cities by default, Aztecs up that to 4. I'd need to source this with other videos though.

Another thing, while classical republic explained the discrepency of amenities across quill's empire for one portion of the video, the problem arises again even after he switches out of that governmens and has no amenity-related policies.

For example. At one point, his empire has 2 unique luxuries; 2 dye and 1 salt. Manchester has -1 amenities at pop 4, which means that either something is causing that city to be unhappy, or the luxuries in his empire aren't providing any amenities to the city. Two turns later Bristol, at 6 pop (-2 to amenities) is at 0 amenities, implying they're recieving both Salt and Dye. Also important to note that at this point there are now 7 cities in his empire. These views happen on turn 122 and turn 124 (video 7).

About 10 turns earlier (still with 2 luxuries)Leeds is also short by 1 amenity, and the two newly acquired Egyptian cities appear to be recieving at least 1 amenity from luxuries. However, war weariness could account for some of this, but if war weariness is empire wide. That might explain why Leeds is at -1 amenities even though there are two luxuries in the empire. However it only accounts for manchester's -1 if one of the luxuries has fallen off.

 

[Atlas627]

Is there some reason you aren't just looking at Amenities from Luxuries rather than the total number of amenities/required?

 

[Amrunril]

Looking at the screenshot, it seems odd to me that, while there are mechanics that provide "negative amenities" (bankruptcy and war weariness), none of them appear to be local in nature. I would have that one of the main advantages of a localized happiness system would be the ability to implement local penalties for factors such as recent conquest and religious/cultural tensions.

 

[King Jason]

Is there some reason you aren't just looking at Amenities from Luxuries rather than the total number of amenities/required?

Because I'm not playing the game and the only way this information is presented to us as viewers is via the city-bar whenever the people in these videos clicks on them. That's how, for example, I was able to draw the correlation that cities suffer -1 amenity at populations 3, 5, 7, 9, and possibly onward. Because that drop in amenity value happens in virtually every city across a number of videos - even in cases where luxuries are present. So at pop 2 with 1 luxury, a city may have +1 amenity, when we check back a few turns later at pop 3, with no luxury changes or any changes to governments or policies, the amenities are now 0.

What I'm trying to do now is deduce how amenities are spread across the empire because where the above system holds true, there eventually becomes inconsistencies where, say, in an empire that has two luxury resources, there exists a pop 5 city that is sitting at -1 amenities instead of 0. Implying that something is causing that unhappiness.

I'm comparing about 10 different videos and there's no consistent thread that accounts for skewed amenity totals apart from the concept that somehow when an empire displays that it has access to two unique luxuries, not all cities are benefitting from those two luxuries.

So I'm trying to figure out why that could be. Localized luxuries and distance are obvious first guesses but those were practically the first two things that I ruled out.

The current leading ideas are either:

Luxuries affect a certain number of cities. So if you have 1 dye in an empire of two cities they both have +1 amenities from luxuries; if you add a 3rd city, then it has no amenities from luxuries until you add another copy of dye.

The number of cities in the empire skews how luxuries affect it globally. Again, the first two cities in your empire may both recieve 1 amenities from the 1 dye that you have. If you build a 3rd city, then it has no amenities because it's amenity bonus is [the total luxuries you have, minus 1]. In this example, you need to add another unique luxury. Doubles don't add up.

Those two examples are just made up for the sake of displaying the idea behind both theories quickly, the numbers of how many cities a luxury might affect are not necessarily a reflection of my findings.

Basically, one idea favors a bit of a hoarding approach, where you'll need to find both variety of luxuries as well as keep up with the quantity. The other favors a "catch em all" approach where you just need a variety of unique resources, This is similar to past games.

Personally, I think it's closer to second idea. I'm not fully convinced that multiple copies of luxuries add anything to your empire, but there's certainly some evidence for it.

 

[stealth_nsk]

I have a feeling what whatever formula they are using now, it's bugged. Civilization games usually used quite simple formulas and the numbers we see so far don't look simple at all.

 

[rastak]

I have a feeling what whatever formula they are using now, it's bugged. Civilization games usually used quite simple formulas and the numbers we see so far don't look simple at all.

Either that or there is something fundamental we're missing.

 

[Cyon]

Any chance stove one with Twitter could all the developers how this works? Or do you think we will get a video explaining amenities soon?

 

[Lucius_]

I'm guessing we'll learn the full mechanics of amenities pretty soon, probably during one of their twitch LP's, like how they sort of showcased religion in the Brazil LP. Maybe they'll do Aztecs or another civ with a bonus to amenities and highlight how the system works a little clearer.

 

[pholtz]

Jason, when you say first two cities... 3rd city, etc. you seem to be referring to the order of their build or acquisition. But I noticed in the England video, the order of the city was also the size of the cities. Could it be that the luxuries get distributed to the largest cities?. So the two largest cities would get all the luxuries, the next two largest cities would get all but one of the luxuries etc. Does that work? (not sure how ties would be resolved).

Or one more possibility. You remove all the luxuries from all the cities, then distribute them to the cities based on lowest amenity to highest? So the cities that need them most would get the most?

 

[Stringer1313]

Whatever the formula ends up being, I really hope it is simple. It should not require hardcore-gaming-level spreadsheets to figure this out.

 

[kaspergm]

Whatever the formula ends up being, I really hope it is simple. It should not require hardcore-gaming-level spreadsheets to figure this out.

Agreed. A flat "N cities per copy of the luxury" seems both logical and simple to understand, so I hope that's the case rather than some of the original theories floating around here, which seemed both very confusing and illogical. If it's only 2 cities per resource, that is rather strict (and means you REALLY want to ally Zanzibar if you go wide).