Specialists

[EmperorFool]

I was playing as Isabella but only found one natural wonder (maybe there was the GBR but I don't think so), and I started with the objective to win science. I won it about ten turns after Korea built their first SS part, on turn 306 in 1872. I never win my science games early in Vanilla. However, they had hit the Future Era long before I entered the Modern Era. I don't know why they didn't build a SS and win. Instead they attacked Persia with whom they were great friends for the first half of the game.

I will say that it was much easier because Lizzy decided to expand along our border and never built a city toward me. Monty was between Korea and me, and everyone on the continent attacked him so we took out his three cities. He had attacked me earlier but I was able to buy a bunch of units (eight? ten?) which is that army I spoke of earlier. I kept them unpromoted the whole game until near the end when I got really bored and started buying stuff in bulk.

I've attached the initial save if you're interested.

Spoiler Details :

I had a very good position in that my close neighbor Monty was separated from me by a two-tile wide bridge, a lake, and a long mountain range. Korea was on that side but not bordering me. Persia was to the north but again too far away to border me. And Lizzy far to the west and settled north-east toward Persia. She never filled in the area between us, letting me delay expansion a bit. I did spike my third city near her capital to nab two luxuries and block her expansion, but it was pretty far past her third city IIRC.

 

[Txurce]

That's interesting - I played Spain once over six months ago, with a version of VEM that led to earlier science victories, and the early gold led to a win so lopsidedly early that I always assumed something was wrong with that game.

If you launched in t306, then you probably weren't cranking population in your cities. That's one thing you could have spent your gold on! That said, it's still a really impressive overall result.

 

[EmperorFool]

Yes, I should definitely have built those other cities earlier and focused more on population. I could have spent my cash much earlier on buildings. I didn't start buying buildings until the Hospital was available. I already felt that I was kicking the AI while it was down.

 

[Thalassicus]

I just thought of an out of the box idea.

What if science came from villages instead of population? This has precedent from Civ 4 cottages providing commerce, used for both gold and science.

• Reduces gold income.
• Keeps villages valuable.
• Shifts power from undeveloped cities (population) to developed cities (improvements).

In other words...

Currently:

• Each 1 gives 1 for free
• Each citizen working a village gives 2

Suggestion:

• Each 1 working a village gives 1 1

particularly reducing Golden Age frequency and length - with the new happiness=1% mechanic, the GA boosts are redundant

This is a great idea, and I'll do it right away! For a year now I've tried altering golden ages to improve the value of excess happiness. It clearly hasn't worked. I think adjusting the science-from-happiness bonus will be more effective.

I'd suggest:

1. mid- to late-game RAs need to be more expensive relative to units and such,
2. mid- to late-game maintenance costs need to go up across the board,
3. CS bribes/"gifts" should cost more late
4. Gold buildings like the Market and Bank may need to be nerfed,
5. Golden ages should give less of a Gold boost, and/or
6. Gold tile yields [for luxury resources in particular, I'd say] need to come down. (I really like VEM's "no bonus Gold for river tiles until improved" rule change for this reason, among others.)

Your overall analysis is exactly my thinking as well. It's a difficult situation to resolve since our point on the curve depends on circumstances of the game.

1. This is easy to do, and an approach I'm pursuing with recent changes.
2. I've mostly focused on unit maintenance costs, which are something the AI handles very well. The problem with building maintenance is early buildings are still available late. There's no way to alter an early building's cost in late game. We can alter empire-wide maintenance, but not in a way the AI can deal with.
3. I've tried altering over-time CS costs a lot. We're unable to do it effectively with our current modding tools.
4. I could reduce the flat gold value on markets/banks. I don't want to weaken them too much. After all, the problem is we don't have enough incentive to get them in the first place.
5. I agree with this and will add some changes for v132.
6. This is somewhat difficult since they're already very low, and we can't do fractional yields. If I could I'd make river gold 1/2 . I'd reduce coastal gold, but that would devalue coastlines in general, which would cause other problems.

However! To add to #6 I could reduce some sources of mid-game gold boosts. For example, I could remove the Harbor +1 on water tiles, and eliminate the +1 boosts located on technologies. Do you think that would help?

At the crux of the matter is that the AI determines gold value via resource trading: 4=240, always.

This would be true if we only got the happiness for 1 turn. However, these deals last 30 turns. 240:c5gold upfront equals 8 each turn, for 4 each turn. The ratio is 2:1. The AI is set to provide somewhat more value to gold than the goal, to encourage players to keep luxuries instead of selling them off.

The Garden (erroneously?) has an Artist slot

This is unintended, left over from when the Garden could only be built along rivers. Thank you for pointing out this bug!

=============

I'll reply more later, but have some things to take care of right now.

 

[Txurce]

I just thought of an out of the box idea.

What if science came from villages instead of population? This has precedent from Civ 4 cottages providing commerce, used for both gold and science.

• Reduces gold income.
• Keeps villages valuable.
• Shifts power from undeveloped cities (population) to developed cities (improvements).

In other words...

Currently:

• Each 1 gives 1 for free
• Each citizen working a village gives 2

Suggestion:

• Each 1 working a village gives 1 1

Out-of-the-box is the way to go when tweaking the same old neural (and balance) pathways isn’t getting us anywhere. This approach is definitely worth exploring.

The first element is the reduction of gold income from villages, increasingly affecting wide empires more than tall ones, and narrowing the gold gap between them. It will also have a philosophically pleasing effect on those bloated AI treasuries. I think this is a fantastic idea, and would implement it right away. It’s possible that all costs should be scaled down slightly as a result, particularly in successive eras – but we could tinker with this after group testing.

The second element – shifting science from cities to villages – raises the question of “excess” population. It can be used to fill specialist slots, but beyond this it contributes nothing except unhappiness. I don’t know how damaging this would be to the AI. For the human player, the megalopolises typical of tall civs could lose some of their edge. However, decisions on farming vs villages + maritimes, still exist and are at least as meaningful.

I don’t know if some of the points I raised about the science element can be explored without testing. If they can’t be, I think both elements are worth trying, perhaps in a beta (!), so a new batch of tall vs wide comparisons can be run.

This is a great idea, and I'll do it right away! For a year now I've tried altering golden ages to improve the value of excess happiness. It clearly hasn't worked. I think adjusting the science-from-happiness bonus will be more effective.

I like the idea of toning down GA's, given the science boost happiness provides.

I could reduce the flat gold value on markets/banks. I don't want to weaken them too much. After all, the problem is we don't have enough incentive to get them in the first place.

However! To add to #6 I could reduce some sources of mid-game gold boosts. For example, I could remove the Harbor +1 on water tiles, and eliminate the +1 boosts located on technologies. Do you think that would help?

Reducing the incentive to build these buildings works against complexity of choice, which is where I think a lot of the game's true strategy lies. Every time we reduce the harbor, for example, it quits being built. I think all buildings need to remain strong - if anything, more necessary than ever. This would be the case if you go with your out-of-the-box idea... which is the main reason I like it so much. It keeps improvements fun and meaningful, but reduces aggregate issues.

 

[Sneaks]

Right now the question of whether to build a Village or Farm rarely if ever exists thanks to Villages being so much better. Add in science to them, and I doubt I would ever build a farm again.

 

[Txurce]

Yes, that's why I think reducing the gold seems more of a no-brainer, while the science element is worth deliberating on.

In theory you would want farms (or Maritimes) to feed specialists and hill workers, with lumber mills and villages in plains being pop-neutral. That leaves villages on grassland feeding a hill worker or specialist. Is this sufficient to negate the need for farms to feed all the hill dwellers and specialists? The goal should be to keep food low enough that a strong specialist economy without farms (or Maritimes) isn't possible.

 

[Sneaks]

I would really not be against making all villages zero food.

 

[Zaldron]

This is pretty amazing. Forget gold - you did everything you wanted to do, including building perhaps every GW you wanted, racking up 20K g, and did it with a "minimal" 15-unit army that's bigger than just about any I ever have when not playing Germany. It reads as if you were playing on Warlord level, but you weren't. Is anyone else having results like this on Emperor?

And out of curiosity, what VC were you pursuing and in what turn did you win?

I *just* finished a cultural win on emperor with France a couple days ago. It was 1625 AD on standard, around turn 230 I believe. This was my fastest win ever by about 50 turns. I never had happy problems and sold like 6 copies of some resource leaving me at 0 pretty much the entire game. I used the money to buy up almost every CS. Then the next game I got bum rushed by Indian elephants and lost my capital, so I'm still content with the risk v. reward.

One thing I'll note is that the +2 culture is a magnificent benefit in the early game because you get the 20% wonder production way earlier.

 

[Zaldron]

I just thought of an out of the box idea.

What if science came from villages instead of population? This has precedent from Civ 4 cottages providing commerce, used for both gold and science.

• Reduces gold income.
• Keeps villages valuable.
• Shifts power from undeveloped cities (population) to developed cities (improvements).

In other words...

Currently:

• Each 1 gives 1 for free
• Each citizen working a village gives 2

Suggestion:

• Each 1 working a village gives 1 1

If you move science from population to villages that could amount to a significant nerf to tall empires. Consider a city with <= 20 villages going from pop 20 to 40. In the current mechanism this would double the science (which can be substantial with NC in your capital). In the village scheme it would be....no increase at all....so there is *much* less incentive to grow new citizens rather than stagnating and plopping down new cities to work every single tile you can.

On a somewhat orthogonal note, are non-riverside farms intended to be a useful improvement to build? Currently I never build them even with fertilization. What about putting a science bonus on *farms*?

 

[Txurce]

I would really not be against making all villages zero food.

This makes sense - farms feed the villages, which produce wealth and (less logically) science - and should restore a meaningful basic choice. The net result is that you have to work harder for both food and gold, making the game more challenging. If it turns out to be too challenging, moving the slider shouldn't be too hard.

 

[Txurce]

If you move science from population to villages that could amount to a significant nerf to tall empires. Consider a city with <= 20 villages going from pop 20 to 40. In the current mechanism this would double the science (which can be substantial with NC in your capital). In the village scheme it would be....no increase at all....so there is *much* less incentive to grow new citizens rather than stagnating and plopping down new cities to work every single tile you can.

On a somewhat orthogonal note, are non-riverside farms intended to be a useful improvement to build? Currently I never build them even with fertilization. What about putting a science bonus on *farms*?

But isn't science currently a function of building multipliers on pop, and to a lesser degree happiness? If so, then wide civs with their higher populations had an edge, which this change would reduce. At the same time, typically tall high-pop cities can afford more specialists... which would especially be the case if villages gave zero food.

(I think putting science on farms would just feel too weird!)

 

[Zaldron]

But isn't science currently a function of building multipliers on pop, and to a lesser degree happiness? If so, then wide civs with their higher populations had an edge, which this change would reduce. At the same time, typically tall high-pop cities can afford more specialists... which would especially be the case if villages gave zero food.

(I think putting science on farms would just feel too weird!)

Don't underestimate the power of the NC, and now the happiness bonus in your capital. Baiscally moving science to villages would have no effect on wide empires (they still work the exact same number of villages and have no cities with more people than can reasonably work them) while tall empires are significantly nerfed in science (because the extra population provides no science).

Let's assume that each city can potentially work up to 15 villages for science.

As an example, a tall empire with three cities of 40, 30, 30 would have 100 base science from pop. With the new mechanism, that would be more than halved, down to 45 science.

The wide empire with a capital of 20 and 8 cities of 10 would still be able to work 95 villages, extracting almost the same amount of science as before the change (a small 5% reduction).

 

[Txurce]

Don't underestimate the power of the NC, and now the happiness bonus in your capital.

I see what you're saying with regard to the above. I had assumed that buildings would still work based on pop... but that population itself wouldn't add to science. In this case the tall civs' more likely early use of the NC, and eventual use of Public Schools and Research Labs, would still benefit (to an adjustable degree) from higher population. I also assumed that the happiness bonus would persist, further adding to tall civs' science output.

 

[Txurce]

I've always felt that happiness was too easy to acquire, and the present consensus seems to be that it contributes to the game being too easy overall. One way to do this and benefit tall civs is to reduce the happiness benefits of buildings, and increase the happiness benefits of NW's and GW's. The ideal way could be to increase the unhappiness of an empire with each successive city, but that may not be possible. There's something to be said for having to struggle to stay in the black.

 

[Seek]

Rather than argue these points, I'll step back and bring up purpose. Thal made a lot of these changes to allow tall civs to compete with wide ones in Science games, as per the "3 Playstyles" tenets. So I would say the question isn't whether there's too much gold (rather than just more gold) - but whether tall civs are now using their increased gold income to blow wide civs out of the water scientifically. This is what I am still testing, and it's where I think the focus should be.

As per above, just saying "I have too much gold" may just be a matter of playstyle. It seems to me that I've always marveled at how much gpt you have late game!

The last two games I've played were both OCC, and I daresay it's overpowered, if nothing else extremely easy, because culture is so readily available - costs don't go up from extra cities, every Wonder helps tremendously (I think lowering the Tradition bonus on Wonders to 1 might be a good idea), all food and culture from CSs is funneled through modifiers, etc. I was acquiring policies every 8-10 turns for much of those games, which leads to rapid bonuses, especially since Piety and Enlightenment don't interfere with each other any more.

In both games I was been able to ally every single CS on the map somewhere around T150, so I think it is empirical at this point to say that there is too much gold available. Which is definitely related to this:

I've always felt that happiness was too easy to acquire, and the present consensus seems to be that it contributes to the game being too easy overall. One way to do this and benefit tall civs is to reduce the happiness benefits of buildings, and increase the happiness benefits of NW's and GW's. The ideal way could be to increase the unhappiness of an empire with each successive city, but that may not be possible. There's something to be said for having to struggle to stay in the black.

Happiness is a non-issue in a small empire, so much of the gold made is from luxury sales, gpt isn't all that high (open borders with large AIs helps though.) I don't think nerfing wide empires is something we need to focus on right now.

(I'll reply to other posts later, but I have to go now.)

 

[Txurce]

The last two games I've played were both OCC, and I daresay it's overpowered, if nothing else extremely easy, because culture is so readily available - costs don't go up from extra cities, every Wonder helps tremendously (I think lowering the Tradition bonus on Wonders to 1 might be a good idea), all food and culture from CSs is funneled through modifiers, etc.

In both games I was been able to ally every single CS on the map somewhere around T150, so I think it is empirical at this point to say that there is too much gold available.

Happiness is a non-issue in a small empire, so much of the gold made is from luxury sales, gpt isn't all that high (open borders with large AIs helps though.) I don't think nerfing wide empires is something we need to focus on right now.

Looking at the three basic points above, I would conclude that there's too much of everything - and not isolate on gold. And there has been a creeping inflation of benefits, especially the combined effects of some GW's and SP's. This happened either as a response to Thal asking what's the most popular(x), or else just miscellaneous complaints. There could well be a benefit to scaling all these down, if the game has become too easy overall. The alternative would be to make the unimproved game more difficult, so accumulated benefits have a higher mountain to climb.

The point about nerfing wide empires has to do with the parallel conversation about changes to VEM making Liberty (an therefore wide) the way to go over Tradition (and tall). This would apply to any dialogue - be it about gold, culture, science or happiness - about making the game tougher.

 

[orangecape]

One thing regarding money that I think is critical is how much money we make from selling junk to the AI. How often do you buy happiness from them for 240G? Never?

The trouble is that temporary happiness boosts aren't good. They do encourage golden ages but they don't raise your population cap because you can't rely on them coming back later on; the civ selling them may be dead, at war, or something else.

I found it was tremendously useful to halve the money coming from selling luxuries. At 120G each my income was slashed dramatically, which I found made gold much more relevant. The weird thing is that the base game doesn't value gpt and raw gold equally in vanilla (I don't know if VEM deals with this, it is a minor issue) so I had to do some weird stuff to make it do so in my mod.

Basically since we all know that we sell luxuries and would never buy them clearly the price is too high. Fix that and you go a LONG way to cutting the supply of gold and making it as valuable as other yields. I have the math and the code behind the gpt and raw gold changes, just PM me if you want it.

 

[Zaldron]

One thing regarding money that I think is critical is how much money we make from selling junk to the AI. How often do you buy happiness from them for 240G? Never?

Actually I'll pay money for luxuries all the time in the mid to late game. Specifically if a CS wants it or one of my cities wants it for WLTK it's clearly worthwhile to spend cash to get the benefit from the lux in addition to the free happiness it brings along for the ride.

 

[wobuffet]

To add to #6 I could reduce some sources of mid-game gold boosts. For example, I could remove the Harbor +1 on water tiles, and eliminate the +1 boosts located on technologies. Do you think that would help?

Absolutely. Experimenting with taking away the +1 Tech boosts sounds like a great idea. (I don't have good enough a sense of coastal vs non-coastal cities to comment on the Harbor; maybe someone else does?)

I've always felt that happiness was too easy to acquire, and the present consensus seems to be that it contributes to the game being too easy overall.

This sounds right.

One way to do this and benefit tall civs is to reduce the happiness benefits of buildings, and increase the happiness benefits of NW's and GW's.

My impression is that tall civs already have it too easy in terms of Happiness and Culture, while wide civs have it too easy in terms of Science and Gold.

Obviously the goal shouldn't be to make all things equally easy for empires of all sizes, but the wide-tall gaps often seem too large. Some combination of jiggling building yields and adjusting/replacing Social Policies could help greatly here.

For instance, reducing or removing lots of %Culture boosts/multipliers and increasing Culture buildings' base + yields would help wide civs compete with tall ones in terms of Culture per turn. The opposite would hold for and , obviously.

One thing regarding money that I think is critical is how much money we make from selling junk to the AI. How often do you buy happiness from them for 240G? Never?

Great point. Slashing AI players' valuation of luxuries would be a fantastic way to make (each unit of) Gold more valuable.

Suggestion:

• Each 1 working a village gives 1 1

I would really not be against making all villages zero food.

Interesting! If it were up to me, I'd say

• rescale/devalue beakers by raising all Science beaker yields, Tech costs, etc. by 75% or so (as discussed in the Yield Equality thread),
• keep per-Citizen yields at 1 per (to preserve some wide/tall balance, see below),
• make Villages -1, +1, +1 at the beginning of the game, increasing to 2 at some point. (0 for Villages regardless of terrain would really hurt cities in Grassland),
• nerf Jungles somehow (2 base is a lot), and
• make Mountains workable for 2 each (and maybe +1 and/or +1 eventually. And I'd make Observatories do something like +2 for Mountains, +15% overall... Cities by lots of Mountains could be useful for non-Inca players!)

To avoid exacerbating the wide-empire Science advantage too much (as moving base beakers from population to tiles would presumably do), I'd also consider adding/revising a Social Policy or two to boost Science output of tall empires (e.g., +15% Science in all Cities with a National or World Wonder).