Feedback: Improvements

[Simon_Jester]

Random not-yet-thought-through idea: What if Plains gave +1 instead of +1? I could even set it so that flatland Plains gave 1/0/1, while hilled plains (gah!) still give 1/1/0.

Hm, I'm not sure I like this. For one thing, it really body-slams city productivity in flat terrain that isn't heavily forested.

That said, it's plausible something like this could be achieved via terrain replacement, though that doesn't solve the issue of Labourers being unable to move into Mountain tiles, and thus the AI's ability to place them sensibly.

But yeah, if I could I would.

Could you code an event that allows you to pay to turn a random peak into... no, that would be useless, since it would almost never appear where creating a pass would be helpful.

 

[Xyth]

The only problem is making them passable to workers but not to other units...

Making them passable to units at all is difficult; even with routes placed on them they're impassable to all but helicopters. The only solutions I've ever seen are SDK based.

I was thinking that it might be possible by building each end of a tunnel on tiles adjacent to mountains*, then have the mountains tiles between each end converted the passable version. I doubt I could ever teach the AI to make effective use of this though.


* Even just doing that requires enabling a Python callback that would cause turn time slowdown.

What effects would Savannah have then?

Hm, I'm not sure I like this. For one thing, it really body-slams city productivity in flat terrain that isn't heavily forested.

Yeah, good points. Disregarding that idea then!

 

[Azoth]

One aspect that I've noticed so far: there's only one Plantation resource (Incense) that ever appears in Desert; every other Desert resource is mined or quarried. How useless do we want Deserts to be? This scheme has the potential to increase their yield by quite an amount.

I think Deserts will remain fairly marginal, even with these changes. Deserts provide no food and the only improvements we are considering will add a little commerce (Plantation) or hammers (Quarry). I expect that Deserts will be no more productive than specialists; and in most cases, much less, considering the civics that boost specialist output, and the GP points they generate. So I wouldn't worry about it.

I would, however, direct you to some slight changes to Desert terrain I suggested a while ago:

Spoiler :

Speaking of terrain and terrain features, I have some suggestions for Deserts and Flood Plains. As it stands, Deserts produce no food, hammers, or commerce. However, they have no other drawbacks, making them ideal locations for cities: founding on any other tile wastes its terrain yield. By contrast, Flood Plains are terrible locations for cities: they revert to Desert when settled and therefore lose their +1 food bonus. Here is what I propose instead:

Desert
2 movement cost
-25% defense modifier

Flood Plains
+3 food, -0.50 health, +1 river commerce
1 movement cost
are not destroyed when settled

This way, Deserts are truly inhospitable for both explorers and settlers. At the same time, Flood Plains make for very desirable city locations, but they are especially vulnerable to attack, since they retain the -25% defense modifier from Desert. They also have a slightly higher health penalty, at -0.50 instead of -0.40. This further opens the door for Desert-specific promotions, namely:

Nomad I: +25% defense in Deserts, available to Recon/Mounted/Gunpowder units
Nomad II: +25% attack into Deserts, double movement in Desert, requires Nomad I

These promotions could add lots of flavour to various Mediterranean/West Asian unique units. They would also balance out the other terrain promotions. Melee units could be Woodmen, Archery units could be Guerrillas, Mounted units would be Nomads, while Recon and Gunpowder units could be all three.

Briefly, I feel:

• Flood Plains should not be destroyed when settled. (The game actively discourages you from settling along Flood Plains at the moment.)
• Deserts should carry a movement and defence penalty. (Stationing and supplying armies in deserts was and is difficult, when compared to grassland or plains.)
• The Nomad promotion should be introduced as a counterpart to Woodsman and Guerilla. (It can now be extended to include Savannah!)

Tell me: is there any way you could design an improvement to affect the yield of an adjacent improvement? For example, would it be possible for Mines to increase the yields of all adjacent Workshops by 1, or Windmills to increase the yields of all adjacent Farms by 1?

Not in a way that the AI could understand it.

Are you saying that it IS possible?
In that case, I suggest we try it even if the AI cannot understand it fully. It would add so much depth to the improvement system!

Frankly, the AI cannot grasp any number of concepts but still manages well enough. For example, the AI knows that it should found cities that claim resources. It also knows that cities must be spaced at least three tiles apart. But it doesn't know, in practice, that if it founds a city here, it invalidates all these other city locations, and thus the possibility of working that resource over there. So you sometimes find that the AI has ignored an isolated resource in its territory. In the same way, the AI might not know directly that Mines increase the yield of adjacent Workshops, but it will come to see that Workshops placed near Mines are stronger than others; and, hopefully, over time, it will work only those Workshops and replace the others. (I find that the AI is constantly replacing improvements: Cottages with Workshops, Mines with Windmills, and vice-versa.)

Shall I include bonuses for adjacent improvements in my next draft?

We could allow Quarries on hills or Desert. The problem is that no one would work Desert Quarries if they produced only 1. Is it possible to adjust yields by terrain type, such that Quarries produce +1, and an additional +1 on Desert?

Not in a way the AI would understand.

Hmm. Lumbermills produce +1, along with +1 along rivers. Are you saying that the AI doesn't understand that Lumbermills along rivers are stronger than others, and that it should construct these first? Does it matter if it doesn't? It would be nice to have the option of varying improvement yield by terrain type.

Speaking of which:
Is it possible to design improvements such that they cannot be built on adjacent tiles; that they must be at minimum two tiles apart?

As for Mines I want to retain their because it's realistic and because we can still afford to up the amount of in game. Mines can only be built on hills so the only improvement they compete with in most situations is the Windmill. You can't build cottages in hills so that comparison is mostly irrelevant. Mines enable (and can discover) some of the most useful resources in game, so that's one big advantage they have over Windmills already. Possibly enough to offset the by itself?

The proper comparison should be between Hills and specialists. Once all the resources have been improved, you're often left deciding between a Grassland Hill Mine and a Plains Farm, or a Grassland Hill Mine and a Grassland Hill Windmill. Assuming +1 is equivalent to -1 (and it will be if Cottages and Workshops also produce ):

Mine: -1/+2 (+1 with Redistribution)
Farm: +1 (+1 with Agrarianism)
Windmill: +1/+1

The Farm and Windmill produce 2 more than the Mine, enough to support a specialist. A Farm+Priest or Windmill+Engineer have the same output as a Mine but are better in every other way. Specialists produce +2. Specialists offer higher yields under certain civics. Specialists are inherently more flexible: you can swap the Priest for a Merchant or another improvement at will. That's why I don't think Mines should produce . In my last game, I built not a single Mine and was no worse for it.

A couple of reasons, I think. Agrarianism is a big one.

Another big reason is that with camps, you can basically put your entire population to work in production-rich forest tiles. In vanilla BtS, for every citizen working a forest-plains and producing two hammers you need someone growing three food. Indeed, I can't think of any terrain that doesn't have either special resources or improvements with high technology that gives you two hammers without also giving you a food shortage.

Build a camp on a forest-plains square, and suddenly your citizen is feeding himself and making two hammers a turn.

You're right. I hadn't thought about that.
Here's an idea: what if Forest Orchards produced +1 and Forest Camps produced +1 instead?
Orchards would require irrigation so fewer high yield forest tiles would be possible.
(In similar fashion, Plantations could take the place of Camps in Jungle and Savannah.)

Hmm. The easiest way to achieve most of that is to simply have them improve via technology, much as they do now. Lacks the time element but I guess there's little point doing the ton of extra coding to try and integrate the mechanic realistically.

I was thinking much the same thing. If you want to limit the growth of Workshops by era, then you might as well link Workshop bonuses directly to technology. That way, you also avoid the "too neat" symmetry of two improvements that grow stronger over time.

 

[Xyth]

I think Deserts will remain fairly marginal, even with these changes. Deserts provide no food and the only improvements we are considering will add a little commerce (Plantation) or hammers (Quarry). I expect that Deserts will be no more productive than specialists; and in most cases, much less, considering the civics that boost specialist output, and the GP points they generate. So I wouldn't worry about it.

Yeah, that's the conclusion I've come too as well.

Flood Plains should not be destroyed when settled. (The game actively discourages you from settling along Flood Plains at the moment.)

Not convinced by the need for this to be honest. It's just not realistic (desert cities aren't built on actual flood plains as they'll get flooded; they're built next to them). If you do build on a Flood Plain you lose the food when it is destroyed but you gain the fresh water health bonus and the commerce/trade benefits of being on a river. And if you build on a nearby desert tile instead you get to develop the entire flood plain, potentially having a better food supply in the long run.

Changing it just means that building a city on a flood plain is always the best idea - in fact it becomes the best of any terrain type (3 :food,: 1 , 1.5 !!!).

The defense penalty should be enough to dissuade building cities on isolated or edge desert tiles.

Deserts should carry a movement and defence penalty. (Stationing and supplying armies in deserts was and is difficult, when compared to grassland or plains.)

The Nomad promotion should be introduced as a counterpart to Woodsman and Guerilla. (It can now be extended to include Savannah!)

I like these changes.

Are you saying that it IS possible?
In that case, I suggest we try it even if the AI cannot understand it fully. It would add so much depth to the improvement system!

Sort of. I could do it by adding additional yield to the actual plot when an improvement is built, then removing it again when the improvement (or its relevant neighbour) is removed or replaced.

Shall I include bonuses for adjacent improvements in my next draft?

Lets try keep it simple for now, and perhaps we can look at this in a subsequent version.

Hmm. Lumbermills produce +1, along with +1 along rivers. Are you saying that the AI doesn't understand that Lumbermills along rivers are stronger than others, and that it should construct these first? Does it matter if it doesn't? It would be nice to have the option of varying improvement yield by terrain type.

The AI does understand about riverside improvements as that's set by a standard XML tag with the appropriate links to the DLL. There's no such XML tag for Deserts so I'd have to do it via the plot method described above. This means the AI won't understand that they'd get any additional yield because it doesn't exist until the improvement is already built.

Speaking of which:
Is it possible to design improvements such that they cannot be built on adjacent tiles; that they must be at minimum two tiles apart?

Yes, but it requires enabled a Python Callback that can severely degrade performance. Best avoided.



I'm working on a draft atm and I'll address on your other points with that when I post it.

 

[Simon_Jester]

You're right. I hadn't thought about that.
Here's an idea: what if Forest Orchards produced +1 and Forest Camps produced +1 instead?
Orchards would require irrigation so fewer high yield forest tiles would be possible.

That's a sensible idea and I approve. In real life, it's very hard to sustain large urban centers by (hunting) camps. The huntsmen may be able to feed themselves in large numbers if the game is good. But that's just equivalent to subsistence food production- in other words, getting two food off the tile. Getting a food surplus out of heavily forested land really isn't very practical, not on a scale large enough to feed major city centers with massive building complexes and specialist labor.

Whereas orchards- systematic agricultural cultivation of food crops- are more productive in terms of edible food.

There are two problems. One is that this displaces the commerce-production role I think the Nature Preserve is supposed to have... or am I confused? The other is that we're removing the primary way to increase food production away from river squares- which means cities away from rivers will revert to BtS sustainable size.

I was thinking much the same thing. If you want to limit the growth of Workshops by era, then you might as well link Workshop bonuses directly to technology. That way, you also avoid the "too neat" symmetry of two improvements that grow stronger over time.

I'd want to limit by era because an improvement that provides +4 hammers (plus civic bonuses) and can be built all over the place is blatantly overpowered in pre-Industrial times. You think hammer overproduction is a problem now...

 

[Xyth]

That's a sensible idea and I approve. In real life, it's very hard to sustain large urban centers by (hunting) camps. The huntsmen may be able to feed themselves in large numbers if the game is good. But that's just equivalent to subsistence food production- in other words, getting two food off the tile. Getting a food surplus out of heavily forested land really isn't very practical, not on a scale large enough to feed major city centers with massive building complexes and specialist labor.

Whereas orchards- systematic agricultural cultivation of food crops- are more productive in terms of edible food.

Do we also want Tundra Camps switching to commerce as well though? The food bonus works really well there, and is needed if we're not allowing riverside Tundra farms anymore. (Farms are made valid by proximity to fresh water - if we don't want Farms on certain terrain types we have to disable this and indicate the terrain/features individually)

I should also mention that Plains Forests will be quite rare on randomly generated maps. Most of them get converted to Savannah.

There are two problems. One is that this displaces the commerce-production role I think the Nature Preserve is supposed to have... or am I confused?

The Nature Reserve grants 1 commerce and 1 happiness to forest tiles so yeah, the Orchard couldn't compete with that. The Nature Reserve isn't available until the Industrial Era though.

The other is that we're removing the primary way to increase food production away from river squares- which means cities away from rivers will revert to BtS sustainable size.

In this scenario Orchards would require irrigation so you could only place them in forest adjacent to fresh water. We can also choose whether we'd want Orchards to spread irrigation or not.

I'd want to limit by era because an improvement that provides +4 hammers (plus civic bonuses) and can be built all over the place is blatantly overpowered in pre-Industrial times. You think hammer overproduction is a problem now...

We're now talking about leaving it mostly as it is in 1.17. So in the Medieval they only grant 1 or 2 (depending on civics) but by the Modern era it will have been boosted by technology to 4 or so.

 

[inasnapp]

That's a sensible idea and I approve. In real life, it's very hard to sustain large urban centers by (hunting) camps. The huntsmen may be able to feed themselves in large numbers if the game is good. But that's just equivalent to subsistence food production- in other words, getting two food off the tile. Getting a food surplus out of heavily forested land really isn't very practical, not on a scale large enough to feed major city centers with massive building complexes and specialist labor.

I totally agree. Speaking as someone who has lived in a rural area of the Untied States, hunters can AT BEST put meat, venison, wild turkey etc, next to beef and chicken in a supermarket meat section (and at higher prices,primarily because of the added work of hunting rather than farming meat) camps should be designed to only sustain a population, rather than allow it to expand.
as far as orchards are concerned, might they be able to be either: a, built on forests and upgradable to forest reserves (their maybe a conflict of purpose here if orchards provide food and reserves provide commerce) or be if an orchard raises the chance of a nearby tile (unworked) to spawn a forest, which would either allow them to either sustain themselves, or provide commerce via reserves or hammers via lumber mills.
on a related note I see little fundamental difference between a farm as one is traditionally understood (grows food or cash crops) and an orchard, which is essentially a farm for trees.

Likewise, what separates an orchard from a plantation?

 

[Simon_Jester]

Do we also want Tundra Camps switching to commerce as well though? The food bonus works really well there, and is needed if we're not allowing riverside Tundra farms anymore. (Farms are made valid by proximity to fresh water - if we don't want Farms on certain terrain types we have to disable this and indicate the terrain/features individually)

We could have, as mentioned, two different camps, one for each. The Tundra camp providing food makes a lot of sense, but the Grassland/Forest or Plains/Forest (and there still are such animals) being commerce instead still makes sense.

I should also mention that Plains Forests will be quite rare on randomly generated maps. Most of them get converted to Savannah.

True, but still... something to bear in mind. Camps mean you no longer have to choose so much between food and hammer output in the very early game, as does Agrarianism. Hence the massive boost in hammers, when you multiply it over many turns.

In this scenario Orchards would require irrigation so you could only place them in forest adjacent to fresh water. We can also choose whether we'd want Orchards to spread irrigation or not.

Good idea, I think- and perhaps Orchards should be available a little later on than Camps, so that it isn't the optimal Stone Age strategy to just park in the middle of the Black Forest and build a megalopolis as fast as happiness permits.

as far as orchards are concerned, might they be able to be either: a, built on forests and upgradable to forest reserves (their maybe a conflict of purpose here if orchards provide food and reserves provide commerce) or be if an orchard raises the chance of a nearby tile (unworked) to spawn a forest, which would either allow them to either sustain themselves, or provide commerce via reserves or hammers via lumber mills.

Having orchards promote forest expansion is a good idea, I think.

on a related note I see little fundamental difference between a farm as one is traditionally understood (grows food or cash crops) and an orchard, which is essentially a farm for trees.

Likewise, what separates an orchard from a plantation?

By that argument a farm is a plantation too...

There are some significant differences- although unfortunately, the civilizations which made the greatest use of tree-agriculture seem to have been the ones of the New World that were effectively annihilated by disease. The main reason is simply so that we have a more or less historically-realistic method for increasing food output from forest terrain, without clearcutting it.

(Hey, perhaps the Orchard can get an extra bonus from Environmentalism!)

 

[Xyth]

PRIMARY IMPROVEMENTS

Farm
Requires Agriculture
• +1
• +1 at Fertilizer
• Requires and Spreads Irrigation
• Can be built on Grasslands, Plains, Flood Plains (Flatlands only)
• Chance to produce Corn, Potato, Rice, Wheat​

Farms remain the only improvement that can produce more than 2 food from a tile without a resource. They're no longer the only improvement that requires Irrigation, but they'll still be the only one that can spread it. I might shift Flax to Farms but to do so requires some tricky art changes. Shall see.

Pasture
Requires Pastoralism
• +1
• +1 at Biology
• Can be built on Plains, Savannah, Tundra (Flatlands only)
• Chance to produce Cattle, Pig, Sheep, *Horse, *Elephant​

The key advantage of Pastures is that they do not require fresh water to be built, balanced by only being buildable on low food terrain. Elephants shift here from the Camp; they and Horses will be given -1.

Orchard
Requires Pottery
• +1
• +1 with Horticulture
• Requires Irrigation
• Can be built in Forest, Jungle (Flatlands only)
• Chance to produce Fruit, Olives, Spice, Wine​

Orchards (formerly Wineries) take over the role currently performed by HR's Camps, but are limited by requiring fresh water (and do not spread it themselves).

Plantation
Requires Calendar
• +1
• +1 with Crop Rotation
• Requires Irrigation
• Can be built on Desert, Savannah, Jungle (Flatlands only)
• Chance to produce Sugar, Cocoa, Coffee, Tea, Incense, Tobacco, Dye, Rubber, Cotton, Flax, Silk​

Like Orchards, Plantations require fresh water and cannot spread it themselves. They won't be particularly useful in Deserts but I can't see a way to boost them there that doesn't also make Jungle Plantations much too strong (remember that Jungle in HR doesn't have a food penalty and gives +1).

Camp
Requires Hunting
• +1
• +1 with Gunpowder
• Can be built on Tundra, Forest (Flatlands or Hills)
• Chance to produce Bison, Deer, Furs​

Camps change to commerce and expand to Tundra. This should help reduce the overabundance of production in the early game, and, along with the Plantation, should help getting the early game economy growing a bit sooner.

Mine
Requires Mining
• +1, +1
• +1 with Geology
• +1 with Railroad or Highway
• +1
• Can be built on Hills
• Chance to produce Gems, Gold, Silver, Copper, Iron, Aluminium, Coal, Uranium​

Mines now grant some commerce to distinguish them from Quarries. Both gain extra production at Geology.

Quarry
Requires Masonry
• +2
• +1 with Geology
• +1 with Railroad or Highway
• +1
• Can be built on Desert, Hills
• Small chance to produce Salt, Stone, Marble, Jade​

Quarries compete with Mines on Hills, and become the only improvement that can be built in Desert without restriction. Desert cities will need good health infrastructure to support them though. I didn't want Mines to have all the good mineral resources so I've shifted Salt and Jade to the Quarry. They made more sense than any other options.

 

[Xyth]

SECONDARY IMPROVEMENTS

Cottage/Hamlet/Village/Town
Requires Employment
• +1 each size
• +1 with Highway (Town only)
• +1​

No change from Azoth's original suggestion. Same as always, except all sizes cause the same unhealthiness and only Towns will benefit from Highways. For reference they can be built on any cleared terrain that provides at least 1 food or has access to fresh water.

Workshop
Requires Artisanry
• +1
• +1 at Machinery, Machine Tools, Chemistry
• +1 with Highway
• +1
• Time to build increases with each tech level​

Experimental. I'm tweaking these to make them the production equivalent of the Cottage. Rather than growing over time, the Workshop will increase output with new technology but will also take longer for Workers/Labourers to build. So in the Medieval era they're relatively weak but quicker to build, and by the Modern era they're more powerful but it takes a lot longer to build new ones. They have the same terrain conditions as Cottages except they are also restricted to flatlands.



COMMENTS

As always, all improvements can be built on a related resource, regardless of terrain and other conditions.

I've deliberately left out interactions with civics for the moment. I'm still working on that aspect and the improvements need to be relatively balanced on their own.

Wetlands and Peat are left out of the scheme. Just didn't feel realistic and we retain the idea of Wetlands being mostly unworkable until the second half of the game, which I like.

I'm still tinkering with the Watermill, Windmill, and Lumbermill. I don't think they need to change much, if at all. Not considering changes for any other improvements at this time,

I'm not going to use the event system or any custom coding for resource discovery, at least in 1.18. Will save a lot of time and effort and we can see if still feels necessary once we've played with the system for a while.

Let me know if you spot any problems or have suggestions.

 

[Azoth]

On the whole, I like your draft.

I think the Farm/Pasture and Orchard/Camp distinction works well;
I support the health penalty for Cottages and Workshops; and
I'm glad that we're moving towards a system in which all resources can be discovered.

I don't think I can properly judge the strength of improvements without reference to civics, but here are a few suggestions:


TERRAIN FEATURES

This is slightly off topic but since it affects improvement balance I thought I should mention it. Can someone explain the rationale behind Savannah, as a feature that adds 1 at the cost of 1? I wasn't around when it was implemented but I see it as further weakening Plains when compared Grassland. Grassland Forest and Jungle are strong terrain features while Plains Savannah is the equivalent of riverside Tundra! I suggest the following:

Flood Plains: +3, +0.40
Forest: +1, +0.25
Jungle: +1, +0.25
Savannah: +1, +1
Wetlands: -1, +1, +1, +0.40

Even with these changes, a Plains Savannah Pasture in 1.18 will be equivalent to a Plains Forest Camp in 1.17: hardly unbalanced, given its rarity. Meanwhile, Wetlands will produce a little more health, to mirror the health penalty from Flood Plains, and as compensation for a lack of early improvement options.


PRODUCTION

With Mines reduced to +1, +1, I no longer think they can compete with Workshops if they keep the health penalty. Workshops must be toned down; depending on civic modifiers, they must lose one or more technology bonuses and see the Highway bonus changed back to +1. Towns should always produce more raw commerce than Workshops produce hammers because commerce and hammers are not equally valuable. 1 is roughly equal to 2, to judge by specialist slots. Engineers produce 2, scientists and merchants produce 3 net, while artists and spies produce 5 net. As for Mines and Quarries, since you are more interested in retaining the health penalty on Mines, I suggest the following:

Mine
Requires Mining

• +2
• +1 with Metallurgy
• +1 with Railroad (not Highway)
• +1
• Can be built on Hills
• Chance to produce Gems, Gold, Silver, Copper, Iron, Aluminum, Uranium

Quarry
Requires Masonry

• +1, +1
• +1 with Geology
• +1 with Railroad (not Highway)
• No health penalty
• Can be built on Desert, Hills
• Chance to produce Salt, Stone, Marble, Jade, Amber

Mines can compete with Workshops on hammers but both suffer a health penalty. Quarries are a weaker production/commerce hybrid without a health penalty. For the sake of flavor and diversity, their bonuses are tied to different technologies, and apply only to Railroads, not Highways. Amber can also be discovered in Quarries. I think I like the resource split between these improvements: Mines produce metals, Quarries produce minerals. It's a good rule, even if life doesn't work that way.

 

[Azoth]

MARGINAL TERRAIN

If Desert Plantations will always be inferior to Quarries, you may as well disable them. Plantations should be allowed on both flatlands and hills; Jungle Hills cannot be improved otherwise. Meanwhile, Desert and Ice can be the site of Wells:

Plantation
requires Calendar

• +1
• +1 with Crop Rotation
• Can be built on Savannah, Jungle (Flatlands or Hills)
• Chance to produce Sugar, Cocoa, Coffee, Tea, Incense, Tobacco, Dye, Rubber, Cotton, Flax, Silk

Well
requires Refining

• +2, +1
• +1
• Requires Irrigation
• Can be built on Desert, Ice (Flatlands only)
• Chance to produce Oil, Gas

I don't think players should be able to build Wells on every Desert and Ice tile in their empire; that's why I asked whether you could code a minimum distance between improvements. "Requires Irrigation" will have to do instead. EDIT: If you're feeling adventurous, you might allow Plantations on Oases.


RESOURCES

I think you should move Flax to Farms if the art allows it. Tobacco is another candidate for Farms while Silk or Incense could be moved to Orchards. However, I disagree with Pastures for Elephants. Elephants were never domesticated and bred as beasts of burden. Young elephants were simply caught in the wild every year and trained. To be honest, I don't know what this change achieves. Pastures and Camps can both be built in Tundra, so Elephants might still be discovered there. Besides, Pastures and Camps already have a good mix of resources. (If anything, Bison are a better candidate for pasture. Bison were always hunted, but their Asian relatives, the yak and the water buffalo, were domesticated.)

Lastly, Nature and Marine Reserves should have a higher chance to produce resources. (Nature Reserves might also be enabled slightly earlier.)

Marine Reserve
requires Ecology

• +1, +1
• +1
• Can be built on Reef
• Slightly higher chance to produce Fish, Shellfish, Crab, Seal, Whale

Nature Reserve
requires Sociology

• +1 from river
• +1
• Can be built on Forest, Jungle (Flatlands and Hills)
• Chance to produce Bison, Deer, Elephant, Furs

 

[Xyth]

This is slightly off topic but since it affects improvement balance I thought I should mention it. Can someone explain the rationale behind Savannah, as a feature that adds 1 at the cost of 1? I wasn't around when it was implemented but I see it as further weakening Plains when compared Grassland. Grassland Forest and Jungle are strong terrain features while Plains Savannah is the equivalent of riverside Tundra!

My reasoning was that although Savannah is technically a terrain feature rather than a actual terrain type, it would act more like the latter. Most improvements can be built in Savannah without any chopping required for example. It was also meant to give +1:commerce by rivers but it looks like I forgot to switch that on.

Savannah: +1, +1

That's too strong, no other terrain combination has 4 base yield away from rivers. Perhaps +1 with no change at all. Although that's the same as Jungle, since they almost always appear on different base terrains the resulting yield would be different. Grasslands Jungle = 2/0/1, Plains Savannah = 1/1/1. I quite like that actually.

Meanwhile, Wetlands will produce a little more health, to mirror the health penalty from Flood Plains, and as compensation for a lack of early improvement options.

I was actually considering lowering Flood Plains to 0.25, it's not like Deserts have an overabundance of food that needs countering.

With Mines reduced to +1, +1, I no longer think they can compete with Workshops if they keep the health penalty. Workshops must be toned down; depending on civic modifiers, they must lose one or more technology bonuses and see the Highway bonus changed back to +1. Towns should always produce more raw commerce than Workshops produce hammers because commerce and hammers are not equally valuable. 1 is roughly equal to 2, to judge by specialist slots. Engineers produce 2, scientists and merchants produce 3 net, while artists and spies produce 5 net.

The thing you're overlooking there is that those specialists are giving specific commerces such as wealth and research; these get multiplied by +X% wealth/research/etc modifiers but don't get multiplied by +X% commerce modifiers. Because of this, and the loss of flexibility, I generally consider the relationship to be 2 = 3 = 3 (approximately). I think Firaxis does too, I've seen it in a few places.

It's much harder to equate production and commerce, as it varies so much over the course of a game and by playstyle. Production is a probably a bit stronger overall than commerce, but its definitely not twice as strong. That said, I have no issue toning down the Workshop somewhat.

Mines can compete with Workshops on hammers but both suffer a health penalty. Quarries are a weaker production/commerce hybrid without a health penalty.

With Plantations being dropped from Deserts, having the Quarry as the +1/+1 option makes more sense too.

For the sake of flavor and diversity, their bonuses are tied to different technologies, and apply only to Railroads, not Highways.

Good call on the techs.

There's a technical reason for these improvements being affected by both Railroads and Highways; each route has a designated 'level' to determine which can be built over which. Roads are set to 1, and Paved Roads are set to 2. I originally had Railroads and Highways both set to 3 but it turned out that this made it impossible to replace one with the other. The AI couldn't cope so I had to set Highways to 4 instead, and change all the improvements that get a bonus from Railroads to also get it from Highways. This fix is coming in 1.18.

Amber can also be discovered in Quarries. I think I like the resource split between these improvements: Mines produce metals, Quarries produce minerals. It's a good rule, even if life doesn't work that way.

Yeah it's not perfect but I think it works well. Not sure about Amber though.

If Desert Plantations will always be inferior to Quarries, you may as well disable them. Plantations should be allowed on both flatlands and hills; Jungle Hills cannot be improved otherwise.

Yeah I'll get rid of Desert Plantations, Desert civs can get their commerce from rivers and Quarries. As mentioned earlier the only Desert resource connected to Plantations is Incense so it's no great loss at all. Plantations on hills seems reasonable.

Given Deserts are now removed from the equation - should Plantations require irrigation?

Well
requires Refining

• +2, +1
• +1
• Requires Irrigation
• Can be built on Desert, Ice (Flatlands only)
• Chance to produce Oil, Gas

I don't think players should be able to build Wells on every Desert and Ice tile in their empire; that's why I asked whether you could code a minimum distance between improvements. "Requires Irrigation" will have to do instead.

I was wondering if we should add the Oil Well to the scheme. Your suggestion seems reasonable, perhaps Tundra should be added as well? Requiring irrigation won't work since Farms can't be built on Tundra and thus irrigation is not likely to make it near Ice tiles.

The is a deterrent from spamming them though; perhaps we lower their yield a bit too. The discovery mechanic only triggers on tiles that are worked by a city so unmanned wells are going to be polluting your city for no real benefit.

EDIT: If you're feeling adventurous, you might allow Plantations on Oases.

I did consider that, it's a bit tricky getting the art to work together. Perhaps something for the future. Sadly there's no performance-friendly way to allow Plantations next to Oases without making them buildable next to any fresh water.

I think you should move Flax to Farms if the art allows it. Tobacco is another candidate for Farms while Silk or Incense could be moved to Orchards.

Other than Flax I don't feel these fit.

However, I disagree with Pastures for Elephants. Elephants were never domesticated and bred as beasts of burden. Young elephants were simply caught in the wild every year and trained. To be honest, I don't know what this change achieves. Pastures and Camps can both be built in Tundra, so Elephants might still be discovered there. Besides, Pastures and Camps already have a good mix of resources. (If anything, Bison are a better candidate for pasture. Bison were always hunted, but their Asian relatives, the yak and the water buffalo, were domesticated.)

I shifted Elephants since Camps can only be built on Tundra and Forests, neither of which are appropriate terrain for Elephants. Pastures on the other hand, can be built on Plains, Savannah, and Tundra - 2 of which are very appropriate for Elephants. I agree that Camps makes more sense than Pastures, but I felt this required less suspension of disbelief than Elephants only ever being discovered in colder climates.

Lastly, Nature and Marine Reserves should have a higher chance to produce resources.

Makes sense.

 

[Azoth]

That's too strong, no other terrain combination has 4 base yield away from rivers. Perhaps +1 with no change at all. Although that's the same as Jungle, since they almost always appear on different base terrains the resulting yield would be different. Grasslands Jungle = 2/0/1, Plains Savannah = 1/1/1. I quite like that actually.

I take your point, but Wetlands are already 1/1/1. Maybe they could be bumped up to 1/2/1? I don't think that's too strong for a relatively rare terrain feature that cannot be improved or cleared for a long time. Most everything can hit 4 yield with improvements, after all. (And I still support the increase to +0.40. Players need some reason to keep Wetlands around until Nature Reserves arrive.)

I was actually considering lowering Flood Plains to 0.25, it's not like Deserts have an overabundance of food that needs countering.

Hmm. I think Flood Plains should probably stay at 0.40. They're already the most desirable terrain type and the only one with a net positive base food yield. Even counting the health penalty, it's 3 - 0.40 = 2.60. Also, variety is always good.

The thing you're overlooking there is that those specialists are giving specific commerces such as wealth and research; these get multiplied by +X% wealth/research/etc modifiers but don't get multiplied by +X% commerce modifiers. Because of this, and the loss of flexibility, I generally consider the relationship to be 2 = 3 = 3 (approximately). I think Firaxis does too, I've seen it in a few places.

It's much harder to equate production and commerce, as it varies so much over the course of a game and by playstyle. Production is a probably a bit stronger overall than commerce, but its definitely not twice as strong. That said, I have no issue toning down the Workshop somewhat.

Raw commerce can more valuable than specific commerce such as beakers and gold; but BtS allows only one commerce multiplier, Bureaucracy, so this is rarely an issue. Even then, BtS Towns produce +1, +7 at full strength. (That's the base yield, plus Printing Press and the Free Speech and Universal Suffrage civics.) Workshops, by contrast, top off at +4. (Again, that's base yield, plus Guilds, Chemistry, and the Caste System and State Property civics.) That's a 2:1 ratio, and I wouldn't say that Towns are too strong compared to Workshops, not when they take 70 turns to mature and are so vulnerable to pillaging.

With its extended technology tree but similar unit and buildings costs, HR is even more one-sided. Simply put, if Workshops produced roughly the same hammers as Towns produced commerce, I would be tempted to replace all my Towns with Workshops in the Industrial Era. I would build Wealth and Research most of the time, of course, but I could easily switch to military, Wonders, or spaceship parts as the situation demanded. With a full set of production multipliers, I wouldn't need to build Banks, Observatories, and the like. All those hammers could go straight into Research or military. That's why Workshops must be toned down. A 2:1 ratio is not necessary, but anything less than 3:2 is problematic, in my view.

There's a technical reason for these improvements being affected by both Railroads and Highways; each route has a designated 'level' to determine which can be built over which. Roads are set to 1, and Paved Roads are set to 2. I originally had Railroads and Highways both set to 3 but it turned out that this made it impossible to replace one with the other. The AI couldn't cope so I had to set Highways to 4 instead, and change all the improvements that get a bonus from Railroads to also get it from Highways. This fix is coming in 1.18.

Makes sense. That reminds me: "No gold cost for improvements" would make for an excellent leader trait or Wonder bonus. Is it possible to code?

Yeah I'll get rid of Desert Plantations, Desert civs can get their commerce from rivers and Quarries. As mentioned earlier the only Desert resource connected to Plantations is Incense so it's no great loss at all. Plantations on hills seems reasonable.

Given Deserts are now removed from the equation - should Plantations require irrigation?

I say 'No.' Plantations produce fairly little commerce so I don't see any reason to limit their spread.

I was wondering if we should add the Oil Well to the scheme. Your suggestion seems reasonable, perhaps Tundra should be added as well? Requiring irrigation won't work since Farms can't be built on Tundra and thus irrigation is not likely to make it near Ice tiles.

The is a deterrent from spamming them though; perhaps we lower their yield a bit too. The discovery mechanic only triggers on tiles that are worked by a city so unmanned wells are going to be polluting your city for no real benefit.

Hmm. On second thought, I agree that the health penalty should be a sufficient deterrent. My only concern is that once players adopt Environmentalism, they will be free to build Wells everywhere. It makes no sense but that's the way the health mechanics work. In any case, I don't think we can lower the yield any further; so late in the game, even Quarries produce +2, +1, counting the Geology bonus. If you allow Wells on Tundra, you might consider a -1 penalty: it would weaken Tundra Wells without affecting Desert or Ice Wells. But even that might be overkill.

I did consider that, it's a bit tricky getting the art to work together. Perhaps something for the future. Sadly there's no performance-friendly way to allow Plantations next to Oases without making them buildable next to any fresh water.

Question: Does anyone think it would be useful to have more terrain features that provide fresh water, like Oases? A Glacier, perhaps, on Tundra or Ice tiles? I find that Lakes and Rivers aren't always well distributed.

I shifted Elephants since Camps can only be built on Tundra and Forests, neither of which are appropriate terrain for Elephants. Pastures on the other hand, can be built on Plains, Savannah, and Tundra - 2 of which are very appropriate for Elephants. I agree that Camps makes more sense than Pastures, but I felt this required less suspension of disbelief than Elephants only ever being discovered in colder climates.

In this case, we'll have to agree to disagree. Elephant Pastures sound far more implausible to me than Tundra Elephants.

Makes sense.

What do you think of Peat being discovered in Nature Reserves? That would limit it to Forests, Jungles, and Wetlands. All told, I'd hate to leave such an important resource out.


FORTS

One last thing: Forts. I don't build them unless I need a canal or my Workers have nothing left to do. There must be a way to make them more interesting. Many mods feature Forts that upgrade over time, without being worked. Typically, it works like this: Outpost --(30 turns)--> Keep --(50 turns)--> Fort
At each stage, the defense, visibility, and unit healing bonuses increase. All Forts act as cities for combat purposes (i.e.: the City Garrison and City Raider promotions apply) but only the final Fort acts as a canal. In some mods, fully developed Forts even provide bonuses to friendly units on adjacent tiles, like Radar Towers in Civilization III. Is that an option for HR?

 

[Simon_Jester]

Mines can compete with Workshops on hammers but both suffer a health penalty. Quarries are a weaker production/commerce hybrid without a health penalty. For the sake of flavor and diversity, their bonuses are tied to different technologies, and apply only to Railroads, not Highways. Amber can also be discovered in Quarries. I think I like the resource split between these improvements: Mines produce metals, Quarries produce minerals. It's a good rule, even if life doesn't work that way.

I'm not sure about amber simply because it's ahistorical- you don't seem to find large amber deposits far inland in real life.

I approve of "railroads not highways," because again in real life, highway transportation is NOT an economical way to run a mine, with very rare exceptions. You need the railroad's ability to haul heavy freight along a predictable route for that.

Marine Reserve
requires Ecology

• +1, +1
• +1
• Can be built on Reef
• Slightly higher chance to produce Fish, Shellfish, Crab, Seal, Whale

Not sure I approve of the food bonus. One of the biggest things about a marine preserve in real life is that it restricts fishing, because modern fishing techniques are so all-consuming and destructive that they can depopulate whole local ecosystems in a few years.

There's a technical reason for these improvements being affected by both Railroads and Highways; each route has a designated 'level' to determine which can be built over which. Roads are set to 1, and Paved Roads are set to 2. I originally had Railroads and Highways both set to 3 but it turned out that this made it impossible to replace one with the other. The AI couldn't cope so I had to set Highways to 4 instead, and change all the improvements that get a bonus from Railroads to also get it from Highways. This fix is coming in 1.18.

Well then, I guess there's nothing to be done for it.

Given Deserts are now removed from the equation - should Plantations require irrigation?

Possibly, but associated resource bonuses shouldn't be- a farm produces less food without irrigation when built on a tile where fresh water isn't available. A plantation shouldn't produce less commerce when built on a resource on such a square;

I shifted Elephants since Camps can only be built on Tundra and Forests, neither of which are appropriate terrain for Elephants. Pastures on the other hand, can be built on Plains, Savannah, and Tundra - 2 of which are very appropriate for Elephants. I agree that Camps makes more sense than Pastures, but I felt this required less suspension of disbelief than Elephants only ever being discovered in colder climates.

In India and Southeast Asia, where tame elephants see the most use, they are often native to forests and jungles, and don't do so well on cleared lands that invite cultivation.

Tundra elephants- well, I like the idea of mammoths not necessarily being extinct.

 

[Xyth]

I take your point, but Wetlands are already 1/1/1. Maybe they could be bumped up to 1/2/1? I don't think that's too strong for a relatively rare terrain feature that cannot be improved or cleared for a long time. Most everything can hit 4 yield with improvements, after all. (And I still support the increase to +0.40. Players need some reason to keep Wetlands around until Nature Reserves arrive.)

I think increasing Wetlands to 0.40 or even 0.50 would be enough. Wetlands can appear on Grasslands, Plains, or Tundra, so they vary in yield quite a bit.

With its extended technology tree but similar unit and buildings costs, HR is even more one-sided.

Not disagreeing, but note that I've increased the cost of many later era buildings and wonders compared to BTS. Still a fair bit of balancing to be done in this area though.

Makes sense. That reminds me: "No gold cost for improvements" would make for an excellent leader trait or Wonder bonus. Is it possible to code?

Not directly, but it's easy to instantly refund the cost. Happens so fast the player won't notice their treasury changing but the only drawback is that the tooltip for the worker action will still display the original cost. I've coded something very similar to this for refunding corporation spread costs under the Free Market civic (more on this later).

At the moment its just the routes that have a gold cost to build, and the costs are fairly low. Is this something worth expanding?

I say 'No.' Plantations produce fairly little commerce so I don't see any reason to limit their spread.

Possibly, but associated resource bonuses shouldn't be- a farm produces less food without irrigation when built on a tile where fresh water isn't available. A plantation shouldn't produce less commerce when built on a resource on such a square;

I'll leave them without an Irrigation requirement for 1.18 and we can see how it feels.

Hmm. On second thought, I agree that the health penalty should be a sufficient deterrent. My only concern is that once players adopt Environmentalism, they will be free to build Wells everywhere. It makes no sense but that's the way the health mechanics work. In any case, I don't think we can lower the yield any further; so late in the game, even Quarries produce +2, +1, counting the Geology bonus. If you allow Wells on Tundra, you might consider a -1 penalty: it would weaken Tundra Wells without affecting Desert or Ice Wells. But even that might be overkill.

Lets try -1:food/+2/+1:commerce/+1 on Desert, Tundra, and Ice, no irrigation needed, and see how it feels.

In this case, we'll have to agree to disagree. Elephant Pastures sound far more implausible to me than Tundra Elephants.

In India and Southeast Asia, where tame elephants see the most use, they are often native to forests and jungles, and don't do so well on cleared lands that invite cultivation.

Tundra elephants- well, I like the idea of mammoths not necessarily being extinct.

I don't mind the occasional tundra or forest elephant, it's just when these are the only types of elephants being discovered that it becomes ridiculous. That's the issue with handling them via Camps.

What do you think of Peat being discovered in Nature Reserves? That would limit it to Forests, Jungles, and Wetlands. All told, I'd hate to leave such an important resource out.

It is a shame to leave Peat out but I don't think that's the right solution.

One last thing: Forts. I don't build them unless I need a canal or my Workers have nothing left to do. There must be a way to make them more interesting. Many mods feature Forts that upgrade over time, without being worked. Typically, it works like this: Outpost --(30 turns)--> Keep --(50 turns)--> Fort
At each stage, the defense, visibility, and unit healing bonuses increase. All Forts act as cities for combat purposes (i.e.: the City Garrison and City Raider promotions apply) but only the final Fort acts as a canal. In some mods, fully developed Forts even provide bonuses to friendly units on adjacent tiles, like Radar Towers in Civilization III. Is that an option for HR?

Upgrading Forts over time is a good idea; tricky to make it happen without being worked but I should be able to find a solution. Defense modifiers are easy to make upgradeable but the other aspects you list are pretty much hardcoded (basically, Forts have a single tag called bActAsCity that gives them these qualities, it can only be set to On or Off). There's bound to be ways of coding alternatives to this though, I'd have to do some research. Lets leave Forts for 1.19 though.

I'm not sure about amber simply because it's ahistorical- you don't seem to find large amber deposits far inland in real life.

Yeah, and certainly not in quantities worth mining or quarrying for. I'm happy to leave Amber as a marine only resource for now.

Not sure I approve of the food bonus. One of the biggest things about a marine preserve in real life is that it restricts fishing, because modern fishing techniques are so all-consuming and destructive that they can depopulate whole local ecosystems in a few years.

We've many marine reserves here in NZ and they help replenish fish stocks, not just in the reserve itself but in the surrounding waters too. That's what I consider the food bonus to represent here.



Had a thought - now that we've got new resources being discovered or produced, perhaps we could also introduce resource disappearance (at a lower rate than discovery of course). Thoughts?

 

[Azoth]

I approve of "railroads not highways," because again in real life, highway transportation is NOT an economical way to run a mine, with very rare exceptions. You need the railroad's ability to haul heavy freight along a predictable route for that.

Yes, this is what I was going for. The question is: will the AI understand that it shouldn't upgrade all its Railroads to Highways; that, in most cases, the extra hammer on Mines and Quarries is worth more than faster transport speeds across those tiles? If so, we can still set the bonus as "Railroads, not Highways." If not, we can't.

Possibly, but associated resource bonuses shouldn't be- a farm produces less food without irrigation when built on a tile where fresh water isn't available. A plantation shouldn't produce less commerce when built on a resource on such a square;

Ah, here's a question: does an improvement that requires irrigation necessarily receive a bonus from irrigation? Let me explain: at the moment, Farms require irrigation (except on resources) and derive their +1 base yield from that irrigation. This means that Farm resources without irrigation produce one less food than they would otherwise. It also means that if an irrigation chain is broken, any Farms affected will produce one less food (and so, in most cases, nothing) until the chain is restored. Will the same rule apply to Orchards? I think it should. Otherwise, players could build Farms to bring irrigation to Forests, build all the Orchards they need, then replace the Farms with higher yield Workshops or Cottages.

Separately, it might be nice to have a +1 bonus for Pastures with irrigation, even though irrigation would not be necessary to build those Pastures.

Not directly, but it's easy to instantly refund the cost. Happens so fast the player won't notice their treasury changing but the only drawback is that the tooltip for the worker action will still display the original cost. I've coded something very similar to this for refunding corporation spread costs under the Free Market civic (more on this later).

At the moment its just the routes that have a gold cost to build, and the costs are fairly low. Is this something worth expanding?

I don't think any other improvements should have gold costs, no. But "improvement costs refunded" would make for a nice leader trait. You could swap it in place of 1% interest on Financial leaders. It would be roughly equal in strength but more intuitive. (I've never liked the interest mechanic, anyway. I can't be sure of this, but I think spending accumulated gold reserves on deficit research, unit upgrades, or rush buys always provides better returns than collecting 1% interest. So, in that sense, it's counterproductive. It also requires fiddling with per era caps and is difficult to balance across game speeds.)

Had a thought - now that we've got new resources being discovered or produced, perhaps we could also introduce resource disappearance (at a lower rate than discovery of course). Thoughts?

Hmm. On the one hand, players don't respond well to penalties and the best games avoid them as much as possible. Resource depletion, no matter how rare, could be very frustrating - especially in the case of food resources, since it would bring city growth to a standstill. On the other hand, resource depletion is very real, and a natural counterpart to resource discovery. On balance, I would avoid it simply because it wouldn't be very fun. If you do decide to implement it, I would stick to commerce and production resources - say, all Mine resources. Empires might easily exhaust their supplies of Silver or Copper, but they would hardly run out of Corn or Pigs. In particular, I would avoid disappearing seafood resources, no matter how true to life. Seafood resources can't be discovered until the Industrial Era; and most coastal cities are nearly useless without their seafood bonuses.


One final thought: since Wine is the only "processed" food resource in HR, wouldn't it be interesting if it never appeared naturally on maps? Empires that wanted access to Wine would have to develop their own wine-making traditions: plant Orchards, and hope their people grew the right fruits to "discover" a Wine resource. That way, Wine would be an especially valuable trade good; and its discovery chances could be bumped up so that several empires would discover Wine in every game.

 

[Keinpferd]

Had a thought - now that we've got new resources being discovered or produced, perhaps we could also introduce resource disappearance (at a lower rate than discovery of course). Thoughts?

Hmm. On the one hand, players don't respond well to penalties and the best games avoid them as much as possible. Resource depletion, no matter how rare, could be very frustrating - especially in the case of food resources, since it would bring city growth to a standstill. On the other hand, resource depletion is very real, and a natural counterpart to resource discovery. On balance, I would avoid it simply because it wouldn't be very fun. If you do decide to implement it, I would stick to commerce and production resources - say, all Mine resources. Empires might easily exhaust their supplies of Silver or Copper, but they would hardly run out of Corn or Pigs. In particular, I would avoid disappearing seafood resources, no matter how true to life. Seafood resources can't be discovered until the Industrial Era; and most coastal cities are nearly useless without their seafood bonuses.

I agree with "all Mine resources", except the really important strategic ones. Bronze and horses I consider as all mine resources, and I don't like to see a mod feature taking them away from me.

"Players don't respond well", sounds like an interesting bit of mind control. I guess, you're right about that in many cases, yet sometimes, punitive elements in games aren't exactly nuclear fallout or a contagious disease: If Mario falls down the cliff for the tenth time, and the player is starting to get angry, this will double the amount of pleasure from beating the next difficulty level!

The standard counter argument against resource depletion has always been its effect on long term strategic planning. If you can't rely on an only iron resource to remain in place, you might discover, it's depleted just in the worst moment, when you needed it most.

But this, as said, can be easily dealt with by exempting the most vital strategic resources like bronze or iron from a realistic depletion feature. Other than that, allow every possible resource to deplete according to realism in written and rewritten history.

 

[Xyth]

Yes, this is what I was going for. The question is: will the AI understand that it shouldn't upgrade all its Railroads to Highways; that, in most cases, the extra hammer on Mines and Quarries is worth more than faster transport speeds across those tiles? If so, we can still set the bonus as "Railroads, not Highways." If not, we can't.

This is a pain to test conclusively. We can try it for 1.18 though and fix it for 1.19 if the AI doesn't cope.

Ah, here's a question: does an improvement that requires irrigation necessarily receive a bonus from irrigation? Let me explain: at the moment, Farms require irrigation (except on resources) and derive their +1 base yield from that irrigation. This means that Farm resources without irrigation produce one less food than they would otherwise. It also means that if an irrigation chain is broken, any Farms affected will produce one less food (and so, in most cases, nothing) until the chain is restored. Will the same rule apply to Orchards? I think it should. Otherwise, players could build Farms to bring irrigation to Forests, build all the Orchards they need, then replace the Farms with higher yield Workshops or Cottages.

Orchards will work just like Farms do currently.

Hmm. On the one hand, players don't respond well to penalties and the best games avoid them as much as possible. Resource depletion, no matter how rare, could be very frustrating - especially in the case of food resources, since it would bring city growth to a standstill. On the other hand, resource depletion is very real, and a natural counterpart to resource discovery. On balance, I would avoid it simply because it wouldn't be very fun. If you do decide to implement it, I would stick to commerce and production resources - say, all Mine resources. Empires might easily exhaust their supplies of Silver or Copper, but they would hardly run out of Corn or Pigs.

I agree that food resources depleting would be crippling and thus they shouldn't. Most wouldn't be realistic anyway; we can assume that one of the key differences between unimproved tiles/resources and improved ones is that the resources can be procured in a relatively sustainable way. So basically I think we can rule out most plant or animal resources straight away. It's really only the ones that are not fully domesticated or are hunted that make any sense depleting: Elephant, Bison, Deer, Fur, Seal, Whale. Most of those go obsolete anyway and that's probably a better mechanic for such resources.

Metals, minerals, and fuels, however, make a lot more sense depleting, in terms of both realism and gameplay (in my opinion). It creates an incentive to secure several sources of a resource in case one depletes and I can teach the AI this. Resources I'd thus consider depleting are: Salt, Amber, Jade, Gems, Gold, Silver, Copper, Iron, Aluminium, Peat, Coal, Gas, Oil, Uranium.

In particular, I would avoid disappearing seafood resources, no matter how true to life. Seafood resources can't be discovered until the Industrial Era; and most coastal cities are nearly useless without their seafood bonuses.

Actually I can't make Marine Reserves discover the sea resources after all. The discovery mechanic is tied the improvement that enables it. So Marine Reserves could only discover Whales and Seals, and if we wanted Fish, etc, to be discoverable it would have to be via Fishing/Harvest boats.

One final thought: since Wine is the only "processed" food resource in HR, wouldn't it be interesting if it never appeared naturally on maps? Empires that wanted access to Wine would have to develop their own wine-making traditions: plant Orchards, and hope their people grew the right fruits to "discover" a Wine resource. That way, Wine would be an especially valuable trade good; and its discovery chances could be bumped up so that several empires would discover Wine in every game.

There's a lot of scope for processed resources for sure, we can look at that in the future perhaps.

 

[Xyth]

PRIMARY IMPROVEMENTS

Farm
Requires Agriculture
• +1
• +1 at Fertilizer
• +1 with Agrarianism
• Requires and Spreads Irrigation
• Can be built on Grasslands, Plains, Flood Plains (Flatlands only)
• Chance to produce Corn, Potato, Rice, Wheat​

Pasture
Requires Pastoralism
• +1
• +1 at Biology
• +1 with Agrarianism
• Can be built on Plains, Savannah, Tundra (Flatlands only)
• Chance to produce Cattle, Pig, Sheep, Horse, Elephant​

Orchard
Requires Pottery
• +1
• +1 at Horticulture
• +1 with Professionalism
• Requires Irrigation
• Can be built in Forest, Jungle (Flatlands only)
• Chance to produce Fruit, Olives, Spice, Wine​

Plantation
Requires Calendar
• +1
• +1 at Crop Rotation
• +1 with Slavery
• Can be built on Savannah, Jungle (Flatlands or Hills)
• Chance to produce Sugar, Cocoa, Coffee, Tea, Incense, Tobacco, Dye, Rubber, Cotton, Flax, Silk​

Camp
Requires Hunting
• +1
• +1 at Gunpowder
• +1 with Redistribution
• Can be built on Tundra, Forest (Flatlands or Hills)
• Chance to produce Bison, Deer, Furs​

Mine
Requires Mining
• +2
• +1 at Metallurgy
• +1 with Railroad
• +1 with Redistribution
• +1
• Can be built on Hills
• Chance to produce Gems, Gold, Silver, Copper, Iron, Aluminium, Coal, Uranium​

Quarry
Requires Masonry
• +1, +1
• +1 at Geology
• +1 with Railroad
• +1 with Slavery
• +1
• Can be built on Desert, Hills
• Chance to produce Salt, Stone, Marble, Jade​