Improvements

[Txurce]

No problem.

I'm in favor of balance and choice. In the case of some of the TP discussion, I think the choices are muddying the clarity game waters need.

 

[Thalassicus]

To clarify a few limitations of the current tools, these options are not possible:

• Non-stacking bonuses (+1 to rivers or coasts, but not both)
• Building/tech bonuses to coastal water tiles (only option is "water" tiles, which includes coasts, oceans, and lakes)
• Inherent bonuses of fresh water or coast adjacency (only can boost near rivers, or boost coasts with specific improvements)
• Tech bonuses near coasts (only can boost fresh water with a tech)
• City/mountain/coast adjacency bonuses with tech (only can be from turn 1)

One thing to remember is AI considerations will be irrelevant when we get full sdk access - we can code it to recognize things.

I think the improvement habits city/river/coast adjacency encourage are super awesome and result in really realistic-looking development patterns, but unfortunately, it seems the AI isn't currently coded to read the per-improvement adjacency tables. When a worker is actually on a tile it does recognize the +4 potential from a trading post is more valuable than the +1, but it doesn't seem to properly "scan" tiles for where to send the worker. As such, when it needs food it sends it to a likely tile where it can get food without reading those tables.

For these reasons and the ICS concern (I also realized in my first test game the on-cost vs one-tile-away city issue Ahriman pointed out), I think it might be best to leave it with just river modifiers and the original harbor bonus. Still, thanks for the discussion and playtesting everyone! City/route/etc adjacency is an idea we can revisit once we have more tools.

Spoiler :

Happy new year, folks!

I'm somewhat sober again and wanted to give you some feedback on the "adjacency" topic. I've played a few hours with the new dev build.

As you might imagine, there have been some issues. Like so often, it is not a question of better or worse, but of which gameplay we want to achieve. I'll admit it right here: the way the current dev build handles tile yields (especially TPs) doesn't work that well. Let me explain why and make some suggestions:


There are 2 basic principles for balancing tile yields:

1) Improvements are equally strong, no matter where you build them (a feature that buffs all improvements equally doesn't count).

2) Improvements are clearly better when built in certain positions, then making them a clear favorite in this position.

Oddly, in vanilla ONLY farms are position-dependent. For all other improvements, the position doesn't matter at all (as long as they are allowed).

We have done 2 very different different things with the changes in the dev build. First we nerfed the position-dependancy of farms by making each improvement equally valuable next to rivers (which then no longer counts).

But then, we have added buffs to TPs for coast and city adjacency, without giving anything similar to the other improvements. Logically, TPs were a clear favourite in these positions now, just like riverside farms were before. TPs are so much stronger in quite a few positions now (when certain boosters stack) that building anything else there would be plain foolish. They also don't need any tech to get so strong, unlike mines/farms. We created a mechanic for TPs that we wanted to get rid of for farms.


The question is: What do we want?

Balancing the game around the first of the two design principles above is rather easy, but less interesting. We could just say that a river (and possibly a shore) provides +1 yield to ANY improvement. So rivers/shores would be "better" terrain, but all improvents would be equally viable there -> Simple balance!

First, the issue with the above way of balancing: All the improvements should get their buff around the same time/era! In the dev version, there was a huge problem with farms only being improved by rivers midgame after civil service, but trading posts had the bonus from turn 1!!! This has to be considered either way.


Secondly:
Is there a way to use adjacency for added fun/complexity while still being balanced?

I think there is! We only need to buff each improvement in a certain terrain. This could mean that farms profit from fresh water, mines profit from vicinity to a mountain and trading posts (=villages) profit from being next to a major city.
These are only examples, but what would we achieve?

- Certain regions favor certain city specializations
- A tile could be epecially suitable for multiple improvements, offering choice (*)
- Possibly more natural/realistic settling patterns
- More strategic choices
- More regional differences, not only river/non-river
- The AI should understand it quite well (it can count yields after all)


I have to think this through in detail, and I have to check the vanilla and TBM techtree for good timepoints to put these yield buffs. I'll provide details later!

In the meantime, feel free to state your thoughts!



(*) EDIT: That's actually the core element why I think this mechanic is better than the one we currently have in the dev build (and vanilla): If multiple improvements have areas where they are better AND those areas overlap, we have an interesting and varied choice very often. Some spots would buff no improvement, some a single one, some would even buff all improvements. But a huge amount of the tiles on a map would buff some, but not all improvements.

The combinations could be city+mountain vicinity, coast+river, river+mountain, city+coast, river+mountain+city, river+mountain+coast, ... The player would have an always different selection of boosted improvements - or he could decide to build an suboptimal/not boosted improvement for a certain reason.

I agree wholeheartedly with the design goals discussed by Tomice in his happy new year post, spoiler'd above.

However, for the same reasons discussed earlier in this post (only can do stacking bonuses, AI concerns) I think we'll have to hold off on adding much depth to the (way too simplistic IMO) improvement mechanics.

It's tough to decide on -1 for riverside farms vs +1 for other improvements, I'm still trying to figure out which might be a better solution. They're not strictly converses of one another.

 

[Txurce]

I like the idea of giving coastal tiles an inherent +1 like river tiles, that would even affect city centers (so as to not discourage building cities on the coast). I'll research this further when I have time.

If you do this, I would do it with a tech trigger like Sailing, rather than right off the top as you indicate later in your post.

 

[Thalassicus]

Yeah, I did some research and discovered neither option is available to us, so I revised the post. It looks like coastal adjacency is a cool idea we just don't have the tools for yet.

 

[Tomice]

I had a fear that the non-stacking bonus would be impossible but the AI issue surprised me. Well, a pity that it doesn't work! I was really fascinated by the concept, as you guys have surely realized.

Nice that you like it, though!

It's tough to decide on

• -1 on riverside farms?
  This works out more cleanly than 0 river base yields with +1 on non-farm improvements, and I think the AI isn't coded to recognize the importance of rivers if they don't have a yield bonus. The question is, should this -1 be inherent or at Civil Service? I'm not sure it makes sense for a tech to reduce your gold income, so I'm thinking perhaps it should be an inherent part of farms next to rivers... until you get civil service, you'd be trading 1g for 1f. This exchange has precedent: clearing jungles and forests to build improvements trades one yield type for another.
• +1 base yield on coast-adjacent land, like river-adjacent land.
• +1 on TPs at economics.
• +1 on mines and lumbermills moved to Machinery.
  This would delay the production boost somewhat, and also indirectly buff Crossbowmen (which are somewhat less-desirable than the knight, treb, or longsword techs right now). I do feel Metallurgy is too late though, and would improve an already-desirable beeline (Riflemen). Ideally we want to put bonuses like this on weak techs.
• +1 on mines at Combustion.
  This would have the desirable side-effect of buffing tanks, which several people (including myself) feel are still underpowered compared to mech infantry, even with the combat mod buffs.

I don't understand your thoughts about farms. -1 with civil service would be ok, but if you make it from start, farms would be weaker than any other improvement there before Civil Service.

I strongly support +1 on shoreside land tiles. Every vanilla guide tells you to stay away from coast. This makes no sense at all, flavor/realismwise.
Unlocking this (and maybe the river bonus) with sailing would have the benefit of making sailing an attractive very early tech, and making early naval scouting, fighting and settling more attractive.

I don't really think the production boost is too early with construction. It is a nice pair with CS. I might be a bad player, but I mostly beeline one of them, then the other, they are both important.

I'm also unsure about the other topics, can't help you there.



EDIT: The impossibility of buffing coasts like rivers is a huge disappointment!

 

[Thalassicus]

The problem with putting it on civil service is, say you have 10 riverside farms when you research the tech. It'll increase your food +10 and gold -10. That might seem sorta weird for your gold income to drop, which could be a problem if you were at or near 0 from trades.

Didn't realize anyone was reading the post as I was revising it, so I've deleted my old thoughts and copied them here.

• 0 base river yield
• +1 riverside resource improvements (like plantations, unchanged from vanilla).
• +4 palace
• +1 riverside non-farms at engineering. This would make riverside improvements be:
  • Farms: +2
  • TPs: +3
  • Mines: +1+1
• +1 mines and lumbermills moved to Machinery.
  This would delay the production boost somewhat, and also indirectly buff Crossbowmen (which are somewhat less-desirable than the knight, treb, or longsword techs right now). I do feel Metallurgy is too late though, and would improve an already-desirable beeline (Riflemen). Ideally we want to put bonuses like this on weak techs.
• +1 TPs at economics.
• +1 mines at Combustion.
  This would have the desirable side-effect of buffing tanks, which several people (including myself) feel are still underpowered compared to mech infantry, even with the combat mod buffs.

 

[Txurce]

The problem with putting it on civil service is, say you have 10 riverside farms when you research the tech. It'll increase your food +10 and gold -10. That might seem sorta weird for your gold income to drop, which could be a problem if you were at or near 0 from trades.

Didn't realize anyone was reading the post as I was revising it, so I've deleted my old thoughts and copied them here.

• 0 base river yield
• +1 riverside resource improvements (like plantations, unchanged from vanilla).
• +4 palace
• +1 riverside non-farms at engineering. This would make riverside improvements be:
  • Farms: +2
  • TPs: +3
  • Mines: +1+1
• +1 mines and lumbermills moved to Machinery.
  This would delay the production boost somewhat, and also indirectly buff Crossbowmen (which are somewhat less-desirable than the knight, treb, or longsword techs right now). I do feel Metallurgy is too late though, and would improve an already-desirable beeline (Riflemen). Ideally we want to put bonuses like this on weak techs.
• +1 TPs at economics.
• +1 mines at Combustion.
  This would have the desirable side-effect of buffing tanks, which several people (including myself) feel are still underpowered compared to mech infantry, even with the combat mod buffs.

This all sounds good. It's extensive enough to make sure it makes these notes make it into the readme!

 

[Tomice]

I agree that a drop of several GPT might be a hard-to-understand mechanic, also very hard to cope with if you suddenly drop deep into red numbers!

I needed some time to understand your countersuggestion: Rivers would only buff RESSOURCE improvements before medieval era - CS would then add food to farms, while Engineering would add gold to all other standard improvements (mines, LMs, TPs).

It might work, although I can hardly find any enthusiasm for the idea after the disappointment about coastlines - I really thought this idea would work This has so little to do with natural settling patterns.

The really good thing about it: It would make all improvements equally viable next to rivers, without timing issues.

On the other hand, I'm not sure if Engineering is the best tech to put such a massive boost on, the path is strong enough with most units on it, happiness buildings, mines, lumbermills and much more. It would also lead to machenery with it's production boost, making CS an unattractive alternative.

I think a naval tech would be better. While sailing is to early and the next one (compass IIRC?) would still be faster to get than CS or Engineering, I see strong benefits from putting a general boost on a naval tech. It would encourage shipbuilding and island-settling very early, which is historic (e.g. phoenicians). A river boost with a naval tech makes sense thematically, too.

Another thing: didn't you say rivers should have the bonus on the tile, not on the improvement? Else the AI wouldn't recognize river worth?


EDIT: I just thought about my current game with 1.09.3 dev, and didn't you have a TP boost for building it on a coastal tile? Is there really no way to make coastlines similar to rivers regarding gold? The above changes should work for coastal land tiles just like they work for rivers, shouldn't they?

 

[Thalassicus]

The reason for having an innate resource-improvement buff is UI limitations. Each individual improvement buffed by a tech is listed separately, therefore any tech which provides the +1g to resource improvements would be overwhelmed. I'm honestly not sure one way or another about the AI thing... they seem to ignore rivers in vanilla anyway, so I'm uncertain if the change to yields makes a difference.

The naval techs are sailing > optics > compass > astronomy. Optics might be an interesting place to put the gold boost. You still have to have trapping/mining to actually build the improvements, and going for optics + those techs means you have to delay calendar and the iron/horse reveals.

It's possible to put coastal bonuses on improvements, but not terrain. Basically I'd list every single improvement type in the coastal bonus table. This would be slightly different effects... 1) wouldn't affect the city center tile 2) undeveloped terrain wouldn't have the extra value. I can work around the first limitation by reducing Harbor maintenance.

Also, what I was saying about the -1 for riverside farms is you'd trade -1g +1f pre-CS. This is similar to clearing a jungle: -1f +1p. Still, I figure it's probably best to go a different route.


Something else I've been considering is instead of adding a bonus to mines in the industrial era, remove the bonus to lumber mills. Basically the net effect of the B-TI mod would be to move this lumbermill bonus earlier in the tech tree (from Steam Power to Machinery) and include one for mines there too.

Also, this interestingly ties in with Druin's suggestion a long time ago (in private messages), way back in September:

Tech: Fertilizer
Additional Effect: Freshwater farms +1
In Game Reason: currently, there is no incentive to research this tech when your empire has even relatively few rivers because it is so easy to allow multiple cities access to those riverside farms.
Real Life Reason: fertilizer is not a substitute for a source of fresh water in farming. Presumably it allows "low water yield" farms to produce slightly better crops and thus +1 . However, it would also allow high water yield farms to produce superior QUALITY crops that would demand a higher price on an international market. Therefore: more commerce.

This is what originally started thoughts of a Fertilizer buff. Since that time I've added the Aqueduct building to enhance city growth, and patch 1.135 also boosted growth. The original +1 vs +1 I discussed with him has a slightly different perspective now, therefore it's reasonable to reassess this change. Basically, we could use his original suggestion of a gold bonus (which would be a nerf, since gold is the least valuable yield per-unit) but move it to a different improvement class, and concurrent with the Civil Service tech (slight buff due to earlier availability).


To sum it all up and simplify, the new mod details would be like:

• +1 on Mines with Machinery.
• +1 on Lumbermills moved from Steam Power to Machinery.
• +1 on coastal improvements.
• +1 on riverside Trading Posts with Optics.
• Farming a desert now requires fresh water.
• Forts deal 1 damage to adjacent enemies and cost 1/turn maintenance (same as roads).
• Net yield of mining gold, silver, and gems is the same as vanilla, post-Machinery (basically, the 1 for resources is moved from the mine itself to Machinery)
• (GP improvement changes)

To put it another way, improvements would be like this (on top of normal terrain yields):

The mod would partially simulate the riverside trading post emphasis Tomice was trying for, while keeping the changes relatively minimalistic, along the goals of the mods. This would also solve concerns of relative mine vs lumbermill value that have been brought up several times in the past. I'd keep the +1 on water tiles for the Harbor to encourage settling on the coast instead of adjacent to it, and possibly buff the Seaport as well (perhaps with a +2 bonus for the city itself, like the Smokehouse's and Monastery's city+resource bonuses).

When we have full sdk access, I'd like to pursue the idea of adjacency-based improvement bonuses similar to the detailed ideas previously posted by Tomice, as an independent mod completely reworking CiV's overly-simplistic improvement mechanics.

 

[Txurce]

On a terrain side note, did you ever nerf the large horse and especially iron yields, while increasing the frequency of those tiles?

 

[Seek]

To clarify a few limitations of the current tools, these options are not possible:

• Non-stacking bonuses (+1 to rivers or coasts, but not both)
• Building/tech bonuses to coastal water tiles (only option is "water" tiles, which includes coasts, oceans, and lakes)
• Inherent bonuses of fresh water or coast adjacency (only can boost near rivers, or boost coasts with specific improvements)
• Tech bonuses near coasts (only can boost fresh water with a tech)
• City/mountain/coast adjacency bonuses with tech (only can be from turn 1)

One thing to remember is AI considerations will be irrelevant when we get full sdk access - we can code it to recognize things.

So a coastal land bonus is possible for buildings? I still think that a sea building would be a better way to go here, to enforce city placement on water. Since all the sea buildings are pretty weak imo, I'd suggest:

• Lighthouse - Limit to +1 on sea resource tiles *only* and add +1 on coastal tiles. Representing local fishing and transportation of goods.
• Harbor - Add +1 on all water tiles. Representing more efficient fishing techniques overall.
• Seaport - Add +1 on all water tiles. Representing int'l trade.

Re: Other TP changes; I played a game with the dev build, and honestly it didn't feel too different. There were a few cases of a "sweet tile, which improvement should I put here?" feeling - but not as frequently as I would have liked. The most noticeable changes were slower gold growth at the beginning of the game, and overall less AI gold throughout the game. I think if tile yield changes are tied to buildings, the AI is much more trainable and will take greater advantage of the changes.

[*]+1 on riverside Trading Posts with Optics.

How does optics justifiably boost river trading? Sorry, but it doesn't make sense to me.. I understand that sailing may be too early for this from a gameplay perspective, but it's much more logical.

@Txurce: The quantity of strategic resources have been lowered, but I have not noticed any increase of nodes. If anything, my last few games have been quite iron- and horse-poor.

 

[Tomice]

On a terrain side note, did you ever nerf the large horse and especially iron yields, while increasing the frequency of those tiles?

Yes he did, and I like it! Although the recent dev build reduces per-tile amounts of iron more than Thal said in his readme.

In vanilla, an iron tile gives you either 2 or 6 units. Thal said to halve it (which would be 1 or 3) while putting it on 50% more tiles, which would result in a 25% overall reduction. But I only found 1 or 2 iron per tile (standard ressource settings), which is a bit harsh.

I think 2 iron for every tile would be best. Still an overall reduction of 25%, but more reliable availability. Those single iron tiles are a bit underwhelming, nothing I'd settle in a desert for.

The recent game was very peaceful, I was friendly with most AIs and they were friends often with each other. I was in the situation very early that I'd piss of 5+ AIs when attacking any of my neighbours, so I decided to conquer the new world instead. What I wanted to say: I can't say much about actual warfare with fewer ressource units, or about other ressources than iron in general.

When we have full sdk access, I'd like to pursue the idea of adjacency-based improvement bonuses similar to the detailed ideas previously posted by Tomice, as an independent mod completely reworking CiV's overly-simplistic improvement mechanics.

I'm glad to hear this, although it's a pity we'll have to wait for months! I'll think a bit more about your suggestions for the current possibilities, there's much to be considered. Guess you're sleeping now anyway

How does optics justifiably boost river trading? Sorry, but it doesn't make sense to me.. I understand that sailing may be too early for this from a gameplay perspective, but it's much more logical.

@Txurce: The quantity of strategic resources have been lowered, but I have not noticed any increase of nodes. If anything, my last few games have been quite iron- and horse-poor.

Renaming the tech is the smallest concern here. The first working telescope wasn't invented before 1608 (http://en.wikipedia.org/wiki/History_of_the_telescope), so optics is in a weird place anyway! Even the compass was invented a few hundred years earlier... Just call the tech "transport ships" or whatever and it's fine. The real concern is gameplay here.

I do think ressource nodes are more common now, but it's probably hard to estimate. I guess one would have to generate a few duel maps in worldbuilder and count the number of iron to get a proof for the change (or is the amount displayed somewhere?).

 

[Tomice]

Sorry, double post.

 

[Seek]

I do think ressource nodes are more common now, but it's probably hard to estimate. I guess one would have to generate a few duel maps in worldbuilder and count the number of iron to get a proof for the change (or is the amount displayed somewhere?).

I almost always play with custom mapscripts so that might have something to do with it. I took a quick look in the lua file at one point to make sure Thal didn't accidentally lower the nodes, but if it's the one I think it is, it's dozens of pages long! I'll give it another go tomorrow.

Btw, thanks for all the work you did on the last couple pages!

 

[Tomice]

Btw, thanks for all the work you did on the last couple pages!

You mean me? Oh, that's just crazy procrastination activity keeping me from doing some more useful stuff I planned

Also, I'm a realism freak, and TPs in their current form are annoying me big time! I'm using poncratias TP-redesigned mod, which helps a lot, but sadly he abandoned it. I gotta remove those spaghetti paths one day myself!


EDIT:

@Thal: About your current suggestions:

The plans for production improvements are good. I don't know if the steam power buff for LMs is necessary, little lategame experience, but it all sounds reasonable.
___

The plans for riverside/coastal improvements are a bit harder to understand. Do you mean that farmed, mined or lumbermilled tiles next to rivers shouldn't yield gold at all? That riverside gold would only come from TPs?

On the other hand, ALL land improvements next to coast would get +1?

Well, this would mean rivers would be nothing special at all until CS/Optics, when the TP and farm buffs for river adjacency would be unlocked. But coasts should have the +1 from start?

This would mean ("+" means like, "-" means dislike):

+ Coasts would be very valuable
- Rivers would be only valuable after significant time
? unlocking the coast-based bonus with a naval tech, too, could be a good alternative (*)
? CS might be too expensive after the nerf


+ 2 improvements boosted instead of one near rivers -> more choice
- log transport in rivers was historically very common, lumbermills often riverside (http://en.wikipedia.org/wiki/Log_driving) (**)

+ overall gold income would be similar for most of the game (less riverside, more coastal gold)
+ the reinstated water tile buff from harbor is good IMO (otherwise those tiles are truly worthless)


(*) further strenghtening naval tech route and naval gameplay while also reducing the river/coast difference

(**) buffing all improvements except mines next to rivers makes no sense, though, at least not without mines being buffed somewhere else, like next to hills. But you could give gold to ALL riverside improvements with optics, to make riverside farms the only exception?


Summary:

I'd put both riverside and coastside gold buffs on compass (possibly renaming it and ev. making it a bit more expensive). Making it an essential tech for everyone would make naval gameplay common in the early game, like it should be (e.g. Phoenicians, Greeks). Islands would be a realistic option to settle on before midgame and empires centered around a body of water (Greek colonization, Roman empire) more likely - see picture below!

Giving all improvements +1 gold next to rivers (except farms) with a tech that comes around the same time as CS would be another option to consider.

CS should be cheaper (beaker cost like all other medieval techs) since it's nerfed, and to keep the "TPs are better than farms on rivers" period as short as possible.

I'm still sthinking about putting a non-military thing on mathematics, to buff archery, the whole path tends to be delayed too long IMO if you don't need siege for offensive actions. Only one gold from TPs before mathematics would be too harsh I guess? Coastal bonus on Optics, river bonus on Mathematics? Currency would be more logic, but might come to late. ALso, the whole path to machinery might be too strong then, compared the CS path.

 

[Thalassicus]

So a coastal land bonus is possible for buildings?

That's not possible either... I was just listing the specific suggestions people made I'd checked to see if it was an option.

I took a closer look at the resource placement code. It turns out it's set up rather strangely: food resources and small strategic nodes read the abundant/normal/sparse setting, but not large-quantity strategic nodes. Also, I realized larger numbers mean fewer nodes (the variable indicates tiles per node). So I was accidentally decreasing food and small-node strategics by -33%. I've now revised it to the proper behavior, increasing quantity of all strategic nodes +50%.

And yes, the AssignStartingPlots.lua file is over 10,000 lines of code. I mark code segments I work on with the "combat mod" comment.

Thanks for pointing out iron was set to 2. The reason this error happened was the the other resources around it are all the same value... Iron's a bit higher (likely due to siege using it).

Something I'd like to figure out is how to scale the graphics as well... all horse nodes for example visually display 2 horses walking around, which is the "small" graphic. I'd like it to show two for 1-per nodes and four for 2-per nodes, if anyone has a tip on where to change that.

The plans for riverside/coastal improvements are a bit harder to understand. Do you mean that farmed, mined or lumbermilled tiles next to rivers shouldn't yield gold at all? That riverside gold would only come from TPs?

Improvement yields are in the table, not terrain yields. The bulleted list is the new totality of the mod, rivers are unchanged from vanilla:

To sum it all up and simplify, the new mod details would be like:

• +1 on Mines with Machinery.
• +1 on Lumbermills moved from Steam Power to Machinery.
• +1 on coastal improvements.
• +1 on riverside Trading Posts with Optics.
• Farming a desert now requires fresh water.
• Forts deal 1 damage to adjacent enemies and cost 2/turn maintenance.
• Net yield of mining gold, silver, and gems is the same as vanilla, post-Machinery (basically, the 1 for resources is moved from the mine itself to Machinery)
• (GP improvement changes)

Can't put anything more on Currency due to UI limitations (if we did it'd be invisible to the player). It's the only tech in vanilla that uses all 5 slots on the UI.

If someone knows an easy way to make these UI elements have 6 slots it'd be a tremendous help...

 

[Tomice]

Ok, I understand - finally
I guess then my last post is useless due to misunderstanding yours...

Yes, I like your changes and will test them when you have a dev build ready. Thanks for your much appreciated effort!

 

[Seek]

On another topic, it's always irked me that oil improvements are so terrible. What do people think about a gold buff?

 

[Bibor]

After playing a lot of post-pach vanilla games I'm hesitant to install this mod again. Not because it's bad but because it might be making terrain a bit too strong.

Lets take, for example, a simple non-riverside hill.
2 without improvement
3 with a mine
4.5 with a factory
6 if you add railroad

Making it +1 with Engineering it becomes 8 @ industrial. And that's excluding bonuses like Hiawatha, Augustus or even simple Armory or Workshop.

My point being: there are significant modifiers to base production in this game. Upping base production can have a nasty sideffect of tiles becoming too strong.

 

[Txurce]

On another topic, it's always irked me that oil improvements are so terrible. What do people think about a gold buff?

Given how few uses there are for oil, that could be a decent way to balance the value of the resource.