Improvements

[Seek]

The real question is how should the yields be valued? This is what I've currently got... 4 = 6 = 3 etc:

"Worth" = 10 * SUMPRODUCT( inverse of weight row; quantity row )

I had just figured out what was going on there when you posted the edit.

I think the values aren't quite right there. One thing to keep in mind is giving multiple types of yields should weigh higher - the reef seems much stronger than old faithful, for example. Otherwise, I think happiness and culture are weighted a little too strongly, and production a little too weakly. (Note I am basing this on gut feeling, not numbers.)
I feel it's more like:
4=3=6= 5 (or maybe 4) =4 (or maybe 3.5)=4

 

[Ahriman]

The purpose of the smokehouse is to make bonus resources better than the surrounding terrain in the early game.

But they're already better than surrounding terrain. Why should they be *massively* better?
But I admit, my concern with the slaughterhouse is much less than my concern for the fresh-water farm boost with fertilizer.
River tiles are already massively superior to other tiles; they get the food boost earlier, and they get gold boost. City sites are already the best in the game, *particularly* in the midgame.
Why would you want to boost them further?

I'm fine with making interesting choices, but I don't think you're achieving this; you're reinforcing the superiority of river sites even more (so that they are utter-no-brainers).
I agree that bonus resources should be attracting you to sites, I just think that the appropriate way to do that is through boosts at techs, rather than just boosts at the very beginning of the game. I think one of the things missing from the Civ5 tree are passive boosts from techs, and I'd like to see cows and the like boosted through tech rather than at the very beginning of the game.

Instead of making them all "meh", it might be fun to both nerf the overpowered ones and buff up some of the others too:

I think these bonuses are far too high, game-breakingly so in some places.
What happens if you happen to start near Krakatoa or Cerro te Potosi? Your economy massively outstrips all the other players, and you race through in tech or buy up all the city state alliances. These things are doubling the size of your economy for the first 50 turns or so.
Those kinds of bonuses that early in the game snowball out of control. Think of how huge it is as an advantage when the AI starts with an extra worker.

The natural wonders are hit-or-miss randomness; its not fun when the game gets determined by whether you get manage to get a good natural wonder or not.
You're also making them no-brainers. If the best strategy is always just take all the natural wonders you can get, then you're on the far-left of your curve.

Its also horribly ahistoric to
Civ power already depends a lot on the terrain near where they happened to start; buffing natural wonders up exacerbates this problem to a really unfun level.

The near-universal response to the superior natural-wonders from the patch was that they were far too high, and needed massive nerfing - including +10 gold. People were ok with the changes to old wonders; I think you should balance around those.

I wanted to add that civ5 has the problem of only three types of regions in gameplay terms:

Actually 4; plains are better than grassland, because hammers are better than food.

 

[Thalassicus]

Did you read the developers' notes I referenced?

The resources are added to make one zone of terrain equivalent to other zones of terrain. This is what I was trying to explain with the overpowered vs equal curve... in vanilla, bonus resources make all terrain equal. The Smokehouse makes bonus resources slightly better, and also boots an otherwise useless building.

Regarding the natural wonders, have you actually tried it? Cerro te Potosi might sound good in theory just looking at the numbers but is not overpowered. In vanilla, luxury resources earn 8g per turn. Therefore, you exchange 2g/turn + instant access for zero city growth. This is a significant tradeoff for one of your first cities to be stuck at size 1. From my experience I've found the 5 from others is arguably better, it's a benefit not available elsewhere.

 

[Tomice]

In case this got lost because I added it as edit and you guys have written smth in the meantime:

I'm really fascinated by the idea of removing the +1 gold from riverside tiles and instead make trading posts earn 1 additional gold next to rivers/coast (possibly after researching a naval tech).

This came in response to Thal's thought that trading posts could possible use a boost from a tech.

The main reason I'm so obsessed with it is that Civ5 fails horribly to simulate human settling patterns, TPs are rather weird in this regard, it's hard to imagine what they should represent in real life? Poncratias redesigned them as villages, but they aren't placed like villages would be.

Please check out my earlier post, too, and sorry if I'm annoying

 

[Seek]

In case this got lost because I added it as edit and you guys have written smth in the meantime:

I'm really fascinated by the idea of removing the +1 gold from riverside tiles and instead make trading posts earn 1 additional gold next to rivers/coast (possibly after researching a naval tech).

This came in response to Thal's thought that trading posts could possible use a boost from a tech.

The main reason I'm so obsessed with it is that Civ5 fails horribly to simulate human settling patterns, TPs are rather weird in this regard, it's hard to imagine what they should represent in real life? Poncratias redesigned them as villages, but they aren't placed like villages would be.

Please check out my earlier post, too, and sorry if I'm annoying

I like this idea, it would make gold more valuable. Maybe a passive tech buff later would be necessary, however. It would mean a pretty big gameshift, so it needs some serious thought before implementation of course.

 

[Ahriman]

in vanilla, bonus resources make all terrain equal.

No they don't. They are more present in zones of the map that have weaker yields from other specials. That is not the same as saying they have the same yield as any other terrain.

The bonus resource tile is still greater than that of a generic basic tile in the early game, and then the same after the tile boosts from techs (like civil service), because the bonus resources don't get tile boosts from techs.

But an individual worker can still only work 1 tile at a time, and they're generating higher yields working a bonus tile than others.
But the sheep/cow etc. isn't that big of a deal.

The far larger point is about the value of rivers, and the imbalance caused by the second food boost at fertilizer.

In vanilla, luxury resources earn 8g per turn

No they don't. You can get a lot of gold *if* there is a neighbor who has good relations with you, and has a lot of gold, and they don't have the luxury already, and its not just profit for you because it boosts your neighbor as well, giving them happiness.
And there is widespread agreement that the AI players pay too much for luxuries, particularly luxuries they don't need, and that they shouldn't be willing to pay so much gold unless they're near zero happiness or negative.

Therefore, you exchange 2g/turn + instant access for zero city growth

Pop 2 city with farmed grassland and Cerro te Posi = 5 food income - 4 food consumption = net food growth.
That's like saying that if you ever work a mine tile, you get zero city growth. Its just not true.

Add in the fact that you can use your ridiculous gold income to buy a MCS alliance, and it matters even less.

From my experience I've found the 5 from others is arguably better,

I agree that 5 culture from a tile is also overpowered.

I'm really fascinated by the idea of removing the +1 gold from riverside tiles and instead make trading posts earn 1 additional gold next to rivers/coast (possibly after researching a naval tech).

Its an interesting idea. But then, I worry that coastal start positions which tend to be loaded with fish would be overpowered relative to other starts.
I like the idea though about having real tension as to which improvement you build in your precious riverside tiles.

The main reason I'm so obsessed with it is that Civ5 fails horribly to simulate human settling patterns

It does fine in encouraging rivers, but I agree that it does a poor job in encouraging coastal settlement. Civ4 at least had trade route boosters only available in coastal cities, but that's gone now.
The big problem with coastal tiles is that you can't build improvements on them, which is why Civ has never really done a good job of making coasts desirable; it never wants tiles without improvements to be as good as tiles with improvements that you have to invest worker time in.

Buffing hilly non-river tiles with the smokehouse would also go rather against historic human settling patterns

 

[Tomice]

I like this idea, it would make gold more valuable. Maybe a passive tech buff later would be necessary, however. It would mean a pretty big gameshift, so it needs some serious thought before implementation of course.

@Thal - No luxury resources yield 8G in vanilla; did you mean they are 8 in total combined yield?

I think the overall gold income would stay the same. A bit less from riverside tiles, a bit more from coastal tiles, maybe lakes, too.

This is not about changing the gold amount available empire-wide, it is about slighly nerfing rivers, making riverside farms less of a no-brainer (by making TPs a equal choice riverside), and making coastal regions more valuable (as they are in real life).

____________


Maybe Thal thought about the income from selling luxuries repeatedly?

____________


@ Ahriman: with "human settling patterns" I didn't mean only cities. Civ5 has trading posts that don't exist in real life (at least not as spammable thing covering the landscape), but it has no representation of villages, suburbs, minor cities... I would really love to see TPs replaced with something that represents non-city human settlements. Poncratias' mod does this graphically to a certain extent, but not for placement. Humans settle near bodies of water most of the time, and next to mayor cities.

IIRC TPs were once designed to be position-dependent in their yields, but this was dropped in the development process.

Well, I don't want to reduce this to a realism and immersion discussion.
Trading posts are boring gameplay-wise, too, in that they always provide the same yield, no matter where, no matter when, no matter of techs.

 

[Thalassicus]

I like the idea about trading posts and rivers, that would make for some interesting tradeoffs.

I was mistaken... resources sell for 450g in vanilla, which comes out to 10g/turn... so Cerro is slightly less income than a luxury resource (since working a luxury generates additional gold).

The only time you're unable to sell off luxuries at a decent price is if everyone's denounced you and/or gone to war with you... which has happened to me several times with too much early success.

Barring this, do you actually have trouble finding trading partners? It might depend on map settings... on immortal+pangaea the AIs always have tons of gold (due to the difficulty bonuses), and it's not a problem finding someone to trade with. Pangaea does make it easier to find trading partners, but since the AI deals with naval warfare so poorly I feel it makes the game much more challenging overall, which is why I use that setting.


Regarding bonus resources, rivers, and such, I'm a subconscious/intuitive thinker and sometimes have difficulty explaining my thoughts clearly...

Do you understand the main goal I'm trying to demonstrate with high vs low balance variance, if that makes any sense? The visual image of the graph is the easiest way I can think of it. The goal of the mod is not to just make everything is equivalent. This might be what you're thinking of. The goal of the mod is to make things more fun by "increasing the choices and interesting decision-making opportunities available to the player," which is why I put that line first in the mod description. Sometimes this means making things more equal (value of buildings), sometimes less (value of terrain). I aim for the middle of the graph where choices are most important.

This is one reason I disliked the handling of the policy/promotion issues in patch 1.135. Rather than fix the balance issues of instant heals and policy-cost-reduction, they removed the choice of saving vs spending these things entirely. It reduced player choice instead of improving it.


--- Update ---
I've got the river thing working; this might be interesting to at least try out. Basically it would make riverside farms the same as lakeside farms.

We have lots of flexibility here, unlike other areas of the game (leader traits have very few options for example).

• Mines we could go either way with... it'd make sense for mines to produce more income with access to a waterway (like in vanilla) and would limit the impact of this on gold income.
• I could also add things like more gold to the palace to reduce the effect on the early game.
• Could add a further +1g to trading posts adjacent to cities, to help offset the lost gold from rivers.
• I can do this for certain technologies like the CS farm bonus. I could put some bonus on a tech like Currency, Machinery, Economics, etc.
• There's unused attributes for city, coast, and mountain adjacency... it'd be interesting to remove the +1g/tile bonus of harbors and instead add a +1g bonus to coasts, like rivers.
• However there's limitations... out of the various tables I've identified all work but one (or I'm not using it properly):
  • Improvement_RiverSideYields
  • Improvement_FreshWaterYields - No effect
  • Improvement_CoastalLandYields
  • Improvement_AdjacentCityYields
  • Improvement_AdjacentMountainYieldChanges
  • Improvement_RouteYieldChanges
  • Improvement_TechYieldChanges
  • Improvement_TechNoFreshWaterYieldChanges
  • Improvement_TechFreshWaterYieldChanges
  So we can't have improvements innately boosted near lakes except with a tech. Placing bonuses on Agriculture might be able to work around that.

To be honest, I find improvements and such underwhelming in Civ V. There's not many of them, and not much decision-making is involved even with the enhancements from this mod. Throwing some more variables into the mix might spice things up a bit.

Also... @Seek regarding the reef. Remember it's a coastal wonder, therefore development options for it are more limited. In many games I play the reef appears in an inaccessible/bad spot for a city to be built. The desert wonders also suffer from this (like the mesa and crater).

 

[Tomice]

Thx for trying the TP change out, Thal!

It is very interesting that those adjacency modifiers are possible

Let's try to think this through systematically:


1) Is the current amount of available gold (empire-wide) ok?

I think there's a fine balance between making things like CS alliances to easy to get (too much gold) and reducing choices (only enough gold to pay upkeeps). Overall, the current overall amount of gold might be roughly ok.


2) Should TPs be buffed in general?

Much of Thal's modding was to make dumb trading post spam less viable, so buffing them could counteract this. Other improvements are buffed through lategame techs (fertilizer), but TPs are buffed through policies. Overall, again no need to buff them.

Conclusion 1: TPs should still have 2 yield on average.

3) What other problems are to be considered?

- Rivers/Lakes only boost farms, making them the ideal choice very (too?) often.
- Riverside tiles are extremely strong in vanilla.
- Coastal regions are not as valuable as they are in real life (or even inferior to landlocked regions).
- Trading posts are boring gap-fillers, because they are equally valuable everywhere, their placement doesn't matter.
- TPs are a weird abomination without RL counterpart. Seeing them as villages increases immersion (especially with a graphics mod)

4) Is there a desirable way to change TP yields?

Suggestion A - small change
TPs are still 2 base yield
Rivers don't give any boost from start.
Civil Service makes riverside farms more valuable (as before).
Currency makes riverside TPs better.
To compensate the lost gold from rivers, coastal adjacency could boost TPs, too.

Suggestion B - larger change
TPs get only one base yield
Rivers don't give any boost from start.
Civil Service makes riverside farms more valuable (+1 as before).
Currency makes riverside TPs better (+1 ).
To compensate the lost gold from rivers, coastal adjacency could boost TPs, too.
To compensate the 1 base yield, TPs adjacent to a city would give +1 .

5) Possible problems:

Suggestion A would probably mean a slight overall increase in available gold, Suggestion B would probably reduce the available gold somewhat. [EDIT: Thal showed that the yield boosts are cumulative. So in variant "B", TPs next to rivers AND cities would have 4 yield, probably resulting in vanilla yields - 2 average - overall)

The reduced yield from rivers in the very early game could mean it's tougher to get enough gold at this stage. [EDIT: Thal suggested boosting gold from the palace to compensate]

6) Advantages:

- Less river dominance
- Coastal regions buffed
- More interesting and realistic TP placement
- Probably understandable for the AI.

 

[Tomice]

@Thal:

Regarding other improvements, riverside/coastal lumbermills would also be realistic to get a benefit. In fact, from a realism point of view many improvements connected to ship transport should be more valuable before combustion engines are invented. But this would make landlocked, riverless regions really bad. Very realistic but maybe ot good for gameplay?

Overall, I'd love to try this out, but it might be best to make this an optional part of Balance-Combined for the beginning.

 

[Thalassicus]

Not if we add bonuses for city adjacency and such. In addition, remember food resources inherently spawn more in low-yield areas. Also, the AI would have no trouble dealing with this since it reads the actual yield values & possibilities. (The AI wasn't coded for the 'within in cultural borders' thing for natural wonder happiness though.)

The effects are also cumulative, as show by this Venn Diagram carved in the side of a planet.

Obviously we wouldn't want to have bonuses in every table like this, it's just demonstrating options. It works for any yield and improvement types. The possibilities created by the CiV programmers far exceed the simplistic final design in this case.

 

[Tomice]

So you put a yield boost on city and mountain adjacency as well as a gold extra for roads under TPs? Nice

 

[Thalassicus]

Yep, the only one we can't work with is an innate freshwater bonus... either it's not read by the game code, or it should be used differently than the others (and is undocumented).

This allows for some very complex possibilities. For example, TPs could have 1g base, and +1g from river, coast, and city adjacency. It'd double their maximum potential while halving it on "bland" terrain. This would likely result in an overall increase in gold supply, yet removing the gold from farms would in turn reduce gold...

I think it's something we can't effectively determine with numbers alone. It'll need playtesting! I'll put together a test dev version.

Also, just think of the complex choices route-boosted yields would create! Should I put a road here for a more direct route, or there for better trading post synergy... it'd make road placement more important than just how well you can solve a shortest-path algorithm between N cities. However incredibly fun that might be though, I doubt the AI could use it effectively so it's probably best to leave out.

 

[Tomice]

for example, TPs could have 1g base, and +1g from river, coast, and city adjacency. It'd double their maximum potential while halving it on "bland" terrain. This would likely result in an overall increase in gold supply, yet removing the gold from farms would in turn reduce gold...

Also, just think of the complex choices route-boosted yields would create! Should I put a road here for a more direct route, or there for better trading post synergy... it'd make road placement more important than just how well you can solve a shortest-path algorithm between X cities. However incredibly fun that might be though, I doubt the AI could use it effectively so it's probably best to leave out.

I think it's something we can't effectively determine with numbers alone. It'll need playtesting! I'll put together a test dev version.

I'd be very happy to be the first test subject! Especially since I had similar ideas a month ago (the italic part) :

http://forums.civfanatics.com/showpost.php?p=9945184&postcount=472

I'm actually quite excited I might be able to play this soon thanks to your skills!

 

[Thalassicus]

Yep, I first looked into all this stuff was after reading your post. I didn't explore it further at the time because I was busy with other things in life for a while.

So here's what I'm thinking...

• 2 trading posts (vanilla).
• 1 for riverside trading posts, mines and lumbermills (vanilla).
• +1 for cityside trading posts (was 0).
• +1 for coastal trading posts (was 0).
• 0 for undeveloped rivers and riverside farms (was 1).
• Delete +1 per water tile bonus from Harbors (back to vanilla) (in City Development mod).
• 6 Palace (was 2) to compensate for early game, as most start locations are on rivers.

I believe this should keep total gold income roughly the same. The coastal increase and harbor decrease will counterbalance one another somewhat... likewise with the cityside increase and riverside decrease. Overall it should 1) shift emphasis away from rivers 2) address concerns about buffed farms.

I've got to head off now, but here's a preliminary version:

 

[Tomice]

I like how this would make all tile improvements equally viable next to rivers. It would break the dominance of riverside farms.

On the other hand, you actually weaken coastal regions, because the harbor bonus aplied to all water tiles, while the TP buff only applies if you choose to build a TP on a coast-adjacent tile. I might be wrong, but it seems more tiles were buffed previously.

Also, where do you compensate the city-adjacence bonus?

What about leaving the harbor bonus, but make TPs provide a base yield of one?

Countersuggestion:

• 6 Palace (was 2) to compensate for early game, as most start locations are on rivers.
• 1 trading posts (reduced from 2).
• 1 for riverside trading posts, mines and lumbermills (vanilla).
• +1 for cityside trading posts (was 0).
• +1 for coastal trading posts (was 0).
• 0 for undeveloped rivers and riverside farms (was 1).
• Leave +1 per water tile bonus from Harbors (in City Development mod).

This way, we'd get:

1 gold from bland TPs somewhere in the back of beyond
2 gold (vanilla value) from TPs next to city, river OR coast
3-4 gold if multiple of the above apply

Maybe the base city tile gold yield should follow the same rules, to not make it better to place a TP on a river delta tile than a city. The water tile buff from harbors was justified before IMO and should be left untouched.

Only playtesting can tell if too many tiles will be neither of the above and overall gold income will be to weak. I don't think so, but the optimal gold city might look vastly different from before. It would be close to a river mouth, or on an island/peninsula, which is much more realistic IMO.

Not sure what to do with roads, though. May they be a possibility to allow decent non-water trade/gold cities? But if we add 1 gold for TPs on roads, they become free, which is no good idea at all (road spam spaghetti!).

 

[Txurce]

Otherwise, I think happiness and culture are weighted a little too strongly, and production a little too weakly. (Note I am basing this on gut feeling, not numbers.)

My gut feeling is that in most cases wonders should be happiness- and culture-centric, rather than production.

 

[Thalassicus]

Wonders or natural wonders?

It seems my friend's still busy so I can stick around a while longer...

The cityside bonus is counterbalanced by reduced farm/undeveloped riverside gold. The reason I want to keep TP base yield the same is to minimize how many changes away from vanilla are necessary. Slight reduction to river gold, slight increase to cityside gold.

The harbor bonus can't be reached until midgame or so, and basic water tiles still provide only 2f2g with it. They're not really all that great even with the buff I added. In contrast, the coastal TP bonus is available earlier and allows TP tiles to reach a higher potential yield in the ideal circumstance of TP across a river from a city on the coast (rare but possible!). Concentrated effects are generally more valuable.

Over the past month or so, I've started thinking merchant slots make more sense as a harbor buff instead of yield bonuses. Harbors still get the +1 merchant specialist slot of the CD mod (0 in vanilla) like before. I could increase it to 2 slots if they feel underpowered again. It's somewhat more realistic since a coastal city is generally more prosperous than an island city... once you start becoming totally surrounded by water your odds of being an important place in the world drop. There's rare exceptions like Hawaii because of their military value.

 

[Txurce]

No they don't. You can get a lot of gold *if* there is a neighbor who has good relations with you, and has a lot of gold, and they don't have the luxury already, and its not just profit for you because it boosts your neighbor as well, giving them happiness.
And there is widespread agreement that the AI players pay too much for luxuries, particularly luxuries they don't need, and that they shouldn't be willing to pay so much gold unless they're near zero happiness or negative.

I've had Cerro, and it did not transform my early game. In fact, it wasn't particularly noticeable, mainly because it paid off like a luxury.

From your first statement I would think lux trading is much more difficult than I know it to be. Also, that its relative value compared to Cerro is lessened by the AI receiving happiness from it. But then your second statement contradicts that, stressing that the AI is already OD'ing on happiness. Contradiction aside, I'm not sure what the AI paying too much for luxuries they don't need on higher levels has to do with the overall issue.

 

[Txurce]

Wonders or natural wonders?

It seems my friend's still busy so I can stick around a while longer...

The harbor bonus can't be reached until midgame or so, and basic water tiles still provide only 2f2g with it. They're not really all that great even with the buff. In contrast, the coastal TP bonus is available earlier and allows TP tiles to reach a higher potential yield in the ideal circumstance of TP across a river from a city on the coast (rare but possible!). Concentrated effects are generally more valuable.

Over the past month or so, I've started thinking merchant slots make more sense as a harbor buff instead of yield bonuses. Harbors still get the +1 merchant specialist slot of the CD mod (0 in vanilla) like before. I could increase it to 2 slots if they feel underpowered again. It's somewhat more realistic since a coastal city is generally more prosperous than an island city... once you start becoming totally surrounded by water your odds of being an important place in the world drop. There's rare exceptions like Hawaii because of their military value.

The cityside bonus is counterbalanced by reduced farm/undeveloped riverside gold.

I meant natural wonders, since we were talking about yields.

Merchants being the buff to harbors makes intuitive sense.

Is the TP change being considered primarily as an indirect nerf of river starts? Again from an intuitive place, my sense is the goal should be what you stated: a tilt toward farms in the early game, replaced by trading posts in the later game. Whether or not this is achieved - in my case, getting off my butt and making the changes - strikes me as the core barometer.

PS - I still think raising the maintenance on workshops is the best way to balance them, as well as feeling right historically.