Realms Beyond Balance Mod for Civ6

Following the conclusion of the Play By Email #7 game at Realms Beyond, I put together a series of suggestions on how to make balance changes to the gameplay. I was frustrated with some of the gameplay elements that had come into stark relief as we took a competitive Multiplayer game into the later stages of the Civ6 tech and civics tree for the first time. What we found there was that many of the gameplay systems that worked well in the earlier stages of the gameplay started to fall apart in the lategame, and I proposed a series of tweaks to ameliorate some of these concerns. Listed below are the changes that I thought would most benefit the gameplay in a theoretical Civ6 mod. (Note that this was written for the non-expansion base game of Civ6 and not anything that came later.)

Main Goals

* Tone Down Scaling: the forest chops / resource harvests and the district costs both scale off of the same formula. It works well in the earlier parts of the game, but the scaling becomes too great as the game continues. RB Mod uses the identical formula at first, then slowly lowers the scaling until it ends up at half the value of the base game at the end of the tech/civics tree.

* Improve City Building: there are too many constraints on city building in the base game of Civ6. The housing cap is set too low (making food fairly unimportant), production is nearly impossible to obtain outside of hill tiles, and city infrastucture is mostly too expensive to be worth constructing. RB Mod makes small tweaks to allow cities to grow faster, have more production available, and make infrastructure more viable.

* Decrease Lategame Costs: we all know by now that the best way to train military in Civ6 is to build cheap early units and then upgrade them into more advanced units. The lategame units are also prohibitively expensive to build. RB Mod keeps the early unit costs the same and decreases the cost of later units to make them more viable. It also increases the cost of upgrading units to make this less of a One Right Choice answer. In order to compensate for more units being trainable and clogging up the map, corps and armies unlock sooner on the civics tree.

* General Quality of Life Fixes: small things that improve the gameplay.

This is not intended to be a total rebalancing mod. France and Spain are not getting buffed to make them equal in power to other civs, the world wonders are not changing their functions, or anything else that comprehensive. The goal is to keep the basic structure of Civ6 intact while making small changes around the edges that will fix some of the more problematic aspects of the gameplay. This is also only a first stab at a potential mod, and I expect that we'll have many changes along the way based on community feedback. For that matter, I don't even know how to do the modding myself. But it's a starting point, and we'll see where it goes from here.

Scaling Formulas

This is the first issue to tackle because it underlies so much else of the gameplay in Civ6. The scaling formula in the base game is based around the number of techs/civics discovered, whichever one is greater as a percentage out of the total number of techs/civics in the game. Since there are 67 techs and 50 civics, in practice most of the time the scaling formula is based on the number of civics discovered. Here's the district cost formula:

=FLOOR(BaseCost * (1 + 9 * FLOOR(100 * MAX(CompletedTechs / 67, CompletedCivics / 50))/100))

This looks like a total mess, but most of the text is devoted to taking the higher number of techs discovered divided by 67 or civics discovered divided by 50. If you have 20 civics discoverd, the formula resolves to:

=FLOOR(BaseCost * (1 + 9 * (0.40)))

That's a lot simpler. The BaseCost for districts is almost always 54 on Standard speed, with unique districts costing half of that (27) and Aqueducts having a BaseCost of 32. This formula works well for the earlier parts of Civ6, but it starts to make the districts extremely expensive in the later stages of the game. They can still be constructed in new cities... but only because forest chopping and resource harvesting use the same scaling formula, and the greater chop/harvest yields match the greater district cost. What we'd like to do is keep the costs roughly the same in the earlier portions of the game, and have the scaling fall off over time so that it's not as large in the later portions of the gameplay. I spent some time playing around with different formulas and eventually came up with this:

=FLOOR(BaseCost * (1 + 9 * FLOOR(100 * MAX(CompletedTechs / 67, CompletedCivics / 50))/100)) - (FLOOR(100 * MAX(CompletedTechs / 67, CompletedCivics / 50))/100) * (FLOOR(BaseCost * (1 + 9 * FLOOR(100 * MAX(CompletedTechs / 67, CompletedCivics / 50))/100)) - FLOOR(BaseCost/2 * (1 + 9 * FLOOR(100 * MAX(CompletedTechs / 67, CompletedCivics / 50))/100))

Now that's a true monstrosity. Let me explain what's going on. We start with the same district cost formula at first. Then we subtract out the same district cost formula minus one half the district cost formua, and scale it by the number of techs/civics discovered. Let me give you an example to help this make more sense, starting with a situation where a player had discovered 10 of the 50 civics. Under the current district cost formula, this would be:

=FLOOR(54 * (1 + 9 * (0.20))) = 151 production

10 civics discovered puts you in the late Classical era of Civ6, and this matches more or less what we see in-game. Under the RB district cost formula, the results would be almost the same:

=FLOOR(54 * (1 + 9 * (0.20))) - (0.20) * (FLOOR(54 * (1 + 9 * (0.20))) - (FLOOR(27 * (1 + 9 * (0.20))) = 151 - 0.2 * (151-75) = 136 production

There's not much difference there, and that's the whole point: the idea is to keep the early game costs about the same. However, the longer the game goes, the more the cost discounts will start to appear. Here's a table with the results across the full tech tree:

The cheaper cost of the districts becomes more and more apparent as the game goes on. Eventually, if the game reached the end of the tech/civics tree, the cost in RB Mod would be half what it is in the current game. This keeps the early game balance of Civ6 intact while making it more viable to build districts as the game hits the middle and later stages. In particular, this should help make Industrial districts more attractive to build instead of them being uselessly expensive for anyone other than Germany.

The same scaling formula is used for forest chops and resource harvests. This is why a forest chop will always produce slightly more than a third of the production required to create a district, no matter where on the tech tree the player might happen to be. The formula is literally the same as before, except that the Base Cost changes from the 54 used for a district to 20 for forests/jungles/marshes and 25 for resource harvests. Forest chops create all production, jungle chops are half food and half production, and marsh clearing is all food. Resource harvests are 25% higher in value because they have a base value of 25 instead of 20, and are either production-based (deer, stone) or food-based (bananas, cattle) or gold-based (crabs, copper), with the crab/copper harvests having double the yield in gold. Using the exact same change to the scaling formula, we end up with the following table:

Once again, the chopping/harvesting values start out identical at the beginning of the game, and slowly scale down in value as time goes on. By the time of the Industrial era, where we were at the end of the PBEM7 game, there would be a very noticeable decrease in chopping yields, from roughly 130 production from a forest chop down to about 90 production. Since the district scaling is changing at the same time, the relative value of a chop/harvest to a district cost remains the same. However, chops/harvests will be less dominant compared to everything else in the game, and that's the goal behind toning down the scaling. Chopping has an important place in the game and we don't want to get rid of it. We simply want it to be less overwhelmingly important in the later stages of the game. By toning down the chopping/harvesting yields as the game goes on and strengthening natural city production, we should be able to achieve these goals.

One caveat: when I tested the observed in-game cost of a district against what the formula says it should cost, I didn't get values that matched up. A district with zero techs completed (discovering a natural wonder and triggering the boost for Astrology let me build a Holy Site with "zero" techs finished) cost 54 production as expected, but then with 1 tech finished it cost 58 production and with 2 techs completed it cost 63 production, in each case slightly less than what the formula says they should cost. I don't know if one of the new Rise and Fall patches tweaked the district cost formula and I couldn't find information about it online. This probably doesn't matter but I wanted to point it out. I also have no idea where in the coding these formulas are stored; hopefully some of you in the community are better at this sort of thing than I am.

This doesn't take into account the district discounting formula, whereby under certain circumstances it's possible to build districts at 40% cheaper than the nominal cost if they are less common than normal in your empire. My preference would be to scrap the district discounting entirely; it makes no sense to create such a huge reduction in district cost, but only under circumstances that are byzantine in complexity and never explained anywhere in-game. I would suggest lowering the cost of all districts to compensate (say, to a BaseCost of 46 instead of the current 54) and doing away with the discounting entirely. I've left this unchanged for the moment in the interests of trying to keep alterations minimal. Interested to hear what others think about this.

I know that the scaling formula also applies to the district project costs and Great Person point yields, and we may as well have them tweaked by the same scaling formula. Does anyone know if there are other mechanics that are affected by the same scaling formula? I think the cost of a trader and the cost to purchase tiles might be others (?)

City Building

One of the other large problems in Civ6 is that the city building doesn't work as well as in past games. It's not so bad in Single Player where you can afford to sink lots of time into building the perfect metropolises, but in Multiplayer, it's rare for any city to get much past size 10. Food rarely tends to matter much past the earliest stages of the game because cities run into the housing limit almost immediately. The designers clearly intended for players to build vastly more farms than they actually do, and in practice farms are useless for the most part because what's the point in sinking builder charges into food production so that your city can hit the housing limit again almost immediately, while sucking up more amenities to boot? We can do better here, making small tweaks to the game's formulas to improve the city building aspects of the game. The idea is to have stronger cities with more natural production, which will also tone down the relative impact of chopping/harvesting everything in the later stages of the game.

For practical purposes, the housing limit is set very low in Civ6. Housing limits growth in Civ6: each city has a housing rating, and when a city has population one less than the housing number, growth slows by -50%. When housing = population, growth slows by -75%. By default, a city on fresh water has 5 housing and will therefore start hitting the growth penalty at size 4. Coastal cities start with 3 housing, and non-coastal cities start with 2 housing and will have the growth penalty at size 1! What the aqueduct does is raise any non-fresh water city to 6 housing. For coastal cities, it therefore provides +3 housing and for non-coastal cities it provides +4 housing. Cities already on fresh water do receive the benefit from an aqueduct of another +2 housing, for a total of 7 housing, making them slightly better than non-fresh water cities even after the aqueduct is built, although to a far lesser degree. Let me summarize with a chart:

Dry City: default housing = 2, initial -50% growth penalty at size 1, full -75% growth penalty at size 2
Coastal City: default housing = 3, initial -50% growth penalty at size 2, full -75% growth penalty at size 3
Fresh Water: default housing = 5, initial -50% growth penalty at size 4, full -75% growth penalty at size 5
Aqueduct: raises non-freshwater cities to 6 default housing, raises freshwater cities to 7 default housing

The designers clearly intended players to build lots of farms to raise the housing cap, run the policies that provide extra housing, and eventually build a bunch of Neighborhood districts. In MP though, none of these things are cost-effective to do and we end up with a bunch of small cities that rarely make it much past size 7. This is not really much fun and it takes a lot of the life out of the district system. It would be much more interesting if cities could grow larger and have a choice to build more districts. There's a simple solution here: increase the housing cap a bit, making food more valuable as an in-game commodity and opening up room for more city production. Here are my suggested tweaks, starting with a global increase to the base housing values:

Dry City: default housing = 3, initial -50% growth penalty at size 2, full -75% growth penalty at size 3
Coastal City: default housing = 4, initial -50% growth penalty at size 3, full -75% growth penalty at size 4
Fresh Water: default housing = 6, initial -50% growth penalty at size 5, full -75% growth penalty at size 6
Aqueduct: raises non-freshwater cities to 7 default housing, raises freshwater cities to 8 default housing

Everything is up one point across the board. Then we make tweaks to the two city improvements that deal with housing:

Granary: 65 production cost, +1 food and +3 housing (up from +2 housing)
Sewer: 150 production cost (down from 200 production), +3 housing (up from +2 housing), unlocks at Scientific Theory (moved up one tech from Sanitation)

These changes to buildings will open up a lot more space for cities to grow upwards. Under the current system, a dry city hits the housing cap at size 1 and even with a granary, hits it again at size 3. Under the proposed changes in RB Mod, a dry city hits the housing cap at size 2, and with a granary doesn't hit it again until reaching size 5. That's enough to make these locations viable instead essentially wasting a settler. With an aqueduct and a granary, it's now enough to get a dry city to 10 points of housing, which would stop the penalty from kicking in until size 9. That's a major improvement from the current state of the game.

For later on in the game, RB Mod makes the sewer cheaper to build at 150 production cost, increases the value to 3 points of housing, and moves the building up a tech on the tree. Moving up a tech has a more significant influence than one might think; the sewer is currently located at Sanitation tech with the horribly unboostable requirement of building two Neighborhood districts. That's doable for Kongo but probably no one else. Sewers are also stuck on one of the most useless parts of the tech tree: behind Astronomy (which has, uh, the Potala Palace wonder and nothing else), Scientific Theory (Oxford University and +1 food on plantations), and then finally Sanitation tech itself. No one bothers to research these techs until forced to do so because they're all weak options. By moving sewers forward a tech, we make it much cheaper to unlock them and breath some life into this part of the tech tree. 150 production cost is quite reasonable for early Industrial era cities. I could see myself wanting this for my mature cities, especially at Scientific Theory tech. The improved sewers along with stronger granaries and higher default housing values should let core cities get closer to size 15 instead of stalling out around size 10. Food will actually start to matter once again, as cities without it will be stall out compared to those that do have food.

Note that the fundamental rules of the game remain in place: fresh water is still better than non-fresh water. It's still important to try and situate dry cities around potential aqueduct placements. RB Mod simply makes some tweaks to make things less punishing and open up space for cities to grow a bit more.

Food Requirements

To go along with the housing changes, RB Mod also tones down the food requirements for population growth. They scale up very quickly in this game, like in Civ5, and this game doesn't have Maritime city states and mass farming to boost the size of every city. Here's the default formula in Civ6:

Food to Growth = 15 + 8 * N + N^1.5

Where N = current population. Plug a value of 1 into the formula and it comes out with 24 food required to grow to size 2, just as observed in-game. The real issue in this formula is the exponential value where N gets raised to a power of 1.5. This is why the food requirements go up so quickly. I played around with some potential alterations to the formula and ended up changing it to this:

Food to Growth = 15 + 8 * N + N^0.5

Watch what this does to the food requirements over time:

At first there's no change and the initial city sizes will remain very similar. Over time though, the food requirements begin to drop off and make it noticeably easier for cities to grow in size. It takes about 600 total food to reach size 9 in the current game, and that same 600 food will take a city to size 10 in RB Mod. Go further down the list and the difference becomes even larger. 1200 food gets a city to size 13 in the current game, while it will take it almost to size 15 in RB Mod. With the housing limit being softened, this should allow for the creation of larger cities that can work more tiles and build more districts. It's a subtle effect and the early game should remain identical in Civ6, but the gameplay in the later stages should be a lot more interesting.

Amenities

Since cities will be growing larger, there's a need for more amenities. This can be done with a simple buff to the luxury resources:

Base value = 5 cities made happy per resource, up from 4 cities (Aztecs and the Zanzibar luxuries go to 7 cities, up from 6 cities)

This could need more tweaking later. For now, it's a starting place.

Cultural Tile Expansion

I can't be the only one who thinks that borders expand far too slowly in Civ6. I think that the designers intended for players to purchase lots of tiles, and while that is indeed necessary at times, it always feels wasteful. The tile picker often makes baffling decisions about what tiles to grab, and for my money's worth, the overall pace of cultural border expansion feels too slow. Way too often, I have cities where the population has outgrown the tiles available to work. Since we're making minor changes to the in-game scaling formulas, we can make a similar quality of life change to the cultural border expansion formula. Here's the current formula in Civ6:

Culture required = 11 + (6 * N)^1.3

Where N is the number of tiles already claimed by culture. Once again, it's the exponent that causes the scaling to go up so quickly. With a small tweak there, we can make it so that borders expand just a little bit faster, with the change becoming more noticeable as the game goes on. Here's my initial suggestion:

Culture required = 11 + (6 * N)^1.25

Which yields this table:

As before, the effects are minor at first and should keep the Civ6 early game pretty much identical, then grow over time as the gameplay continues. Mature cities will have more tiles within their borders available to work, and more places to construct districts without having to spend money on tile purchases.

Tile Improvements

The biggest problem for Civ6 in terms of tile improvements is that flatground tiles are almost useless. The intention on the part of the developers was to farm them, but we've already seen how building lots of farms just isn't very useful. Sinking production into builders to place farms so that cities can grow a size or two and then hit the housing cap against is a poor use of the game's limited production. What we really would want is a way to convert flatground tiles into something that can yield production, making them worth having and changing the gameplay so that it's less reliant on hills. Fortunately, the game already has this available:

Plant Woods: moved from Conservation civic to Construction tech

Civ6 does have a way to turn flatground tiles into production centers in the form of planting woods. Unfortunately, this comes so late at Conservation civic that it's almost pointless. We didn't even make it there in PBEM7 in a game that was entering the Modern era. Gameplay balance would be vastly improved by moving this earlier on the research trees, much earlier, timed so that it becomes available right around the same time as lumbermills come online at Machinery tech. I'll suggest Construction tech for the moment: somewhat logical from a thematic perspective, roughly where we would want this ability to unlock, and makes Construction a much more useful tech to chase after (currently there isn't much at Construction that makes people want to research it, just siege towers and the Terracotta Army).

The idea is that the ability to plant woods works in tandem with lumbermills to provide a new option for flatground tiles. A forest + lumbermill pair can add 2 production to any flatground tile, and 3 production for any flatground tile next to a river. Remember those sad cities from past PBEM games where there's a riverside location that looks like it should be beautiful, but there's no hills present and therefore the city is basically worthless? This change would allow those spots to become very useful locations, as one would expect that they should be. There's no fear about chopping abuse here either because "new growth" forests do not produce production when chopped down. (It would be nice if we could come up with some kind of graphical distinction between new growth and old growth forests. Something to think about for the future.) If the game ever reaches Steel tech, all of the lumbermills get +1 production, taking a non-river grassland lumbermill to 2/3 yield and a riverside grassland lumbermill to 2/4 yield, same as a grassland hill mine.

Of course, it requires builder labor to do this. Players will have to decide whether it's worthwhile to invest 2 builder charges into dropping down a forest + lumbermill pair. I suspect this will definitely be worth it for riverside tiles, more questionable for non-riverside tiles. Still, it makes things a lot more interesting, right? We can move away from the current dynamic where non-hill, non-forest locations are virtually useless. The gameplay will retain its current setup where hills are better than non-hills and riverside locations are better than non-riverside locations. However, parts of the map without hills or rivers or forests can actually become viable locations now with some builder labor, and not just at the absolute end of the civics tree.

There's a smaller change that I'll also suggest:

Camps: +1 food immediately, +1 production and +1 gold at Mercantilism civic

Right now, the camp improvement yields +1 gold for improving it and then +1 food and +1 production at Mercantilism. This is a pathetic result for improving the tile, and it makes camps almost pointless in the early game. Don't bother wasting a builder charge. By changing them so that the initial improvement is +1 food, this makes camps comparable to farms and their resources worth improving in the early game. I honestly don't know why Firaxis didn't make this change a long time ago.

Builders

RB Mod will also make some small changes to builders to let them keep pace with the other tweaks to city growth. Right now, builders scale up in cost by 4 production cost each:

Builder cost = 50 + 4 * number of previous builders

So 50 production, then 54 production, then 58 production, and so on. You get the idea. RB Mod will make this very simple adjustment:

Builder cost = 50 + 3 * number of previous builders

It's the same principle, the early game cost of builders is essentially the same and then they get slightly cheaper as compared to the base game as time progresses. RB Mod then makes two other quality of life changes for builders:

Ilkum policy: +50% production towards builders (up from +30%), obsoletes with Public Works policy / Civil Engineering civic instead of Serfdom policy / Feudalism civic

This will let players train builders more quickly with the larger production modifier. If +50% production is fine for settlers and military units, I don't see any reason why it can't be in place for builders as well. This also avoids the awkwardness of having +50% modifiers on virtually everything else that gets built and only +30% modifier on builders, which never made sense to me. The second change would delay the obsolescence of Ilkum as a policy so that it can be paired together with Serfdom if desired. If someone wants to tie up two precious policy slots by running both Ilkum and Serfdom at once, I see no reason why they shouldn't be allowed to do so. This will have the net effect of making individual builder charges cheaper to produce, allowing builders to keep up with the faster pace of city growth. I'm a bit concerned that this may be overtuning the builders a bit, but China can easily get 7 charge builders and it doesn't seem to break the game for that civ. We could always revisit this if the changes have gone too far.

Finally, there won't be any changes to the settler scaling. It seems to be in a good place right now, with the map filling up over time with additional cities at a solid pace. Colonization policy and the 30 production incremental costs for settlers are working OK in my book.

Decrease Lategame Costs

Another problem with Civ6 is that the lategame units and buildings are too expensive to be cost-effective. It's much better to train earlier units and upgrade them with gold into their lategame versions than it is to build the later versions of the units. In some cases, this gets to the point where unlocking new units is a poison pill, and players are better off *NOT* researching techs, with the privateers into submarine replacement at Electronics being the best-known example. RB Mod will scale down the costs of the lategame units and buildings to make them more viable while simultaneously increasing the cost of unit upgrades. The goal is to reward players for unlocking new techs and more advanced units/buildings, not penalize them.

Unit Upgrade Formula

This is the source of a lot of problems in Civ6. Building older units and upgrading them is too effective compared to building newer units. Ever since my first game of Civ6, I've thought that the upgrade costs should be more expensive to stop unit upgrading from being the One Right Choice solution for military. Here's the current formula:

Upgrade Cost = 10 + 1.5 * Production Difference

To use a basic example, warriors cost 40 production and swords cost 90 production, making the upgrade cost 10 + 1.5 * (90 - 40) = 85 gold, which gets rounded off to 90 gold. But in practice, it's normally half of this because Professional Army policy cuts upgrade costs in half, turning that warrior into sword example in a 45 gold cost. This is a ridiculously good rate of return, with gold exchanging for production at better than 1:1 rate, when at all other times gold converts into production at a far inferior rate. Rush buy costs are 4 times the production cost in gold, and it doesn't take a genius to see that upgrades are a vastly better deal. (You can upgrade a warrior into a sword for 45 gold with Professional Army policy or rush-buy a sword for 360 gold cost. I mean, come on.)

RB Mod will tone down the power of unit upgrades in two ways. First, by changing the upgrade cost formula:

Upgrade Cost = 0 + 2 * Production Difference

This removes the flat cost of the upgrade while increasing the cost scaling. For modest early game upgrades, like the slinger into archer upgrade (35 to 60 production = 25 difference), there's essentially no change at all. The warrior into sword example (50 production cost difference) goes up from 85 gold to 100 gold cost. Units with a huge difference in production lost, like changing quadriremes into frigates (160 production cost difference), go up a LOT in upgrade pricing, from 250 gold per upgrade to 320 gold per upgrade. Then we tackle the second aspect of the problem by nerfing Professional Army policy:

Professional Army: 33% discount on all unit upgrades (down from 50% discount). Here's a table showing the combined effects:

The quadrireme to frigate ends up almost twice as expensive in practice, going from 125 gold each to 213 gold each. That effectively rules out building 20 quadriremes and upgrading them all at once since there simply won't be enough gold available. The net effect is that upgrading units still remains a powerful strategy, but it's a lot less of a one-right-answer approach to playing the game. If players want a modern army or navy, they may have to - wait for it - train modern units after discovering the appropriate tech. I know, shocking right? This should help a lot in terms of toning down the effectiveness of mass upgrade strategies.

Unit Costs

Unit costs are rebalanced across the board in RB Mod, with the goal of keeping the early game costs unchanged while making the later era units more viable. I have no idea what the designers were thinking when they made the lategame units cost 500 and 600 production apiece. As stated above, one of the problems with Civ6 is that the unit production costs scale up to the point where it's almost impossible to train new units. This is one of the long-running problems with the One Unit Per Tile system, which needs to avoid the map getting overly crowded with units. RB Mod will attack this issue by keeping early game unit costs intact and gradually making late game units cheaper so that they can be trained in a more reasonable time frame. This combined with the improvements to city growth and city production will hopefully avoid players from having new units that take uselessly long to build. To prevent the map from being overcrowded, corps and armies are moved up earlier in the civics tree. As we saw in the PBEM7 game, the map became vastly less crowded as soon as corps were available as an option.

I spent a long time going through the full unit list in Civ6 and playing around with different unit costs. This is what I came up with:

It's the same idea that's been used throughout RB Mod, keeping the early game unit costs identical while gradually decreasing the costs of the lategame units. The Ancient and Classical units are essentially left intact, other than very small cost decreases to the Chariot Archer and Ngao Mbemba to help them out a little bit. The Medieval units are where the cost scaling starts to kick in, with crossbows dropping from 180 to 150 cost (knights have the same decrease), and muskets decreasing from 240 to 200 production cost. In both cases, that's 83% of the original cost and makes these units more viable for training with normal production. The cost discounting increases slightly for infantry, and then it increases a lot for the endgame units like mech infantry and machine guns. The one unit that has slightly less than normal cost discounting is the field cannon, which I've deliberately given less of a cost discount because it's such a dominant unit in its era.

I've also been thinking in terms of unit upgrade costs. The well-known archer to crossbow upgrade goes from 190/95 gold cost under the old system to 180/120 gold cost under the new system. The Roman warrior to legion upgrade goes from 115/55 gold cost to 140/93 gold cost. These upgrades are still viable, but players are probably going to have to train more contemporary units, not just fling gold at a bunch of old units. I see this as a positive change.

Pretty much the same deal here with the cavalry units and the siege units. They match the cost scaling downwards in the melee/ranged units and get larger as the game continues. Tanks should be a lot more viable to build at 380 production cost as opposed to 480 production cost.

The naval units require a bit more explanation. The cost scaling is greater for the naval units than for the land units, and that's due to another change: the naval production policies are dropping from +100% to +50% added production. This is the same thing that Firaxis did with the cavalry policies, which were originally +100% production and dropped down to +50% production. It's just too abusable for chopping/overflow purposes to have those +100% production cards in the game, sorry. Because of this policy change, the ships have been scaled down in cost more significantly than the land units. I scaled down the naval melee units a bit more than the naval ranged units because the naval ranged units are where the real power lies. Frigates in particular aren't any cheaper at all; they were effectively 140 production cost with the old version of Press Gangs, and now they will effectively cost 145 production with the new Press Gangs. That's deliberate because frigates were way too good in the base version of Civ6, and too much of the gameplay turned into who could build the largest fleet of frigates. With ironclads now moving to 250 production cost as compared to 220 production for frigates, the melee ships are much more viable. We may see more mixed fleets of ironclads and frigates as opposed to 80% frigates with a handful of upgraded caravels.

I've also been careful here to make sure that the unit upgrade costs follow the goal of making them more expensive. The quadrireme to frigate upgrade previously cost 250/125 gold, and we certainly saw plenty of that in past games. Now it moves to 260/173 gold cost, just short of 50% more expensive (and I would make it more expensive still if I could without wrecking the rest of the game balance). Players can probably manage to upgrade 10 quadriremes at that rate, maybe 15, but that's about it. Chopping out ships will also be a lot harder; in the Industrial era, an old chop with Press Gangs would produce about 125 * 2 = 250 production, roughly equal to a new ship. A new forest chop produces about 90 * 1.5 = 135 production, which is well short of a frigate's cost even with the newly cheaper ship costs. This should tone down the "throw down junk city to chop out ships" phenomenon significantly.

The air units have also been scaled down quite a bit in the hopes that this will make them more viable. I don't want to go too far here because air units are extremely powerful if they ever get built. At the same time, in the base game the cost of building an Aerodrome district and then a single Fighter at 500+ production is just absurd. Let's try this for now and see if it helps.

Unit-Related Policies

There are a number of small quality of life changes to be made to the policies related to units. The first changes are ones that I've already mentioned:

Maritime Industries: +50% production towards Ancient/Classical era naval units
Press Gangs: +50% production towards Renaissance/Industrial era naval units
International Waters: +50% production towards Modern, Atomic, and Information Age naval units

All of these are down from 100% production bonuses. The idea is to cut down on the chopping/harvesting + overflow tricks that these policies lend themselves towards. You can probably see where the next part of this is going:

Limes: +50% production towards defensive buildings

No more special overflow abuse for Limes policy, it gets the same as Agoge and the other unit production cards. (There is still the Monarchy + Limes extra production though, which is fine because Monarchy is a bit weaker than the other midgame governments.) The city walls have been made cheaper to compensate, which is detailed below in the buildings section. Here's another change referred to before:

Corps/fleets: unlock at Guilds civic, up from Nationalism civic
Armies/armadas: unlock at Nationalism civic, up from Mobilization civic

These changes shift the arrival of corps/armies forward by about one era in each case. Moving armies forward to Nationalism civic is an easy change; I debated about where to put corps before setting on Guilds. It's not a great fit thematically, but it's better than the other options in the Medieval part of the civics tree. Mercenaries would be the logical fit, but that civic is a little too easy to beeline, and I think that would make corps come online a bit too soon. Moving corps to Guilds civic transforms a rather lackluster current civic into a much more desirable target, and the need to go through both Civil Service and Feudalism to reach Guilds will keep anyone from beelining there too quickly. The late Medieval period is right around when I think corps should unlock, and then armies appear at the beginning of the Industrial period. Together, they should keep the map from getting too cluttered with units despite cheaper unit costs and more general production available to cities. (I am not worried about more advanced players using corps/armies to defeat less advanced players. If you fall behind in tech/civics in Civ6, you should expect to be in serious danger.)

Several policies are moved up on the civics tree:

Levee en Masse: unit maintenance reduced by 2 gold/turn, unlocks at Nationalism civic (up from Mobilization civic)
International Waters: +50% production on Modern, Atomic, and Information era naval units, unlocks at Mobilization civic (up from Cold War civic)

By making Levee en Masse unlock sooner, it will help players with large militaries support their costs a bit easier. We're expecting there to be more total units on the map with these changes in RB Mod, and this is the main way to help control the costs. As for International Waters, the policy inexplicably unlocks far too late in the base game, long after submarines and battleships are available. Moving it from Cold War civic up to Mobilization civic is an obvious fix.

Then there are a bunch of changes to the military card policies so that they don't obsolete obnoxiously early:

Agoge: +50% production on Ancient/Classical ranged/melee units, obsoletes at Grand Arme policy / Nationalism civic (instead of Feudal Contract policy / Feudalism civic)
Bastions: +6 melee / +5 ranged defensive city strength: never obsoletes (instead of obsoleting at Public Works policy / Civil Engineering civic)
Chivalry: +50% on Medieval/Renaissance/Industrial cavalry units, obsoletes at Military First policy / Rapid Deployment civic (instead of Lightning Warfare policy / Totalitarianism civic)
Discipline: +5 strength against barbarians, never obsoletes (instead of obsoleting at Native Conquest / Colonialism)
Feudal Contract, +50% production on Medieval/Renaissance ranged/melee units, obsoletes at Military First policy / Rapid Deployment civic (instead of Grand Armee policy / Nationalism civic)
Grand Armee, +50% production on Industrial/Modern ranged/melee units, never obsoletes (instead of obsoleting at Military First policy / Rapid Deployment civic)
Limes: +50% production on defensive buildings, never obsoletes (instead of obsoleting at Public Works policy / Civil Engineering civic)
Maneuver: +50% production on Ancient/Classical cavalry units, obsoletes at Lightning Warfare policy / Totalitarianism civic (instead of Chivalry policy / Divine Right civic)
Maritime Industries: +50% production on Ancient/Classical naval units, obsoletes at International Waters policy / Mobilization civic
Press Gangs: +50% production on Renaissance/Industrial naval units, obsoletes at Military First policy / Rapid Deployment civic (instead of International Waters policy)
Retainers: +1 amenity for cities with a garrisoned unit, never obsoletes (instead of obsoleting at Propaganda policy / Mass Media civic)

A couple of comments here. The main idea is to stop penalizing players from discovering more advanced civics, with the most famous examples being the way that Feudalism currently obsoletes Agoge policy and Exploration civic obsoletes Maritime Industries. Players should not be deliberately skipping civics that they want to have in order to prevent the obsolescence of useful policies. All of the unit production card policies are shifted later down the civics tree before they become obsolete. Bastions and Discipline and Retainers, all useful policies, never go obsolete at all. (I don't know why Firaxis had them disappear; the designers were overly zealous in trying to limit the number of policies available for players to pick between.) Limes also never obsoletes now. It currently obsoletes in the game at Civil Engineering civic, which made sense back when Civil Engineering granted 200 HP city defenses. However, that was shifted to Steel tech and Limes was left unaltered, still obsoleting at a civic that no longer unlocked the innate city defenses. We can fix this in RB Mod.

Miscellaneous Unit Changes

Machine Guns: changed from 1 range to 2 range

This is an obvious fix - why are machine guns worse than field guns?! We don't care about real-world accuracy here.)

General Quality of Life Fixes

This is the last category for other changes that don't go into the larger headings. I'll start with passing through the building list, where many of the lategame buildings have been scaled down in cost similarly to the units, and then finish up with whatever random stuff didn't fit anywhere else.

Building Costs

Like the lategame units, the lategame buildings take forever to construct in Civ6. We can scale down those costs a bit so that cities aren't required to spend 500 production on a single structure. Here's another list:

I'll start with the city center buildings. The three types of city walls have all been scaled down due to the nerf to Limes policy. Ancient walls go from 80 production to 60 production, which keeps them at the identical cost after applying the Limes nerf and makes them a bit cheaper for anyone not running the policy. Medieval and Renaissance walls go down a bit further to make them more viable build options. Someone in Monarchy government can knock out medieval walls quite cheaply, and that may be a smart choice for cities in contested border regions. Then the granary and sewer changes mentioned in the city growth section also appear here for sake of completion.

Most of the other buildings are scaled down in cost at about the same rate as the unit costs. Research labs, banks, and stock exchanges are scaled up in benefit slightly to make them more appropriate given the costs involved. For the Encampment buildings, I deliberately kept the scaling more minimal, since those buildings have some powerful military aspects. No need to go overbuffing them and making Encampment districts too strong. I did lower the stable to the same cost as the barracks though; it made no sense to have the stables cost 33% more. I think that may have been an artifact of the release version where cavalry units had +100% on their production cards. The Entertainment district buildings have been buffed for the tier 1 arena, and then the cost scaling dialed up much higher than normal for the zoo and stadium. At 300/420 production cost, it's much more viable now to have a single Entertainment district providing extra amenities for the core of your empire.

The Harbor and Holy Site buildings are basically fine, I left the Holy Site buildings mostly alone because faith is already a powerful currency in Civ6. The Industrial buildings were a real mess though, and I've made them significantly cheaper. The workshop costing 195 production to produce 2 production/turn is just a horrible deal in the base game. Nearly 100 turns payback on the upfront investment for that building is awful. I've dialed that way down to 120 production, and hopefully that along with cheaper factories will be enough to make at least a couple of Industrial districts a viable choice. It should be worthwhile having a few of them without ending up in the release version situation where endless factories/plants all buffing one another made things a bit too crazy.

Theatre districts are also basically OK, just getting the standard cost scaling discounts for RB Mod. I dialed up the discounting a bit higher for the Aerodrome buildings in the hope of making them more useful. Has anyone ever built a single one of them in our PBEM games (?) It's hard to know where they should be situated without any evidence from past games. Again, airplanes are really powerful in Civ6, they're just ungodly expensive to build in the base version of the game.

Economic Policies

There are several policy changes made to improve the economic side of the game:

Expropriation: +50% production towards settlers, tile purchase cost reduced by 33% (up from 20%)
Ilkum: +50% production towards builders (up from 30%)
Land Surveyors: tile purchase cost reduced by 33% (up from 20%)
Public Works: +50% production towards builders (up from 30%) and +2 builder charges

The first is the builder change mentioned in the city growth section, taking Ilkum and Public Works policies up from 30% production boost to 50% production boost. The second change for Land Surveyors and Expropriation discounts tile purchases by 33% instead of 20%. I've always felt that Land Surveyors was a good idea for a policy, and that the benefit was simply set too small to make it worth running very often. By increasing the discount to 33%, it should make this policy a much more viable choice for when cities need to purchase additional tiles.

Just as we did for the military policies, there are also several economic policies that go obsolete way too early. As part of the quality of life changes, these policies will obsolete later on the civics tree (or not at all):

Corvee: +15% production on Ancient/Classical wonders, never obsoletes (instead of obsoleting at Gothic Architecture policy / Divine Right civic)
Gothic Architecture: +15% production on Medieval/Renaissance wonders, never obsoletes (instead of obsoleting at Skyscrapers policy / Civil Engineering civic)
Ilkum: +50% production towards builders, obsoletes at Public Works policy / Civil Engineering civic (instead of Serfdom policy / Feudalism civic)
Urban Planning: +1 production in all cities, never obsoletes (instead of obsoleting at Colonial Office policy / Exploration civic

For the wonder building policies, it's better to stop these from ever expiring because players often build wonders from earlier in the game at much later dates. Since those wonders are still sitting there later on, there's no reason to obsolete the policies in case someone wants to make a run after them. As mentioned before, Ilkum now doesn't obsolete until the arrival of Public Works policy, letting players run both Ilkum and Serfdom together if desired. If they want to tie up two whole policy slots with that combo, that's their choice. Urban Planning also sticks around now for the whole game instead of inexplicably disappearing at Exploration civic. I considered doing something with the policies that add housing to cities, but I figured that it wouldn't be necessary with the changes being made to base housing and the granary/sewer improvements.

Religious Beliefs

Defender of the Faith: Combat units gain +5 strength when in the borders of a friendly city following this religion (down from +10 strength)

The religious system is pretty much fine in Civ6 as it is. Defender of the Faith is the one thing that comes across as a little too strong, so RB Mod will knock it down to +5 strength. That's still a huge benefit but without making the user almost invincible in defensive combat.

Roads

Ancient roads gain Classic era benefits (bridges over rivers)
"Classical" era roads moved to Medieval era, movement cost = 0.75
"Industrial" era roads moved to Renaissance era, movement cost = 0.66
"Modern" era roads moved to Industrial era, movement cost remains at 0.50

This is a little bit tricky because there are only graphics for four types of roads in Civ6 right now, and the names correspond to the era in which they are unlocked. RB Mod will provide the bridges over rivers benefit starting with the default Ancient era roads. Then the "Classical era" graphic roads will unlock in the Medieval era, with movement going to 0.75 cost on roads, followed by the "Industrial era" graphic roads unlocking in the Renaissance era and having 0.66 movement cost. This is the biggest change since it will allow the standard 2 move units to travel 3 tiles across roads. Then the "Modern era" graphic roads unlock as normal starting in the Industrial era. The net effect is to make unit movement a little bit less painful than it is in the unmodified base game. I think Firaxis created the current road movement under the impression that players would be going through 400 turn games that reached the end of the tech tree every time, not the 150 turn games that we run. Since there are different road types in the game's coding, we may as well take advantage of them.

Wonders

Venetian Arsenal: Receive a second naval unit each time you train a naval unit in the city where the wonder is built.

This really does seem like the best way to fix this wonder. The other wonders are OK; the goal of RB Mod is not to go through and rebalance all of the civs and leaders and wonders. At least not in this incarnation.

War Weariness

War weariness is obnoxious and annoying in Civ6. Here are the details on the current mechanics:

Warmonger points = (base 16 + era points(8/16/24/32)
* 1 for combat in allied lands
* 2 for combat in foreign lands
* 3 for killed unit
War weariness is applied to both parties in combat, attacker and defender, so it always goes up.

-50 for every turn at war
-2000 for declared peace
-200 for every turn at peace

The biggest problem with the formula is the scaling cost due to "era points". Not only will players have more units on the map in later eras, the era points also go up and cause the combats to produce more war weariness. My suggestion for RB Mod is to remove the scaling component and simply have a base 16 points with the modifiers for combat in allies/foreign lands and killed units. That would simplify things as follows:

Warmonger points = 16 * 1 (if combat in allied lands) * 2 (if combat in foreign lands) + 16 * 3 (if killed unit)

This would cut war weariness roughly in half initially, and by more than that in later eras. Open to other suggestions on this issue from the community.

Conclusion

OK, that's all for now. It might look like a long change list, and I suppose that there is indeed a lot here. However, nearly everything in these posts addresses cost scaling in some way. There are very few changes that fundamentally alter the gameplay, with the tweaks to builder costs, moving up corps/armies, and pushing the "plant woods" tile improvement much earlier on the tech tree being the biggest ones. I'm not interested in going through and rebalancing all of the different elements of the gameplay. The goal is to make small changes the fix the costs in the base game, while leaving the gameplay that we enjoy intact. This is only a starting point, and I'm very interested to see where the community takes things from here. That said, these changes are intended to be applied together as a whole, and picking them apart into bits and pieces likely will not yield the desired results. If we're serious about create an RB Mod as a community, we'll also need to do a lot of testing to find things that can't be grasped from spreadsheet numbers alone. As I said in the first post above, I also have no idea how to make these actual changes. So... yeah. I'm glad that we have a lot of people with coding and modding experience in this community to handle that side of things.

Unfortunately, as it turned out no one with modding experience proved to be particularly interested in taking up this challenge. There was a brief initial discussion of creating an official RB mod and then the project shortly fell by the wayside and was ignored. Since I couldn't do the programming myself, it effectively killed this off as a project. Rather than see all of the effort that I put into theorycrafting disappear into the void, I decided to archive my ideas here on the website in case they might prove useful one day down the road. Think of this as a time capsule of sorts for what kind of changes might have been helpful in fixing some of the gameplay problems that crept up in Civ6 without venturing into a wholesale reworking. It's too bad that nothing came out of this, but sometimes that happens in life. Not every project winds up being seen through to conclusion. Regardless, I hope that this was an interesting read for anyone who's spent time digging into the Civ6 mechanics. Thanks.