Civ6 has been out for a little over a year at this point. With six patches in the rear view mirror, it looks as though we have the definitive non-expansion version of Civ6. Now that the initial hype has died down, some of the most egregious bugs and exploits of the release version have been patched out, and I've had the chance to play a full Multiplayer game in addition to almost a dozen Single Player games, I wanted to take a chance to assess the state of things. Overall, I've been pleasantly surprised at how well Civ6 has held up to the stress testing of the expert turn-based strategy gaming community. While there are still significant issues, especially in the performance of the AI, the mechanics of the game themselves have proven better than I expected, and have steadily improved with each patch released. If you didn't like the look of Civ6 at release then it's still not going to satisfy you, but if you were on the fence initially it might now have been polished enough to be worth your time.
This might seem like a strange topic to begin with, but expansion is a mechanic at the core of the empire building genre of strategy games. How different Civilization games have chosen to address expansion has had a profound effect on the way in which each of them has functioned. As a quick recap, Civ1/Civ2/Civ3 placed very few limits on expansion, making endless swarms of cities (the notorious "Infinite City Sprawl") the optimum way to play the game. The only limiting mechanic to stop expansion was corruption, which would cause cities further away from the player's capital to lose production and commerce, until eventually they became hopelessly corrupt and essentially useless. Corruption was a failed mechanic in multiple ways, not only doing little to stop expansion (since corrupt cities were still always a net positive and better in the player's hands than in someone else's) but also serving to be profoundly frustrating and anti-fun. Civ4 removed corruption completely in favor of city maintenance: in a reversal of the earlier games, buildings were now free to construct while cities themselves cost gold each turn to maintain. Cities therefore usually started out as net negative money losers, and eventually over time would become profitable as they grew in size and added markets and libraries and such. This was an excellent system overall if somewhat limiting in the early to middle portions of the game. Civ5 did away with city maintenance and replaced it with global happiness, a mechanic where all cities shared the same happiness pool. Each additional city also increased the cost to research social policies and (in the Brave New World expansion) technologies by 5 percent each. This had all sorts of terribly negative effects on gameplay, which I've chronicled in great detail elsewhere on the website. In a nutshell, this system essentially made expansion beyond a small handful of cities counterproductive and destroyed the entire rationale behind the game. Civ5 was an empire-building game where building an empire became pointless.
Civ6 was a gigantic step forward simply by scrapping the per-city science and culture penalties from Civ5. I wrote about this change in my pre-release article, and that remains very much true in the finished version of the game. Global happiness is gone, and Civ6 is back to being a game of expansion and competition for land and resources again, just as it should be. If the player fails to expand in Civ6, they will eventually fall behind those who do and struggle to win the game. Based on reading some of the online forums, this came as a rude surprise to many players who were introduced to the series by Civ5. No, sitting on four cities and otherwise never expanding is not good strategy in the rest of the series.
Let's examine expansion in a little more depth. Civ6 introduces a new mechanic by making settlers (and builders) scale up in cost with each one produced. The first settler costs 80 production and each additional one costs another 30 production (initially 20 production before the Summer patch), so 80 then 110 then 140 and so on. This means that the marginal cost of each settler keeps going up, forcing the player to weigh the usefulness of another city against building up the existing cities further with districts or other infrastructure. At some point, it becomes questionable if cramming another city into the desert or tundra is really worth the escalating cost. That said, because the cost of the next settler remains static until it's actually completed, as cities scale up in size and production over time, eventually it starts to become worthwhile to send out more settlers into unclaimed portions of the map. This creates a staggered pace of expansion that many other past Civilization games have tried and failed to implement. Aside from the scaling cost of settlers, there are no limitations on expansion, no corruption or maintenance costs or escalating policy costs. Bigger is always better in Civ6, and that's a huge plus as far as I've concerned. There's no need to have those silly "Wide vs Tall" arguments from Civ5; the correct answer is both Wide AND Tall. More population is always better in Civ6, however the player can manage to get it.
In practice, the scaling settler costs works pretty well at stopping the gameplay from being an endless sea of settler builds as in, say, Civ2. The other mechanic that indirectly works to limit expansion is the housing issue. While the player can put cities down anywhere in Civ6 (so long as they are at least 4 tiles from an existing city), effective city locations tend to be limited by the availability of fresh water for housing. When cities hit the housing cap, growth slows by first -50% and then -75%. The best tile yields in the world will stop a city from seeing much growth once those penalties kick in, and completely dry terrain causes the housing penalty to kick in at size 1! As a result, there's fierce competition in the early game for choice city locations located on fresh water rivers and lakes. Generally speaking, dry locations tend to require more technology to become useful, which is an interesting reflection of real world concerns. In actual Civ6 gameplay, expansion organically tends to happen in a staggered series of phases, claiming first the river locations and then branching out later to more marginal spots as technology and production improves. This is a pretty elegant system overall and it mostly works well in practice.
The one place where expansion falters is in relation to the AI civs. It's still a little too easy to shortcut this whole system by attacking the AI and stealing their cities away from them, which does not scale up the cost of settlers further. (The counter only increases when you build a settler; it's not based on the number of cities in the player's possession.) Making this matter worse is the fact that settlers are capturable in Civ6, a bad design decision that also serves to undercut the scaling cost mechanic. The AI does a poor job of protecting its settlers, and they can often be seen wandering around unguarded, easy pickings for the player. All of the fast finishes in the Game of the Month competition tend to revolve around sucker-punching the AI as hard and as fast as possible to get a leg up on the expansion curve. This is a mantra that I'll repeat a lot: the mechanics here are solid, however the AI's inability to understand them hurts the overall system. One thing I do have to point out is that the AI has been getting better at expansion with every patch that comes out. The release version was just sad how they would sit there and do nothing to expand the whole game. I'd say that the AI has now reached the "passable" level, but more work is still needed here. Especially on the top difficulties, the AI should be rapidly expanding like they do in Civ3 and Civ4. There's no reason for them not to be even more aggressive at pushing out settlers.
The biggest change in Civ6 was the decision to split out city infrastructure into districts that are placed on the map separate from the city center tile. Districts each have their own type, such as the Campus for science and the Encampment for military, and the relevant buildings associated with each type are constructed out on the map in their associated district. The placement of the district also grants benefits depending on the location of the surrounding terrain, such as mountains granting an adjacency benefit to Holy Site districts and next door mines helping to boost the production of Industrial districts. This was an interesting idea in theory, and it works quite well in practice too. I love the fact that the terrain matters so much in terms of locating the districts, and deciding which ones to emphasize has a noticeable influence on a city's development. If you construct an early Theatre district at your capital in a sweet spot next to multiple world wonders, it makes a big difference to your culture output. It's also neat how having the districts out on the map makes them inherently more vulnerable to attack. In one of our Multiplayer games, we've already seen a player put a Campus in a good location for adjacency bonuses, but also in an exposed position relative to other players. That Campus district was the target of an eventual attack, and when it was pillaged, it was a significant setback. Of course the AI doesn't do a good job of understanding how to take advantage of this, but the mechanic itself works reasonably well.
As for the problems with the district system, one issue that's come up repeatedly is their scaling costs. Districts become more expensive as the player researches more technologies and civics. This is never explained anywhere in-game and has frustrated many players. I actually think the scaling costs are pretty reasonable; districts tend to remain an expensive build (usually about 10-15 turns) throughout the game for an average city. There's been a lot of grumbling about putting down a new city and seeing a district build time of 100 turns, but that's a bit deceptive as new cities in the late game can benefit from a ton of methods to get up to speed quickly: cash-rushing infrastructure, trade routes, builder actions, superior tile yields via technology, etc. What the critics of this system are ignoring is that everything else in the game scales up to match the increased cost of districts. A forest chop and a trade route will knock out even an expensive lategame district in a dozen turns in a new city; I haven't found it difficult at all to get new cities up to speed as the game progresses. The biggest issue here is poor information about what's going on from Civ6's abysmal interface, not so much the mechanic of the scaling district costs themselves.
Under the same system used in Civ5, science and culture are divorced from gold in this game, and there is no spending slider to adjust as in the earlier Civilization games. Gold has the same uses in this game as it did in Civ5: purchasing tiles outside of a city's cultural borders, rush-buying things to completion, upgrading units, bribing the AI civs in diplomacy, etc. Districts and some of the key city infrastructure cannot be rushed to completion with money, however, and that's a good restriction to reign in the power of magic purchasing sprees. Gold generation tends to be unbalanced in Single Player, where it's easy for the player to amass large fortunes by selling off unwanted resources for gold/turn and there's little danger of military attack. This is once again mostly a problem with the AI and not the mechanics themselves; in Multiplayer, against other dangerous human opponents, gold generation is nicely balanced. It's critical to amass a lot of money to support the large militaries needed to be safe, and woe be it to the player who can't afford to upgrade units to the next generation of technology. Unfortunately, without the need to maintain lots of expensive units in Single Player, this system falls apart and the player typically winds up with more gold than can be spent. It's useful for lategame Patronage of Great People if nothing else though.
Trade routes are a related feature with a lot of interesting strategic implications. Some readers may recall that I wrote how the trade routes added in the Brave New World expansion for Civ5 were the best addition to that game, even if the implementation was a little bit off. The trade routes are back for Civ6, and this time they are the only practical way to construct roads between cities (the sad military engineer unit doesn't count). The trade routes were a little too good in the release version of Civ6, where each Commercial district and Harbor added a trade route, and they have wisely been scaled back so that each city can only add one trade route under normal circumstances, either from a Commercial district or a Harbor district but not both. Trade routes are still great though, adding food and production to their targeted city as well as gold to the wider empire. One of the ingenious ideas that the designers had was to tie the trade routes into the district system, such that a destination city with Encampments and Industrial districts adds a production bonus. It can be a good idea to have one city stack up the Encampment/Commercial/Industrial/Harbor districts to unlock 1 food / 5 production trade routes and jumpstart the production of new settlements. Other barren parts of the map might need to import food to grow in a reasonable time frame, and that's possible as well. The trade routes are affected by about a dozen policies and they can be heavily customized depending on the player's needs.
The trade routes tend to work even better in a Multiplayer context. Because the little caravans exist out on the map itself, they can be pillaged by enemy units for a substantial boost of money. Each trader unit also has to be built individually, and it can be a tough decision as to when to fill out those trade routes. In the middle of a hot war there's often no production available and losing several trade routes at once can be heavily damaging. We've also seen some creative uses of the trade routes when it comes to city states. They don't produce much in the way of food or production, but trade routes to city states tend to generate lots of gold. In the ending stages of one of our Multiplayer games, oledavy was running his entire economy off of a dozen trade routes going to one of the last remaining city states on the map. When it was captured by a rival player, his income plummeted by abouy 75% and he had to reconsider his plans to upgrade a bunch of military units to the next generation of technology. This was fascinating stuff and it helped to turn the city states into real bones of contention between players. On the other hand, the biggest drawback to the trade routes is the fact that they require a lot of micromangement from the player. This is one reason why I've limited myself to Small and Standard maps in Civ6. Managing five trade routes is fun and interesting whereas managing 35 trade routes is tedious. Stay away from the larger maps and you'll likely have more fun with this mechanic.
Civ6 also replaces workers with builders, units that only have a limited number of charges before vanishing. As I wrote in my early preview of Civ6, if you're going to use the One Unit Per Tile system, this is a much better way to implement tile improvements. I hated the extremely slow rate at which Civ5's workers would improve the landscape, and the One Unit Per Tile rules prevented players from pairing them up to knock out improvements at a faster rate. None of this is an issue with Civ6's builders, as they instantly complete whatever tiles they intend to improve. Deciding when to train builders is a fascinating strategic choice, as there's a legitimate decision between opening with a builder before settler or a settler before builder depending on the local terrain. It can also be a very difficult decision on which tiles to improve with a builder's limited charges. Builders scale up in cost with each one completed in the same fashion as settlers (50 production for the first one and then 4 additional scaling production for each subsequent builder), and I find myself training them throughout the game. At a more advanced level, builders can also chop forests/jungles and harvest all of the game's bonus resources for a one-time infusion of food, production, or gold. This is a powerful tactic that must be understood to compete at the top levels of play in Civ6, akin to the Slavery mechnic in Civ4. By the later stages of the tech and civics tree, builders can turn almost any part of the map into fertile, productive terrain that's worth settling. It always feels like there's more work for the builders to do, and that's a good feeling to have.
On a related note, it's such a feeling of pleasure to have the terrain back to being so important again! One of my biggest criticisms of the gameplay in Civ5 was that all of the land tended to feel the same. Where my cities were planted didn't seem to matter that much because all of the tile yields had been heavily nerfed and cities gained so much of their food/production from factors outside of the map. Civ6 is back to being a game where terrain matters enormously: desert and tundra patches are much, much weaker than grassland once again. Hills and forests are both excellent to have, and which resources are located at the start matter a great deal. Tile improvements are also back to having real yields, with mines adding +1/+2/+3 production as the player advances through the tech tree, and farms picking up more and more food depending on their adjacency bonuses. The ability to plant forests and lumbermill them for bonus production revolutionizes the whole gameplay when it becomes available, and I wish it didn't come 2/3rds of the way through the civics tree. With the district placement bonuses added into the mix, the map is back to being a living entity once more.
Let me close this section with some criticisms of Civ6's economics. The biggest, most glaring issue was detectable within my first five minutes playing the game: the user interface is horrible! I have no idea what the designers were thinking when they put it together. Accessing the city screen requires drilling down into a half dozen menus and submenus. Nothing is located in the same place, and the interface refuses to update numbers as the player looks at different options in the production queue or moves around citizens to different tiles. Mouseover tooltips are either confusing or nonexistent. The diplomacy screen with AI leaders lacks a basic matrix to display all of their relationships at once, and the victory conditions are so poorly explained that there's no clarity about what's going on or why. The release version of the game didn't even tell the player which tile was going to be grabbed on cultural expansion and wouldn't let players rename cities! Was it THAT hard to do a simple text replacement?! It's still shocking to me that an expensive game like this could have such a shoddy and poorly designed interface given the attention that clearly went into polishing the mechanics. I still have no idea what they were thinking here. The patches have made some modest improvements, adding badly needed icons in a number of places for clarity, but much more work needs to be done here. Hopefully the expansions will just steal some of the community-designed interface mods and incorporate them into the main game.
My other biggest criticism of the city-building and economy mechanics: the AI doesn't understand how to use them. This is the inevitable result of designing such complicated systems based around district placements and One Unit Per Tile rules. In past Civilization games, it was enough to give the AI a bunch of production bonuses and let them build everything in their cities. Now they can't do that because most of the game's buildings require districts as preconditions, and the AI struggles to put up much of a challenge. While they have gotten a lot better since the pitiful AI of the release version, the AI can still be out-researched on Deity difficulty without much trouble. The AI can only win the game via early military rushes - it cannot play the longterm economic game at all. What I can't understand is why the designers haven't responded to this by giving the AI more cheats to improve its performance. Doing something as simple as giving the AI bonus free housing on the higher difficulties would help enormously. This would let them grow their cities much larger (therefore picking up more science/culture from population) and also unlock more districts via the population requirements. Yet the AI doesn't get any housing cheats at all right now from difficulty as far as I can tell. Why not? They clearly need more help than they're getting at the moment. This is the single most disappointing aspect of the whole economic side of the game: the AI's inability to understand the system undercuts nearly all of the difficulty. Single Player games have gotten boring largely because the AI falls enormously far behind in every game after the first 100 turns. If the designers can't fix the AI, they really should dial up the cheats so that we can have a competitive game on Immortal and Deity.
Research is an outgrowth of the economic system, but Civ6 made enough changes to the standard Civilization formula that it warranted its own section. Scientific advancement has been around since the first game in the series, with research used to unlock more advanced technologies that offer a wide-ranging series of benefits. After culture was introduced in Civ3 and tied to expanding city borders, one of the recurring questions has been how to make culture more useful as a game mechanic. Civ4 tied cultural output to city defenses, which was somewhat useful while still falling short of true importance. Civ5 shook things up by having borders expand one tile at a time (instead of popping in large rings) and tying culture to the social policy system. For all of my many criticisms of Civ5, these were good ideas to make culture more useful as a concept and deserved to be expanded upon and implemented in a more balanced fashion.
Civ6 follows this exact logic. The social policies in Civ5 were effectively a second tech tree using culture as the currency instead of beakers. While the game did its best to disguise this fact by gating off many of the social policies by era advancement requirements and created a series of brancing paths, it didn't change the fact that social policies were a cumbersome second path of advancement using a different counting tool, culture instead of beakers. Civ6 embraces this concept and simply creates a civics tree tied to culture that advances alongside the tech tree. This is a brilliant small tweak that makes the whole mechanic more intuitive to understand and presents the information about the civics in a fashion that's easier to read. It also creates situations where a civ may be strong in science but weak in culture, or vice versa - I've played out both situations and they create some interesting dynamics. The more advanced civics are very powerful and players who fall behind on the civics tree can find themselves in serious danger. One of the biggest reasons why I was able to win the first Play By Email Multiplayer game at Realms Beyond was by picking Rome and riding their dominant early game culture into a position of supremacy.
In a related mechanic, the social policies of Civ5 have been replaced with policy cards in Civ6. Each of the game's governments has a set amount of available policy card slots in four different types (Military, Economic, Diplomatic, and Wildcard) and policy cards of each type can be dropped into each one. Policy cards are unlocked as the player advances through the civics tree and can be swapped for free every time a new civic is discovered, typically about every 5-10 turns. Skillful use of the policy cards will make or break each game; reaching Colonization's +50% bonus to settler production enormously speeds up expansion, Professional Army's half cost unit upgrades can swing the military outcome of whole games, Rationalism's doubled science output from Campus district buildings will leave players who lack it in the dust on research, and so on. Unlike the one-off bonuses in Civ5's social policies, things like "a free settler appears near your capital", everything in the policy cards only modifies an existing bonus, never creating something out of thin air. Critically, they are an active gameplay element, not a passive one. Instead of something being given to the player out of the ether, the player chooses to improve something that they were already doing. And the benefits from the policies can be shifted around over time as the situation demands. The policy cards are therefore both powerful and flexible, offering up a wealth of tools to intelligent players and make it clear that strong cultural output is vital. Falling an era behind in civics is almost as bad as falling an era behind in technology, maybe worse in some situations. The civics and policy cards are a gigantic winner from a design standpoint and I can't give them enough praise as a mechanic.
If there's a weakness to the dual tech/civic trees system, it's that the later stages of both of them wind up feeling bland. There isn't quite enough content to fill up all 80 techs and 55 civics at the moment, which is something that may be addressed in expansions. The Atomic and Information eras are dull affairs, mostly serving up an endless sea of new military units that aren't important because the player has long since supplanted the AI. Maybe if we ever get a Multiplayer game that reaches that deep into the trees it will be a different affair - maybe. This has always been a problem for the Civilization games, and it's particularly acute in Civ6.
Another new mechanic in Civ6 are the "eurekas" and "inspirations" for science and culture respectively, which everyone has taken to calling the "boosts". The idea is simple: each technology and civic has a miniature quest associated with it, and if fulfilled, it will grant half of the beakers or culture for that particular research goal. For example, quarry a stone resource and you will knock out half of the cost of the Masonry tech. The boosts start out quite easy and get progressively more difficult and esoteric as the game goes on, until some of the last ones are unboostable except with a Great Person. Most of the opinion on this mechanic from the Civ community was negative when the game released, but it's turned out to work relatively well in practice. Balancing out which boosts to pursue alongside other goals creates an interesting dynamic in the gameplay. Sometimes it's better to wait to finish researching something until one of the little quests can be fulfilled, and at other times it's better to ignore the "saved" beakers/culture and press on regardless. This creates real strategic tradeoffs. It also increases the replayability of the starting terrain in each game, as certain boosts may be trivial to get in one game and then be almost impossible to unlock in others. My opening research strategy therefore tends to be dictated not only by the land around me, but also by which boosts are easy to unlock, and which ones will require additional time and should be delayed until later. The key to the whole system is that the boosts never change from game to game, which allows players to strategize around them. Some of the early reviewers, most notably Tom Chick, wanted all of the boosts to be completely random and unknowable from game to game. That would have been a disaster, however - it's the predictability of the boosts that makes the whole system work. While the system isn't perfect (especially in terms of the AI which again has zero understanding of the mechanic), I believe it's been more of a net gain than a net loss. At the very least it hasn't been as harmful as many feared before release.
As a final issue for this section, let's discuss the Great Person mechanic briefly. Civ6 changes how Great People work by uses a mechanic similar to the Founding Fathers in Sid Meier's Colonization. The Great People all have unique functions and are recruited individually by "purchasing" them with their respective currencies. Great Merchant points fill up in a civilization-wide bar until someone has enough points to recruit the first Great Merchant, then a new one becomes available and the process starts over again. Great People can also be patronized using faith or gold, albeit at very high prices that will normally be out of reach. The Great People have a wide range of different functions, and depending on the path a player may be pursuing, some can end up being much more useful than others. Great Generals and Great Admirals are extraordinarily powerful for military games, while they will be almost useless for a pure builder campaign. Both the Cultural and Spaceship victory conditions are highly dependent upon recruiting key Great People, if not always the type one might expect (there are some Great Merchants that are arguably the best to have for the Cultural victory).
This system works decently well on the whole. However, it suffers from several flaws that could easily be fixed via patching or expansions. The biggest problem is that only one Great Person of each type is up for recruitment at a time, and there's no way to choose which one a player might prefer to have. They can "Pass" on a Great Person but that means waiting for someone else to recruit the one currently up for grabs, and against the AI that might mean waiting for a very long time indeed in some cases. It's annoying to be forced to take a Great Person that you don't want when the luck of the dice could have given you a much better one instead. The era of the Great People is also determined by a formula based on the average advancement of all civs in the game, which typically means that the Great People from the Classical and Medieval eras are rushed through in a flash with only one of each type appearing. This is again highly frustrating when desirable choices like Hypatia (+1 science to all libraries) never even appear. This mechanic has an incredibly easy fix: make all Great People always available once the era requirement is met and let them be recruited however players desire. If someone wants to save all their Great Engineer points the whole game and then dump them all into Modern era Engineers, that's their choice. This would be more fun and more strategically interesting, especially for Multiplayer. It would also improve game balance by making more of the early Great Generals/Admirals available in Multiplayer, and maybe it would actually make Great Artists useful by letting players pick the ones they wanted for the theming museum bonuses. It's such a simple fix that I really hope this happens at some point.
Diplomacy was an area that was poorly implemented in the release version of Civ6. The AIs would repeatedly sneak attack for no clear reason, often doing so through weird "joint wars" where two AI leaders attacked simultaneously. They didn't appear to put much consideration into the state of current relations before attacking, and with the enormous warmongering penalties in the release version of the game, all of the AI leaders would wind up detesting one another as they fought repeated meaningless wars against each other. It was an ugly situation and not a lot of fun to play. This was compounded by an awful series of bugs involving the trading screen, bugs where it was possible to get the AIs to hand over thousands of gold for free, or trade away all their cities for nothing in exchange. It was patently obvious that diplomacy was unfinished when Civ6 was sent out to the public and the mechanic rightfully came in for plenty of scorn.
A year later, the system is definitely working better although I would hesitate to shower it with too much in the way of praise. It's much more possible now to make friendships with individual AI leaders, and I've been able to sign plenty of alliances when my gameplan happened to line up with their agendas. There are also some basic things that improve relations across the board that are worth doing with everyone: signing Open Borders and sending a trade route will both provide a big fat bonus, and I've found them worthwhile to avoid unwanted conflicts. The random joint wars are almost entirely gone now, and the AI seems to be smarter about picking and choosing its conflicts. They are more likely now to attack civs with whom they have unfriendly relations, not seemingly random neighbors with no pretext. I like the fact that each AI leader has its own pair of agendas for each game, one that's always present and one that's picked randomly before each game begins. I don't like that these agendas are hidden until the player unlocks more levels of spying - neat idea in concept, but it's horribly unfun to have an AI leader that hates you and won't tell you why. In my experience, if you mind your own business and stick to a peaceful game, the AI leaders will mostly leave you alone in turn. However, if the player repeatedly launches unprovoked wars of conquest, expect to be hated by everyone else, and deservedly so. The warmonger penalties have been greatly tuned down since the game was released, when one war declaration at any time after the early stages of the game would make your civ an instant global pariah. War weariness has been tuned down as well. Even with those changes from patches, I still think the warmonger penalties are set too high. The diplomatic and unhappiness penalties in the later eras are too severe and stop conflicts from happening in a way that's not conducive to an interesting endgame.
If I had to summarize diplomacy in one word it would be "functional". It's no longer broken and it mostly works as expected. The diplomatic interface is still terrible though, and the whole system remains weaker than what we had in Civ4 a dozen years earlier. That's about the best praise that I can offer for this mediocre setup.
City states are a happier story. They were introduced in Civ5 as individual cities that would never expand and offered benefits in exhange for befriending them. However, it was clear that the designers of Civ5 didn't have any real idea of what to do with city states after introducing them. City states offered random quests that gave out huge rewards in exchange for trivial tasks like connecting a silver resource, and influence could be bought through a ridiculous and gamey "pay to befriend" gold spending mechanic. Civ6 ties the city states to the district system and implements a new envoy mechanic to simplify the process of how alliances form. The city states in Civ6 again have individual types (Scientific and Commercial and Culture and so on), and gaining additional envoys provides extra benefits associated with the district of that type. The first envoy with an Industrial city state grants +2 production in the capital; the third envoy grants +2 production to every Industrial district, and then the sixth envoy grants another +2 production to every Industrial district. Each city also has a unique benefit that it grants to its "suzerain", the civ that has the most envoys invested in it. Many of the suzerain benefits are quite useful to have, like Stockholm's bonus to Great Person generation or Nan Madol's ability for cities to ignore fresh water requirements, and competition for the suzerain benefits can be fierce.
The city states do still offer quests, but in a much more intelligent manner. The first to meet a city state gets a free envoy, and achieving a quest grants another free envoy. If the player wants more envoys, they must be unlocked through the civics tree or by running policy cards that generate more envoys. This still creates a space for city state quests to play an important role without making them the all-important factor that they were in Civ5. You'll still get an envoy sometimes due to dumb luck, but it will only be one envoy at a time, and other civs who put more emphasis on the envoy system will easily be able to outpace a single random envoy. We've found from our Multiplayer games that competition for city states can be fierce and players will work hard to fulfill quests to get more envoys. The AI is certainly obsessive about competing for suzerain bonuses, although this is one area where they do cheat on higher difficulties and get envoys handed to them for free. The city states also present an interesting strategic choice when it comes to keeping them around for their district/suzerain bonuses versus attacking and annexing them. Generally speaking, Commercial and Scientific ones are some of the best to keep around, while Militaristic city states are likely to get the axe. To summarize this section, the city states in Civ6 are basically what the city states in Civ5 should have been. The mechanics surrounding them are well designed and have proven to work effectively in both Single Player and Multiplayer. I wish that the diplomacy mechanics with the AI civs functioned as smoothly as the ones involving the city states.
Civ6 continues to use the One Unit Per Tile gameplay mechanic from Civ5. This is the rule that only allows a single unit to appear on each tile, the single most radical gameplay departure since the inception of the series. I was a vocal critic of the decision to introduce One Unit Per Tile in Civ5, and I continue to believe that it's a mistake for the Civilization series to use it. The Civ games are designed to portray empires on a grand scale, and it's a loss from both a theming perspective and gameplay mechanics when battles are limited to a handful of units on each side. The problems that existed in Civ5 are still very much present in Civ6; for example, unlike dedicated tactical wargames the map isn't large enough in Civ6 to accomodate One Unit Per Tile maneuvering. This means that units must get very expensive very quickly as technology advances to keep the map from overflowing with units and getting stuck in the dreaded "Carpet of Doom" phenomenon. It's way too common for units to get choked up on one another and be unable to move, leaving the player with a frustrating experience. Civilization should be a game of grand strategy where the focus is on military production and logistics, not a game of tactics where the player can win through battlefield manuevering alone. One Unit Per Tile was and continues to be a bad fit for this particular series of games.
With all that said... if Civ6 was going to use One Unit Per Tile as a gameplay mechanic, it's implemented about as well as it could be. The first and most immediate change comes in the form of the worker units, which have been replaced by builders with their limited number of charges. The workers in Civ5 were a painful relic of earlier Civ design, taking forever to add tile improvements and gumming up the roads for movement purposes while they were mining or farming or whatever. The system of builder charges is a much better fit for the One Unit Per Tile mechanics. Similarly, the designers have split off the other non-combat units into their own strategic "layers" where they don't block or interact with military units. Religious units now have their own movement layer, finally, as of the last non-expansion patch. They were a huge pain in the earlier versions of Civ6 where they moved on the same strategic layer as combat units. Traders also get their own layer, which means that theoretically you could have a combat unit, a builder, a trader, and a missionary all standing on the same tile at the same time. These little tweaks do a lot to help relieve some of the congestion on the map.
As a second mechanic related to improving One Unit Per Tile, Civ6 also introduces the idea of corps and armies. This is a mechanic that allows military units to be combined together and pool their strength into a single unit. Combining two units together creates a corps and adds +10 to the unit's strength, while pooling three units together for an army adds +17 strength. (Naval units have the same ability and are known as fleets and armadas.) Sometimes it's better to keep units separated to get more total attacks against a target, and sometimes it's better to use corps or armies to get the job done. If there's a weakness here, it's that the corps and armies don't arrive until the halfway point on the tech tree. The gameplay would be improved by enabling this ability from the very start of the game; if someone wants to join their first two warriors together for a stronger unit, and lose the opportunity cost of having two units out on the map, I see no reason why they shouldn't be able to do so. This is a smart mechanic that helps to reduce the number of units out on the map in the lategame.
The corps and armies wouldn't work without a well-designed combat system, however, and fortunately Civ6 does indeed have excellent battlefield mechanics. Everything in the game keys off of a single strength number, and if two units with the same strength value attack one another, each will take a random number between 24-36 damage, with 30 being the average value. From that starting point, the bigger the disparity in strength between the two units, the more damage dealt and the less damage taken. If one unit has an advantage of 10 points in strength, they deal 50% more damage and take 50% less damage, for a more than 2:1 disparity overall. (They would deal about 45 damage on average and take about 20 damage.) With an advantage of 17 points in strength, a unit deals 100% more damage and takes 100% less damage (dealing 60 damage and taking 15 damage) for a 4:1 ratio. These numbers are no coincidence, and this is why the corps and armies provide +10 strength and +17 strength respectively. It sounds silly to combine a pair of units together until one realizes that the new corps is roughly twice as effective in combat as either of the units individually. If a unit can reach an advantage of 30 points in strength over another unit, it reaches the realm of one-shot kills while taking almost no damage in return. This is difficult to pull off in practice but advacements in technology and unit promotions can make it happen.
From this basic math of how the strength values interact, the rest of the combat system flows naturally. Even small disparities in combat strength start to become hugely valuable. Rough terrain can add as much as +6 strength to a defending unit, which is worth about 25% more damage dealt and 25% less damage taken. Fortifying on a tile adds +2 strength on the first turn, then +4 strength the second turn, and +6 strength the third turn, capping at that +6 value. A fully fortified unit on rough terrain is worth more than a corps and can be extremely difficult to dislodge. There are flanking and support modifiers as well, with defending units gaining +2 strength from each adjacent friendly unit and losing -2 strength for each enemy unit standing next to the attacker. Unit promotions can further increase strength values, and damaged units lose combat strength as they become injured, losing -1 strength for each 10% of their missing health. The whole system ends up working very well in practice, with the tactics of individual unit placement hugely important to the outcome of each battle. This is a combat mechanic that has to be played against humans to get the full benefit, and we've already had some epic confrontations in the limited number of Multiplayer games conducted thus far. We had a naval showdown in the PBEM4 game where three major fleets clashed with each other at the same time, and it might as well have been Trafalgar out there with frigates and ironclads literally crossing the T against each other to maximize their damage output. When two large forces are clashing and guided by human minds, it's a thing of beauty to watch.
I'm mentioning human minds for a reason here, because the same crippling problem regarding One Unit Per Tile remains from Civ5: the AI cannot use this system at all. Not even a little bit. This is the single biggest flaw undercutting the entire game and it's the main reason why Civ6 is a better Multiplayer game than a Single Player game. While the AI was capable of building large stacks and throwing them against the player's cities in Civ3 and Civ4, the AI cannot understand how to maneuver units under the One Unit Per Tile rules. As a result the AI cannot pose a military threat to the player in Civ6, not even on Deity difficulty, and much of the gameplay falls apart from there. The whole military side of the game collapses without a credible opponent; many of the Military social policies never need to be used, city walls can be ignored, and the AIs are there to serve as punching bags for capturing cities and settlers, not actual opponents who need to be respected and feared. When we started the first Civ6 MP games, we were surprised to find that the Military social policies were actually really good! We had just never needed most of them because the AI opponents were so laughably inept. The need for a large (and expensive) military is also how the economic side of the game is supposed to be held in check, with advanced units getting more and more expensive over time, therefore keeping civs from piling up gigantic gold/turn incomes. This works perfectly in our Multiplayer games where everyone is always frantically struggling to get more money to maintain and upgrade their forces. In contrast, every Single Player game always seems to accumulate thousands and thousands of unspent gold since a token military can keep the AIs at bay. It's the AI at fault here, not the gameplay mechanics themselves. The AI's failure at One Unit Per Tile combat is the biggest reason why I continue to hope that this system will be scrapped in future Civilization games.
I'll limit myself to two more points on the subject of combat. The first relates to this game's movement system: Civ6 institutes a change whereby a unit must have enough movement points to enter a tile, rather than always being able to enter a tile so long as there are movement points remaining. In practical terms, this means that a warrior with two movement points can't move onto a flatground tile and then move onto a hill. It needs both of its movement points to enter that hill tile. This new mechanic makes movement horribly slow for most of the game, and drags out exploration of the map for a long time. It also prioritizes ranged units heavily, as they can often shoot at melee units without being hit in return. Offensive warfare almost requires a Great General or Great Admiral (which provide +1 movement and +5 strength to units in their respective eras) for melee units to be able to get on top of the ranged units effectively. Roads are similarly almost useless in the early stages of the game, barely providing any additional movment points until they are upgraded to their Industrial and Modern era forms. Everything involving movement in Civ6 is needlessly tedious and painful to use. I suppose the counter argument is that it makes unit placement in tactical combat extremely important because of the difficulty of maneuvering around the battlefield, but I still find this to be a bad system. The word is that lead designer Ed Beach insisted on using this movement system over the strong objections of everyone else on the development team, and while I think Ed Beach did good work overall in Civ6, this is one place where the rest of the team was correct. There's no reason for movement to be this frustrating and counterintuitive.
Finally, I can't conclude discussing combat without mentioning the barbarians. Their presence was dialed up significantly in Civ6 and the early game typically has barbarians crawling around everywhere. Generally speaking I approve of having barbarians as a more prominent element to cut down on the farmer's gambit openings that often prevailed in earlier Civilization games, but the problem with the barbs in Civ6 is that they aren't implemented evenly or fairly. Barbarian camps will send out scouts to search for cities, and if found, the scout will return to the camp followed by a series of barbarian units spawning inside to attack. While this works well enough much of the time, it's very possible for the barb scouts to be completely uncatchable by the player's slower warriors and slingers. Even worse, if there happens to be a horse resource within 5 tiles of the barbarian camp, the camp will spawn faster, stronger horse units instead of the normal warriors and archers. Nor are there any restrictions about where barbarians can spawn; in our PBEM2 game, Singaboy had a barbarian camp appear right outside his capital and start spamming units in his face on Turn 5. There was nothing he could do about this and it effectively knocked him out the running for victory before he even had his first unit built. I would love to see some coding put into the barbarian mechanics so that barb camps can only spawn a set distance away from existing cities and ditch the ridiculously swingy horseman spawns. It's patently unfair that sometimes the barbarians will have little to no effect while at other times they will be deluging the player with units in the first dozen turns of the game, and it's mostly down to pure luck as to what happens. If nothing else, create more options other than Barbarians "On/Off"; why not have the old "Restless", "Roaming", and "Raging" barbarians? More work can be done here to make this a less random experience.
The natural finishing place for this review of Civ6 is with a discussion of the victory conditions. Unfortunately the four victory conditions are poorly designed for the most part, unintuitive and undocumented for the player, combined together with an AI that has no understanding of them. The Domination victory condition is the most straightforward, retaining Civ5's victory condition that involves capturing the capital city of every other empire. I've never liked this idea since it feels like the definition of a "gamey" mechanic. Why do the capital cities have this mystical power that causes them to be the arbiters of victory or defeat? It's frustrating that they cannot be razed for this reason, although Civ6 fortunately does allow the player to raze city states. It's actually possible to win the Domination victory without doing anything at all, if the other civs happened to all lose their own capital cities to someone else. That would never happen in Civ6 because the AI is terrible at combat, but it is possible. I would vastly prefer the old Domination mechanics based on population and land area, or even population alone since the map generally doesn't get filled up with borders in Civ5/Civ6. Why not have a rule where controlling 65% of the world's total population makes you the winner? If anyone has that much population then the game is over, period, without any need for capital city shenanigans. The current Domination victory lends itself too much to gamey exploits like sucker-punching the AI to take its capitals without having the ability to hold them afterwards. I'll take a pass on this one.
The Spaceship victory condition is probably the best designed of the group, based on reaching the end of the tech tree and building the spaceship parts. For whatever reason the spaceship parts are way too expensive, however, costing 900 production, 1500 production, and then 1800 production for the last three parts. The Spaceport district alone costs 2000 production to build! And this is after dialing down the costs from the 3000 production they cost in the release version. There is a way to build them quickly, of course, however it requires recruiting the correct Great People in the later stages of the game - and that's hoping that the correct Great People are there at all since it's a dice roll on which ones appear. For the poor novice player who doesn't know any of this stuff, they're going to finish the tech tree and then spend the next 50 turns sitting around slowly waiting for the parts to get built. Again, there's no reason for any of this tedium. The Spaceship victory is the economic victory condition in Civilization, and if someone reaches the end of the tech tree first, they've already demonstrated mastery of the game's economics. Making them jump through hoops to build the spaceship itself doesn't add anything to this process. I can live with this setup but it's far from my favorite design choice.
The Cultural victory condition is all kinds of wrong from a design perspective. For starters, the interface never explains in the slightest what's going on or what's required to win via culture. In my very first game of Civ6, I won by culture and was left clueless as to what was happening. Obviously I needed to accumulate tourism to win this way, but where were all of the numbers coming from? What did those tourist fractions on the Cultural Victory screen next to the other civilizations mean, and why were they going up and down? It was all a mess and nothing has been changed or improved in the patches since then. Once I did learn what was going on from a forum thread at CivFanatics, the system did start to make sense. Domestic tourists are generated by your culture and defend against someone else winning a Cultural victory. Foreign tourists are generated by your tourism and are how you go on the offensive to win a Cultural victory. As it turns out, accumulating culture itself doesn't do anything to win a Cultural victory (heh). Only tourism does that. A Cultural victory takes place when you have more foreign tourists than anyone else has domestic tourists, a silly way to measure a victory condition that makes the whole thing highly contingent on what the AI civs are doing. It's ridiculous to me that the best way to win a "Cultural" victory is to go kill everyone else with high domestic culture because eliminating them from the game will lower the tourism requirement. This is stupid and flows directly from a poorly designed victory condition.
Furthermore, it seems clear that the designers themselves didn't understand the system that they created. Tourism is supposed to be mostly generated through building world wonders and recruiting Great People, specifically Great Writers, Great Artists, and Great Musicians. These Great People create various Great Works, which are then stored in "slots" created by Theatre district buildings and certain wonders. However, the designers miscalibrated the numbers and this is not at all how the system works in practice. Great Writers are pretty good but Great Artists are terrible, with a small tourism output unless the player can recruit a ton of them for the very difficult "theming bonuses". Given that not all of the Great Artists will even be available in each era, this makes it nearly impossible to theme art museums with their works. Instead, most of the tourism comes from archaeologists with their artifacts and then seaside resorts that can be added along the coasts, with their own complicated and largely undocumented rules for placement. The Archaeological museums are much easier to get the theming bonus, and the designers didn't seem to realize that seaside resorts can be built without limit, with their effectiveness magnified by the Eiffel Tower and Cristo Redentor wonders. I find it absurd that success in the Cultural victory condition relies mostly on moving archaeologist units around the map to dig up ruins and build resorts on a whole bunch of coastal tiles. Do I even need to mention that the AI has zero understanding of this system and can never pursue it effectively? This victory condition needs a complete redesign; something along the lines of Civ4's three Legendary cities (50k culture) would be a lot more elegant.
Then there's the Religious victory condition. The goal under this mechanic is to convert every other civ to your state religion, and the player must have founded their own religion to be eligible for this condition at all. The Religious victory is difficult to pull off and works reasonably well; my biggest issue is that it's tedious to move so many missionary units around the map. This mechanic was a lot less polished in the release version of the game since missionaries shared the same strategic layer with combat units and the two types kept getting in each other's way constantly. The final non-expansion patch established a separate strategic layer for religious units and vastly improved this victory condition, to the point where it's actually kind of fun to play on occasion. The patch also finally shed some light on the religious spread mechanics, making religious pressure from one city to another visible and listing the number of turns for a city to convert. This information was badly missing from earlier patches and kept most of what was going on in the dark. With the addition of guru units to allow religious units to heal away from Holy Sites, the religious side of the gameplay is now working pretty well overall. It's worth trying out in a game or two just for a change of pace.
One positive thing about these four victory conditions is that they do play out quite differently. A Religious game is very different from a Cultural game, which is different in turn from a game with Domination as the focus. This helps to add a lot of replayability to Civ6 as the player tests out different civs that may be better suited to different victory goals. On the flip side, the lack of some kind of diplomatic victory condition is a significant drag on the gameplay. Without some kind of United Nations mechanic in the background, it's far too easy for the player to be incredibly evil towards the AI civs with no repercussions. This is something that will likely be added in expansions, hopefully in a better format than the one used in Civ5. (Is it too much to ask that the series return to voting by population like in Civ4?) Finally, I have to mention that although Civ6 does include a short victory cinematic for all four victory conditions, there's still no map replay or Hall of Fame to look back at after winning. The "Dan Quayle" leader ranking also remains hopelessly broken, with the score for each game equal to the in-game score on the last turn with no other modifications. There's no bonus for winning on higher difficulty and no bonus for faster finishes, which means that this game has a weaker scoring system than even the original Civilization back in 1991. There's also no formal Map Editor or Pitboss for Multiplayer, and the modding tools remain scant despite many promises to make them available. It's pretty clear that Civ6 was rushed out the door at the end of 2016 before a lot of this stuff was finished and none of the patches have addressed these incomplete features. The patches have done excellent work thus far fixing other, more serious, gameplay issues but I continue to hope that these things will get some developer attention down the road.
Civ6 is a better game than I expected it to be. I was cautiously optimistic at the time of the game's release and thus far it's more than lived up to the tepid expectations that I had. The game gets a lot of things right: expansion is back again, meaningful tile yields have returned, the district system is a big winner, the double research/culture trees are fun, the policy cards add a ton of strategic decisions, city states are infinitely better handled than their implementation in Civ5, and the combat system is surprisingly rich from a tactical perspective. There are two crippling flaws in Civ6 that still need to be addressed for this to go from a good game to an excellent one. The first is the user interface, which has been improved slightly but still remains a baffling mess that limits and hides information from the player. The whole city screen interface should be tossed into a radioactive fire and rebuilt from the ground up. If nothing else, the designers should add more mouseover information and stop trying to hide all of the actual numbers. No, I don't want to see only "5 turns" on a build queue or turns until city growth - give me the actual production and food numbers! Stop hiding information for no reason!
If nothing else, the interface problems can be handled through community mods. They aren't the biggest problem. Far more serious is the Civ6 AI, which doesn't understand how to use most of the mechanics that make up the gameplay. Yes, the AI has gotten better since the horrendous release version AI that sat in place and never expanded. The AI does an OK job of settling the map now and they're generally pretty good about founding religions and chasing after Great People. However, the AI can't play the military side of the game at all - not even a little bit - and the AI fails rather spectacularly at the economic side of the game as well. The only threat from the AI always occurs in the first 100 turns, where they can be dangerous through the sole function of early rushes using their free starting units. They also get enough cheats that they will have a science and culture lead in the early stages of the game. And yet if the player can survive the first 100 or so turns, the AI always completely collapses after that. Always. They can't attack cities effectively, they can't research their way through the tech tree, and they can't pursue any of the victory conditions in a coordinated fashion. The player is guaranteed to win if they have any clue what they're doing once they reach the midgame. It's profoundly depressing and has effectively killed our community's interest in Civ6 Single Player. Dumb as the AI may have been in past Civilization games, they could at least get to the end of the tech tree with their huge bonuses and compete in the space race.
I'll continue to lobby for the designers to vastly dial up the cheats that the AI is getting for the top difficulty levels. If they can't compete fairly, then jack up those bonuses until they can compete. Right now the highest difficulty levels are a joke; in my Immortal difficulty game with Poland, the top AIs were struggling to break 150 beakers/turn on Turn 200, at a time when I was over 500 beakers/turn. That would never, ever happen in Civ4. The biggest thing that Civ6 needs is an AI opponent capable of playing the game, not the various doodads and whistles that the designers are planning to put into the upcoming expansion. I understand that that's not how capitalism works, and in order to sell more copies of the game they need to advertise new features. However, if we're being serious about what would make Civ6 better, it's an AI that can move units around effectively and build districts in a logical fashion, not Golden Ages and Loyalty and Governors or whatever.
If you can ignore the AI's inability to play the game, then Civ6 is a pretty good game. The typical player likely won't even notice or care about these things. Don't get me wrong, I certainly don't think that Civ6 is casualized or "dumbed down" or whatever else derogatory buzzword is in vogue these days. Quite the opposite: the game's biggest problem is probably that it's too complicated, since the AI certainly can't understand most of the features and I suspect a lot of players have also been left confused by the district mechanics. The key difference between Civ6 and Civ5 is that this game has mechanics that actually *WORK*. The AI doesn't understand how to play either game, but the vastly superior mechanics in Civ6 make it an excellent Multiplayer environment if nothing else. I suspect that this game's future in our community will mostly be centered around Multiplayer, where we can get around the issues with the AI. For myself, I might have a few variant ideas that I still want to experiment with in a Single Player environment, and we'll have to see what the expansion(s) will bring.
Overall, Civ6 might be a bit of a mixed bag, but it does do enough things right for me to characterize this one year review under the heading of "exceeds expectations." Thanks for reading.